100
Advanced Discretization Techniques
Immersed Boundary Methods have garnered significantly increased attention over the past ten to fifteen years. Their central principle involves extending the domain of computation to a larger one, typically with a simple shape that is easy to mesh. On this extended domain, a finite element-type computation is performed, distinguishing between regions interior and exterior to the original domain. Known under terms like fictitious domain or embedded domain methods, this central principle has been in use since the 1960s. Recent renewed interest is driven by innovative, efficient algorithmic developments and mathematical analyses showing optimal convergence despite the presence of cut elements. Additionally, the compatibility of these methods with various geometric models and their application to many new engineering challenges have contributed to their popularity. Numerous variational versions of Immersed Boundary Methods have been developed, such as CutFEM, the Finite Cell Method, Unfitted Finite Elements, the Shifted Boundary Method, Phi-FEM, and Immersogeometric Analysis, to name a few.
This minisymposium will focus on variational types of Immersed Boundary Methods. Key topics include mathematical analysis, adaptivity, advanced quadrature, data structures, and parallel scaling of algorithms, along with integration with CAD models and non-standard geometric representations, and applications. The scope of this minisymposium is broad, including applications in solid mechanics, heat transfer, CFD, fluid/structure interaction, and other types of domain coupling. Additionally, the connection between Immersed Boundary Methods and meta-algorithms, such as those used in Uncertainty Quantification, Reduced Order Models, Machine Learning and Artificial Intelligence, Direct and Inverse Problems, and Topology Optimization, among others, will be addressed.
Keywords:
Embedded Domain Methods, Fictitious Domain Methods, Variational Methods
Iterative Coupling Algorithms and Enriched Finite Element Methods (e-FEMs) â such as the Generalized/Extended FEM â are powerful and complementary techniques for the solution of complex problems in computational mechanics. Both approaches are particularly well-suited for simulating multiscale phenomena, fracture and damage evolution, moving interfaces, and other challenging problems.
In recent decades, e-FEMs have matured significantly, with research focused on improving conditioning, robustness, and efficiency. Some of the recent developments include the Interface- and Discontinuity-Enriched FEMs, which present alternative techniques for handling both weak and strong discontinuities. Additionally, iterative coupling algorithms, such as the Iterative Global-Local algorithm (IGL), have gained attention for their ability to non-intrusively couple commercial and research software.
This minisymposium aims to bring together researchers from universities, national labs, and industry â from engineering, applied mathematics, and computer science â to exchange ideas and present recent advances in coupling algorithms and enriched finite element techniques. While contributions to all aspects of these methods and their implementation are invited, topics of particular interest include:
⢠verification and validation; accuracy, computational efficiency, convergence, and stability of e-FEMs and coupling algorithms.
⢠new developments for immerse boundary or fictitious domain problems, flow, and fluid-structure interaction, among others.
⢠applications to industrial problems with multiscale phenomena, localized non-linearities such as fracture or damage, and non-linear material behavior.
⢠acceleration techniques for coupling algorithms.
⢠coupling algorithms for multi-physics and time-dependent problems.
Keywords:
Enriched Finite Elements, Multiphysics problems, Non-intrusive coupling
Meshfree, particle, and peridynamic methods offer a new class of numerical methods that play an increasingly significant role in the study of challenging engineering and scientific problems. New and exciting developments of these methods often go beyond classical theories, incorporate more profound physical mechanisms, and serve as exclusive numerical tools for addressing computational challenges that were once difficult or impossible to solve by conventional methods.
The goal of this minisymposium is to bring together experts working on these methods, share research results, and identify emergent needs towards more rapid progress in advancing the important fields of meshfree, particle, and peridynamic methods. Topics of interest for this minisymposium include, but are not limited to, the following:
⢠Recent advances in meshfree, particle, and peridynamic methods, and their coupling with other computational methods such as isogeometric analysis, material point method, and the finite element method
⢠Immersed approaches for non-body-fitted discretizations
⢠Enrichment of basis functions for non-smooth approximations
⢠Integration of physics-based and data-enabled approaches
⢠Enhancement of meshfree, particle, and peridynamic methods by machine learning algorithms
⢠Strong form collocation methods
⢠Stabilization for under-integrated Galerkin methods
⢠Methods for coupling multiple physics and/or multiple scales
⢠Parallel computation, solvers, and large-scale simulations
⢠Recent advances in challenging industrial applications: modeling extreme loading events, additive manufacturing, and disaster mitigation
⢠Methods enabling a rapid design-to-analysis workflow
Keywords:
Meshfree, Particle Methods, Peridynamics
Polytope elements provide great flexibility with respect to meshing complex or evolving geometries. Polytope meshes can be refined locally and generated automatically using digital images or CAD data. In the latter case, the geometrical description of design objects and the numerical analysis of physical problems on these objects can be linked by using the same basis functions through the concept of isogeometric analysis. However, established numerical techniques such as the finite element method typically rely on triangular or quadrilateral element shapes and cannot be used on polytope elements straightforwardly.
Polytope discretizations also play an important role in geometry processing, where the aim is to display geometry as accurately as possible while simultaneously reducing the computational cost of graphics applications. This can be achieved by clustering best-fit regions, thus generating polygonal surface meshes. Since numerous open questions exist with respect to using polytope-based discretizations both from the computational mechanics and geometry processing point of view, one aim of this Minisymposium is to bring together researchers from both communities.
With respect to numerical solution techniques on polytope meshes various approaches exist. While contributions dealing with all types of computational solution techniques for physical problems on polytopal meshes are invited, special focus is placed on the scaled boundary finite element method (SBFEM) and its isogeometric variant (SB-IGA). The SBFEM in its original form is a semi-analytical technique that not only excels in facilitating the use of polytopal meshes, but also in representing radiation damping and stress singularities accurately. We therefore also welcome contributions that develop SBFEM or apply it to challenging problems, irrespective of the use of polytope elements. Possible areas of application include fracture mechanics, wave propagation, topology optimization, structural acoustics, linear and nonlinear solid mechanics, contact, plate and shell analysis, coupled problems and multi-scale modeling.
This Minisymposium aspires to gather researchers from geometry processing, numerical mathematics and computational mechanics who aim to combine forces to fully leverage the potential of polytope discretization and analysis methods and experts developing the SBFEM and other polytope-based techniques for fruitful moments of exchange of ideas and inspiration.
Keywords:
Mesh generation, Polytope elements, Scaled boundary finite element method, Virtual element method
Structure-preserving numerical methods have emerged as a cornerstone for the reliable simulation of nonlinear PDEs in applied sciences and engineering. For time-evolution problems, it is crucial that numerical schemes reflect fundamental thermodynamic properties such as energy dissipation, entropy production and preservation of conservation laws. These properties underpin the existence and uniqueness of physically meaningful weak solutions and are essential for robust error analysis through relative energy and entropy techniques. This minisymposium will focus on recent advances in the design and analysis of structure-preserving discretization techniques, including finite element, finite volume, and discontinuous Galerkin methods, as well as structure-aware time integration schemes. Emphasis will be placed on methods that guarantee discrete analogues of energy or entropy stability, enabling accurate and robust simulations even for highly nonlinear or multiscale problems. We will highlight both theoretical developments and practical applications, such as energy-stable schemes for phase-field models, entropy-consistent approaches for turbulent flows, and structure-preserving algorithms for coupled multiphysics systems. Special attention will be given to the interplay between discrete stability properties and a priori error estimates, as well as the challenges of implementing these methods in high-performance computing environments. By bringing together experts from numerical analysis, computational physics, and engineering, this minisymposium aims to foster a comprehensive discussion on the state-of-the-art in structure-preserving numerical methods and to identify promising directions for future research and applications.
Keywords:
Advanced Numerical Formulations, Nonlinear PDEs, Stable Approximations, Structure-Preserving Schemes
For problems in structural mechanics, locking is a well-investigated and a well-treated issue in the context of the finite element method (FEM). Based on the geometric characteristics, shear locking, membrane locking and trapezoidal locking can occur, while depending on the material properties, volumetric locking gets triggered. Recently, locking phenomena have been detected that only occur in nonlinear problems. There have been several ideas to alleviate the effects of locking, for instance, reduced integration techniques, the assumed natural strain method, the discrete strain gap method, the linked interpolation method, mixed methods, and many more. However, with emerging discretization techniques, this issue reappears, demanding further attention. In the recent decades, existing ideas from the framework of FEM have been extended, or novel approaches have been established to treat locking across these newly evolved discretization methods.
This mini-symposium aims to unify ideas and facilitate discussions on strategies explored to mitigate locking effects while using state-of-the-art discretization techniques to solve problems in small-strain and finite-strain solid and structural mechanics. This includes various topics like, but not restricted to, treatment of locking in the field of standard FEM, isogeometric analysis, scaled boundary FEM, virtual element method, meshless methods, collocation methods, lattice Boltzmann methods, machine learning-based approaches, and vector and matrix finite elements.
This mini-symposium is dedicated to Prof. Ekkehard Ramm on the occasion of his 85th birthday for his inspiring contributions in this field.
Keywords:
Computational Structural Mechanics, Discretization techniques, Locking, Machine Learning, Mixed Methods
The design of structure-preserving numerical methods for coupled problems has become an increasingly attractive research field in recent years. Structure-preserving schemes come with the promise of enhanced numerical stability and robustness. They can be viewed as extension of conserving schemes, which were previously developed in the context of conservative Hamiltonian systems with symmetry, to coupled dissipative systems. The coupling of several fields makes the design of structure-preserving schemes particularly demanding. On the other hand, the interaction of different fields may cause numerical instabilities when applying standard discretization techniques. Structure-preserving methods have the potential to correctly reproduce coupling effects in the discrete setting and are thus less prone to numerical instabilities.
The discretization in space and time of coupled problems is strongly affected by the way in which the underlying field equations are written, including the choice of variables. The structure of the underlying balance laws is built into specific descriptions such as GENERIC, metriplectic dynamics or the port-Hamiltonian formulation which thus might be of advantage for the design of structure-preserving schemes.
The present MS aims at bringing together researchers from different fields dealing with the design of structure-preserving discretization methods for coupled problems. Applications may focus on both dissipative solids as well as fluids. Specific applications may deal with, among others, large-strain thermo-elasticity, electro-magneto-mechanics, thermo-chemo-mechanics, or structure preservation for control.
Keywords:
GENERIC, Metriplectic Dynamics, Multi-physics, Port-Hamiltonian Systems
The Lattice Boltzmann (LB) method has emerged as a robust and efficient numerical method. Originally developed in the context of fluid mechanics, it continues to find more fields of application as researchers adapt it to their problems of interest. In this symposium, we would like to bring together researchers working on the use of the LB method to new problems of interest in applied mechanics and mathematics including, but not restricted to, solid mechanics, heat transfer, wave propagation, materials science, multiphysics problems, etc.
Researchers are encouraged to present new developments that will help to understand the theoretical underpinnings of LB methods, enabling the generalization of the method and understanding it as a powerful and general-purpose discretization technique for problems in computational mechanics and applied math. Also, results that provide insights into connections of LB to other well-established methods are welcome.
Keywords:
Advanced Numerical Formulations, Lattice Boltzmann
The concept of isogeometric analysis (IGA) has had an especially great impact in the field of thin structures, i.e., shells, plates and beams. The high continuity properties of isogeometric discretizations allowed for a variety of novel, highly efficient shell formulations, like rotation-free Kirchhoff-Love shells and hierarchic formulations for Reissner-Mindlin and solid shells, which show significant advantages over well-established shell formulations within traditional finite element methods. At the same time, shell analysis allows for the realization of the isogeometric paradigm, i.e., performing structural analysis directly on CAD geometries, which typically are surface models. Today, isogeometric shell analysis is already used in industrial applications, e.g. in the automotive industry. The efficient treatment of complex CAD geometries with multiple trimmed patches is a key aspect here and represents still a vital field of active research. Besides shells, a variety of novel and efficient formulations for plates and beams have been developed over the years, taking advantage of isogeometric discretizations.
The proposed mini-symposium invites all contributions from the field of isogeometric analysis of thin structures, both from method development and application. Typical topics are expected to be, but not restricted to:
⢠Isogeometric discretizations for shells, plates, membranes, beams, rods and cables
⢠Locking and un-locking in isogeometric structural elements
⢠Patch coupling
⢠Analysis of trimmed surfaces
⢠Coupling with solids and fluids
⢠CAD integration
⢠Industrial applications
Keywords:
Beams, CAD integration, Isogeometric, Locking, Plates, Shells, Thin Structures
This mini-symposium highlights recent progresses in advanced discretization techniques for applications in linear and nonlinear structural mechanics. The objective is to achieve computational optimality by enhancing accuracy, robustness, and efficiency in the simulation of complex structural responses.
Contributions are invited on theoretical developments and practical applications, including but not limited to:
⢠Polytopal mesh discretizations and generalized polygonal/polyhedral formulations;
⢠Virtual element formulations, including stabilization techniques and stabilization-free variants;
⢠Mixed finite element methods: assumed stress/strain formulations, Hybrid-Trefftz approaches, and variational principles such as HellingerâReissner and HuâWashizu;
⢠Accurate discrete strategies for addressing plasticity, fracture mechanics, and geometrical nonlinearities;
⢠Methods for alleviating locking and improving convergence behavior;
⢠Verification and validation approaches focusing on accuracy, computational efficiency, convergence, and stability;
⢠Industrial applications demonstrating the real-world impact of advanced discretization techniques.
Keywords:
Nonlinear Mechanic, Polytopal Meshes, Virtual Element Method, Advanced Discretization, finite elements, mixed formulations
In this minisymposium we seek to highlight challenging problems in computational solid mechanics that require rapid modeling building and mesh adaptivity for solution. We focus on finite element and other emerging discretization methods for large deformations and the accompanying inelasticity, contact, localization, and failure. Discussion will center on Lagrangian descriptions and the necessary computational components to resolve, preserve, and evolve the fields that govern these processes. Prototypical material systems may include, but are not limited to, polymers, structural metals, and biomaterials.
Topics of interest:
⢠Novel methods for discretization
⢠Tetrahedral, hexahedral, and other 3D element technology
⢠Local remeshing including topological changes and smoothing
⢠Field recovery and mapping of internal variables
Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energyâs National Nuclear Security Administration under contract DE-NA0003525.
Keywords:
discretization techniques, element technology, mesh adaptivity, solution strategies
Mixed-dimensional PDE systems arise when coupling unknown fields defined over domains of different topological dimension. They characterize a broad range of relevant problems in many scientific and engineering fields, such as fluid flows in fractured porous media, the design of very large floating sea structures, coupled cortex-cytoplasm dynamics in living cells. They can also be used to impose non-standard interface conditions on a lower-dimensional embedded subspace, e.g., through a Lagrange multiplier.
The aim of this minisymposium is to share and discuss the latest advancements, challenges and perspectives around the numerical approximation of mixed-dimensional PDEs, with a special interest in problems with moving boundaries and interfaces, and advanced discretization techniques. Topics of interest range from modelling aspects, mathematical analysis, computer implementation, solvers and innovative applications. We will address both parametric and immersed boundary approaches, including (but not limited to) arbitrary Lagrangian-Eulerian, unfitted finite element methods, phase-field, virtual element methods or formulations based on tangential differential calculus.
Keywords:
interface-coupled multiphysics, moving boundaries and interfaces, unfitted finite elements, Mixed-dimensional PDEs
Interfaces are ubiquitous in computational science, in problems ranging from material boundaries to different physical phenomena and mathematical approximations in fluid and solid mechanics. In this minisymposium, we will focus on âcut-cellâ methods, where a smooth background mesh is modified only in regions near a sharp interface. This approach has been applied in finite volume (âembedded boundaryâ), finite element (âNitscheâs methodâ, primal methods), and finite difference (âimmersed interfaceâ) approaches. The challenges for these methods have traditionally been higher-order accuracy and stability in the presence of small or arbitrary cut cells.
For accuracy, questions arise around boundary representation, which ranges from simple approaches using grid line intersections, to more complex representations using splines, implicit functions, and constructive solid geometry. We will present results from several different approaches, and discuss the difficulties encountered using higher-order bases, curved boundaries, and compatibility with higher-order interior methods. For example: What is the âcontaminationâ of a global solution when using less accurate boundary methods? And, is it possible to maintain global accuracy and convergence even with very complex boundaries?
The second challenge is around methods that are both higher-order accurate and demonstrably stable in the presence of small or arbitrary cut cells. For elliptic problems, stability implies positivity of matrix operators, norm-boundedness of solutions and eigenvalues, etc. Recent progress from both mathematical foundations and numerical linear algebra will be presented. Stability for hyperbolic problems is more subtle, in terms of time step restrictions and wave properties relative to a regular interior scheme. Traditional approaches address any time-step restrictions using cell merging (to avoid small volumes) or implicit methods, both of which have consequences for dissipation and preserving accuracy for wave-like phenomena and stability with moving boundaries. We will have several examples from more recent progress on mathematical theory to heuristics that are practically effective at maintaining both stability and accuracy, even when using explicit time integrator methods with very small cut cells.
We anticipate two sessions with a total of approximately 12 speakers.
Keywords:
cartesian grids, embedded boundary methods, voxel-based meshes, Cut-cells, structured grids, Wetting dynamics
The numerical approximation of Partial Differential Equations on polygonal and polyhedral meshes has been gaining growing interest within the scientific community. Polytopal grids offer a flexible and efficient framework for addressing challenges such as hanging nodes, diverse cell shapes within a single mesh, and non-matching interfaces. This versatility makes them particularly suitable for solving problems involving complex inclusions (as in geophysical modeling) or intricate and potentially deformable geometries, such as those found in basin and reservoir simulations, fluid-structure interaction, crack propagation, and contact phenomena. Over the past few years, numerous discretization techniques, such as the Virtual Element Method, the Hybrid High Order method, and the Discontinuous Galerkin method, tailored for polygonal and polyhedral meshes, have emerged, revealing strong interconnections. This mini-symposium aims to gather both experienced experts and early-career researchers to exchange recent advances and to foster a shared understanding and collaborative direction in the field.
Keywords:
advanced discretization techniques, complex geometries in simulations; , numerical methods for PDEs, Polygonal and polyhedral meshes
Numerical methods for nonlinear hyperbolic equations are known for being especially complicated to derive. This comes from the nature of the equations, which may develop nonregular solutions and nonuniqueness of the solution. To guarantee a proper convergence of the discretization, several structures of the system should be ensured: conservativity and entropy inequality are necessary conditions to ensure the right solution is captured. Additionally, the positivity of some physical variables (e.g. density, internal energy), is usually necessary not only for ensuring the physical relevance of the solution, but also for avoiding the computational failure of the code.
Among these considerations, some hyperbolic systems are known to include implicit differential constraints, also called involutions. This can be, for example, the conservation of the vorticity for the wave system (linked with the low Mach number accuracy problem), the conservation of the solenoidal character of the magnetic field for the Maxwell system or the Magnetohydrodynamics system, the curl of the deformation tensor for the hyperelastic system in solid mechanics.
These involutions are additional constraints with respect to the conservation laws. Numerical schemes that rely on the direct discretization of the conservation laws typically fail to respect these involutions, therefore innovative strategies must be investigated.
The aim of this Minisymposium is to review a large spectrum of recent advances in numerical methods for ensuring all the structures that were described in this abstract, and to present examples and applications e.g. to electromagnetics, fluid mechanics and structural mechanics.
Keywords:
Finite Element Methods, finite volume, hyperbolic systems, Scientific computing, Stable Approximations, Structure-Preserving Schemes
Polytopal methodsânumerical methods capable of handling general polygonal and polyhedral meshesâhave experienced significant growth over the past decade in both the mathematics and engineering communities. Notable examples include Virtual Elements, Hybrid High-Order methods, PolyFEM, Polytopal Discontinuous Galerkin methods, and Mimetic Discretizations.
These methods are particularly well-suited for addressing engineering problems in fluid and solid mechanics, owing to their remarkable flexibility in representing complex geometries, interfaces, and heterogeneous media. They also offer enhanced capabilities for mesh adaptivity due to the capability of managing efficiently hanging nodes, and elements with highly general shapes. Furthermore, polytopal methods have inspired ideas to design elements that are novel also on classical meshes, both in fluid and solid mechanics.
This minisymposium aims to bring together mathematicians and engineers to discuss recent advances in polytopal methods with a particular focus on applications in mechanics.
Keywords:
Numerical Methods in Mechanics, Polytopal Methods
In modern computational science and engineering, the ability to solve complex multi-physics and multi-domain problems has become essential across a wide range of applications, from fluid-structure interaction and geomechanics to biomedical modeling and additive manufacturing. A common challenge in such simulations is the need to couple independently generated computational grids that may differ in resolution, structure, or discretization schemes. The mortar method has emerged as a powerful and flexible approach for enforcing weak continuity across non-conforming interfaces, enabling accurate and stable coupling between subdomains without requiring mesh conformity. By enabling flexible grid generation and facilitating the coupling of different physical models across subdomains, mortar methods play a critical role in advancing the scope and fidelity of computational simulations.
This minisymposium provides an overview of recent advancements in the mortar finite element method as a robust framework for domain decomposition and grid coupling. The mortar approach supports heterogeneous discretizations, making it particularly well-suited for applications such as contact problems in solid mechanics, reservoir modeling in subsurface flow, and coupling structured and unstructured grids in large-scale fluid dynamics simulations. The focus is both on the mathematical formulation and on representative examples that demonstrate the versatility and effectiveness of mortar methods in practical engineering problems. Special attention is given to the role of mortar spaces, projection operators, and interface integrals in ensuring numerical accuracy and stability.
Keywords:
Multi-Domain, Multi-Physics, Non-Conforming Grids, Mortar Method
Particle in Cell (PIC) methods have been used since the beginning of plasma simulations to approximate the Vlasov-Maxwell equations or related reduced models. Even though they are noisier than grid based discretizations, they are still more efficient for 6D simulations as they are not as much affected by the curse of dimensionality. As for other models, it has been recognized that the exact preservation of key invariants such as the divergence constraints in Maxwellâs equations is essential for accurate long time simulations. These constraints actually come from the noncanonical Hamiltonian structure of the Vlasov-Maxwell equations. Field discretizations based on Discrete Exterior Calculus (e.g. Finite Element Exterior Calculus or Mimetic Finite Differences on dual grids) coupled to a particle method for the Vlasov equation yield in a natural way to a finite-dimensional noncanonical Hamiltonian system that can then be solved using strategies from geometric numerical integration of ODEs [1,2]. In particular splitting methods, either on the Hamiltonian or on the Poisson bracket are good choices.
On the other hand, adding collisions to the system in terms of the Landau or some approximate models, stills conserves energy, momentum and energy at the continuous level and adds the important property, namely the Boltzmann H-Theorem that stipulates that the entropy must grow. The full Vlasov-Maxwell-Landau system possesses a so-called metriplectic structure. There have been recent works devoted to the discretization of the collision operator that preserve this metriplectic structure including the H-Theorem at the semi-discrete level [3,4].
The mini-symposium will focus on the recent advances in the field as well in the development of original numerical methods as in the development of Open Source code infrastructures.
Keywords:
structure preserving, Particle in cell (PIC)
This mini-symposium addresses both theoretical and practical aspects of particle-based computational methods that can be effectively used for solving a variety of problems in solid mechanics, fluid mechanics, fluid-structure interaction, heat transfer, and many others. Contributions dealing with the discrete element method (DEM), the particle finite element method (PFEM), the smoothed particle hydrodynamics method (SPH), the material point method (MPM) and the moving particle semi-implicit method (MPS), among others, are welcome. Likewise, the coupling of these methods with other established numerical procedures, such as the finite element method, the finite difference method and meshless techniques, is considered.
Keywords:
Discrete Element Method, MPM, MPS, Particle Methods, PFEM, Smoothed Particle Hydrodynamics
Meshfree and particle methods have been developed in the field of computational mechanics by taking advantage of their robustness against dynamic changes in free surfaces and propagation of discontinuities. While the advantages of these methods derive from their meshless nature, these features can conversely pose difficulties in the treatment of boundary conditions and in problems of multiphase flows with high density ratios. The purpose of this mini-symposium is to provide discussions for researchers of the meshfree and particle methods to share their recent knowledge and advanced insights. The topics are mathematical theory, discretization schemes, multi-resolution techniques, multi-physics analysis, boundary conditions, accuracy, adaptive analysis, parallel processing, large scale analysis, applications, verification and validation etc. for the particle methods.
Keywords:
Fluid Dynamics, Mesh-free Method, Multi-physics Simulation, Particle Method
This minisymposium brings together researchers studying geometric mechanics and structure-preserving discretizations at the continuous level, discrete level, and the interface between the continuous and discrete levels. We welcome contributions related to all areas of geometric mechanics and structure-preserving discretizations including (but not limited to) novel geometric mechanics formulations; structure-preserving spatial, temporal and spatiotemporal discretizations; structure-preserving reduced order models; structure-preserving scientific machine learning; and geometric optimal control.
Keywords:
discrete exterior calculus, finite element exterior calculus, geometric mechanics, Metriplectic Dynamics, Port-Hamiltonian Systems, Structure-Preserving Discretization, symplectic integrators, Variational Methods
200
Atomistic, Nano and Micro Mechanics of Materials
The advancing of nanotechnology has enabled the fabrication of high-performance functional materials and the continuing miniaturization of mechanical devices or systems. To facilitate the manufacturing and applications of nanoscale materials, it is vital to understand their mechanical properties. There have been extensive experimental, theoretical, and computational efforts at atomistic scale to understand the mechanical behaviours of nanomaterials.
Besides, the thermal transport property of the nanomaterials is another fundamental characteristic that determine their usages. Depending on the application, materials are required to have a high thermal conductivity or a strongly suppressed thermal conductivity. For instance, for energy saving in both residential and commercial buildings and thermoelectric devices, there has been a continuing search for high performance materials with a low thermal conductivity. In comparison, a high thermal conductivity is required for the electronic packing to enable efficient heat removal and transfer.
The diversity of low dimensional nanomaterials has provided a great potential to construct novel nanostructures with required mechanical and thermal performance. This mini-symposium intends to bring the recent progress on atomistic simulations for the mechanical and thermal transport properties of nanomaterials, which serve as effective tools to guide experiments or predict novel nanomaterials.
Keywords:
first principle calculations, molecular dynamics simulation, nanomechanics, nanoscale thermal transport
The extreme miniaturization in modern technology demands a deeper understanding of the unconventional, fluctuation-dominated mechanics of materials at micro- to nano-scales [1,2]. At these scales, dislocation-mediated plasticity undergoes a radical shift: sub-micron metallic samples exhibit high, but highly scattered, yield strengths. However, this behavior is compromised by intermittent strain fluctuations that undermine forming processes and threaten structural stability. As a result, a comprehensive theoretical framework that quantitatively links material characteristics to mechanical fluctuations has yet to be established. This poses significant challenges to conventional engineering models of plasticity, which were developed to describe the smooth macroscopic behavior of bulk materials, whether in isotropic, anisotropic, or crystal plasticity contexts.
This session welcomes modeling tools that address these fluctuation-dominated mechanics and provide insight into the behavior of materials at micro- to nano-scales. Contributions that offer innovative theoretical frameworks or computational approaches to better understand and predict dislocation-mediated plasticity and its effects on material performance are highly sought after.
Keywords:
Dislocation-Mediated Plasticity, Micro- and Nano-Scale Materials, Fluctuation-Dominated Mechanics
The mathematical formulation of natural and synthesized material for modelling plastic behavior usually consists of differential equations, algebraic equations, and algebraic inequalities. This mathematical combination of ingredients causes the numerical difficulty, hence special treatment for computation is needed. The sub-stepping integration, the return-mapping integration, the angle-based integration, and the exponential map integration have been developed from this background. In recent years, mathematical modelling on plastic behavior of materials pinpoints the material behavior under multiaxial cyclic loading. Under the multiaxial cyclic loading, the yield surface of the material translates, expands, and distorts. Towards the accurate prediction of yield surface evolution, several new plasticity models have been proposed and studied in recent years. Current work suggests, combining the kinematic, the isotropic, and the distortional hardening rules would a crucial point. However, this development greatly increases demands on computing power and require careful examination of numerical convergence. Furthermore, dedicated optimizations of numerical implementation are crucial for computational efficiency of mixed hardening predictions, numerical verification, and calibration.
This MS provides a platform for researchers to exchange his/her work which is especially related to but not limited to the investigation on the computational plasticity in metals, polymers, biological materials, and other materials. The computational investigation includes not only the continuum-based but also the particle-based approach whereas the computational modelling contains single-scale as well as multi-scale simulations. Theoretical research and numerical study are all welcome.
Keywords:
Cyclic Loading, Hardening and Softening, Multiaxial Loading, Plasticity, Yield Surface Evolution
Soft materials comprise a diverse class of substances that are highly deformable and exceptionally sensitive to external stimuli. Many of these materials offer unique advantages, including biocompatibility, superior energy absorption, and the capacity to finely tune their mechanical properties. Such characteristics make them central to emerging technologies spanning biomedical engineering, flexible electronics, soft robotics, and energy harvesting.
Understanding and predicting their mechanical response is crucial for the design of next-generation devices. Unlike traditional engineering materials (e.g., metals or ceramics) soft materials exhibit complex, nonlinear, and inelastic behaviors. These may include large deformations, viscoelasticity, plasticity, permanent set, anisotropy, and phenomena like the Mullins effect. Capturing these behaviors requires modeling strategies and computational techniques that extend beyond classical approaches.
This symposium will assemble leading researchers in computational mechanics who are pushing the frontiers of soft material science and engineering. Topics will range from the formulation of robust constitutive models for nonlinear elasticity, rate-dependent responses, and irreversible effects, to the development of advanced numerical methods. These methods include higher-order discretization schemes, efficient constitutive integration algorithms, and multiphysics coupling frameworks. Emphasis will be placed on cutting-edge applications where computational innovations can accelerate design, optimization, and deployment. Examples include soft robotic actuators, biomedical devices, tissue scaffolds, adhesives, and drug delivery systems.
Keywords:
Biomechanics, Computational Methods, Inelasticity, Multiphysics, Soft materials
This mini-symposium will present recent advances in machine learning (ML)-assisted multiscale design of materials [1], emphasizing computational mechanics and data-driven innovations across a wide range of applications. Early ML approaches accelerated materials discovery by screening large databases and uncovering complex links between atomic structures and properties. More recent developments in high-accuracy universal ML interatomic potentials and generative models have greatly improved predictive capabilities and sped up the discovery of materials across scales, from atoms to bulk components.
ML now enables first-principles-level predictions of electronic, optical, and mechanical properties for large systems, opening new possibilities for structural, functional, and multifunctional materials. At the mesoscale, ML-assisted microstructure reconstruction and physics-informed models for solving partial differential equations are advancing the understanding of microstructureâproperty relationships critical for multiscale modeling.
The integration of ML platforms into autonomous laboratories, combining quantum mechanical simulations, large language models, and experimental testing, is transforming the traditional approach to materials synthesis. Alongside showcasing these advances, the symposium will address ongoing challenges in scalability, interpretability, and coupling with multiphysics simulations.
This symposium will invite contributions in:
1. ML interatomic potentials and surrogate models for accelerated evaluation of materials properties.
2. Generative and inverse design approaches for structural, functional, and multifunctional materials.
3. ML-assisted multiscale modeling for predicting and linking properties across atomic, microstructural, and macroscopic scales.
4. Autonomous and closed-loop materials discovery platforms, integrating simulation, optimization, and experimental validation.
Keywords:
interatomic potentials, materials design, multiscale modeling, Machine learning
The ever-growing world population demands material consumption and drives material discovery. Material design is gradually breaking through the assumption of continuum system. The bottom-up combination of different microstructures into complex building blocks has opened unprecedented opportunities in materials science and engineering. Micro/nano structural materials can extensively take advantage of the physical, chemical and mechanical properties of micro/nanoscale units, achieving more excellent mechanical properties and functions compared to traditional materials. Traditional mechanical studies have been unable to establish the theoretical framework of advanced materials at multiscale, resulting in more common scientific problems emerged from interdisciplinary fields. Although modern physics edged mechanics out into the wilds of engineering, we should note that there is plenty of room in the cross field of mechanics and advanced materials. The development of computational multiscale mechanics has covered from the physical edge at atomistic scale to the framework of continuum theories. It is of the utmost importance to extend the models and methods of computational multiscale mechanics that would provide in-depth understandings for the structure-property-function relationships of materials. The topics of this minisymposium focus on the computational multiscale mechanics for microstructure design of hierarchical materials, low-dimensional materials, biomimetic structural materials, biomass-derived sustainable materials and so on. Toward the digital age and sustainable future, this minisymposium seeks to boost the computational mechanics-guided design for high-performance materials.
Keywords:
hierarchical microstructures, mechanics design of materials, micro-nano mechanics, multiscale mechanics, theoretical modelling
Materials in a fusion reactor must withstand intense radiation, high temperatures, and interactions with the plasma. This MS focuses on to understand and predict the behavior of materials in the extreme environment of a fusion reactor, using computational methods such as density functional theory, molecular dynamics, and kinetic Monte Carlo, etc. The topics of this MS include but are not limited to:
1. Fundamental Irradiation Damage Mechanisms: The high-energy neutrons produced by the fusion reaction can produce irradiation defects after collision cascades. This topic covers understanding of the fundamental mechanisms of formation of irradiation defect and evolution of the materialâs microstructure.
2. Plasma-Material Interactions (PMI): The materials that directly face the super-hot plasma are exposed to intense heat and particle bombardment. Sputtering, erosion and the hydrogen/helium retention in the material are a major safety and fuel-cycle concern. This topic includes prediction of change in surface morphology over time and understanding of diffusion and trapping of defects.
3. Mechanical and Thermophysical Property Changes: This topic explores how the irradiation damage and PMI affect the material's properties, including changes in hardness, ductility, thermal conductivity, etc.
Keywords:
Irradiation Damage, Nuclear Fusion Materials, Plasma-Material Interaction
300
Biomechanics and Mechanobiology
Computational mechanics and numerical methods play an increasingly significant role in the study of biological systems at the organism, organ system, organ, tissue, cell, and molecular scales. New and exciting applications of computational mechanics go beyond the classical theories and incorporate biomechanical mechanisms inherent in biology such as adaptation, growth, remodelling, active (muscle) response, and inter- and intra-patient variables. Synergies among fundamental mechanical experiments, multi-modal imaging and image analyses, new mathematical models and computational methods enable studies of, e.g., microphysical (mechanobiological) cellular stimuli and response, structure-function relationships in tissues, organ and tissue integrity, disease initiation and progression, engineered tissue replacements, and surgical interventions.
The goal of this minsymposium is to promote cross-fertilization of ideas and collaborative experimental and numerical efforts towards more rapid progress in advancing the overall field of computational biomechanics. To this end, contributions considering the following topics are particularly welcome: coupled analyses of chemo-mechanical processes; methods coupling multiple scales and/or multiple physics; growth and remodelling of biological tissues; characterization and impact of inter- and intra-patient variability; applications with clinical impact or potential clinical impact; new constitutive models; mechanobiology and cellular mechanics; applications of medical images and image analyses in mechanics; mechanics of pathological processes; and experimental methods and computational inverse analyses towards model calibration.
Keywords:
Biomechanics, Computational Mechanics, Growth and Remodelling, Mechanobiology, Multi-Scale, Soft Tissues, Physical Experiments
The brain is one of our most complex organ systems. Despite decades of intense research, many fundamental processes and diseases are still not fully understood. While a large body of literature exists regarding the role of chemical signaling in regulating brain function, only recently the important contribution of mechanical stimuli has been discovered. To fully understand brain development, homeostasis, and pathology, it is therefore key to explore mechanics as an important puzzle stone. This endeavor is challenging as brain tissue is not only ultrasoft, biphasic, and highly heterogeneous, but also continuously changes its microstructural composition and architecture as well as its mechanical properties â in close relation to evolving brain function. Living brain cells actively sense and respond to their mechanical environment, leading to sophisticated coupling effects.
Computational models based on nonlinear continuum mechanics can valuably complement experimental studies to systematically understand the mechanisms underlying mechanics-regulated biological and biomedical processes [1]. In addition, they can allow for data transfer across species and scales, and eventually pave the way for personalized clinical predictions. This minisymposium will cover novel experimental and modeling approaches, including computational solid and fluid mechanics as well as data-driven modeling. It is targeted toward advancing our understanding of brain function and dysfunction, and eventually diagnosis and treatment of neurological disorders.
REFERENCES
[1] S. Budday, T. C. Ovaert, G. A. Holzapfel, P. Steinmann and E. Kuhl, Fifty Shades of Brain: A Review on the Mechanical Testing and Modeling of Brain Tissue, Archives of Computational Methods in Engineering 27 (2020) 1187â1230. doi:10.1007/s11831-019-09352-w
Keywords:
Computational Mechanics, Coupled problems, Brain mechanics
The digital representations of the human body within the field of biomechanics and mechanobiology become progressively sophisticated with the evolution of computational methods in applied sciences and engineering. As a result, there is a growing need to review, survey and discuss state-of-the-art advances and identify key research directions that will shape the future development of the human body models (HBMs). To tackle these challenges, this Minisymposium covers the emerging fields related to both computational approaches finite element method (FEM) and multibody (MB) dynamics, and addresses the following topics of human body modelling:
1. Passive and Active HBMs. In addition to all the existing features of passive models, active HBMs incorporate muscle activation and neuromuscular control systems. The Minisymposium aims to explore how these complementary muscular system sub-models and codes evolve and converge to address complex biomechanical challenges of the human reflex and movement modelling.
2. Neuromuscular control algorithms for Active HBMs. The Minisymposium intends to investigate the implementation of various neuromuscular control strategies and algorithms with subsequent integration to anatomically accurate HBMs. Different solutions to unique challenges that require interdisciplinary collaboration are accepted, ranging from PID controllers and muscle length feedback to EMG-based control and Reinforcement Learning.
3. Multi-industry HBM applications. The Minisymposium seeks to survey deployment of different types of HBMs in multiple industries like automotive safety, healthcare, medical device development, sports biomechanics, ergonomics and workplace safety.
4. HBM Personalisation. The Minisymposium pursues to analyse applications of patient-specific modelling techniques, which incorporate individual anatomical, physiological, and anthropometric parameters to create tailored HBMs. Such models can simulate individual physiological responses, enabling precision medicine and personalised safety systems.
Keywords:
Biomechanical Applications, Computational Methods, Human Body Models, Neuromuscular control, Biomechanics
Bone is a mechanosensitive, hierarchically structured tissue whose function and adaptation are governed by mechanical and biological processes acting across multiple spatial and temporal scales. Computational modelling has become an indispensable tool to investigate and integrate these processes: from molecular signalling and cellular mechanotransduction to tissue-level adaptation and whole-organ biomechanics.
This minisymposium will focus on recent advances in computational modelling of bone growth and adaptation that address the challenges of multiscale modelling frameworks, simulation procedures, graphical representation, and validation against experimental data. We aim to highlight approaches that connect fundamental mechanobiological mechanisms with boneâs macroscopic behavior, including continuum- and microstructural modelling, image-based simulations, and multiscale or multiphysics frameworks.
Particular emphasis will be placed on methodologies that bridge scales, either through homogenisation and nested modelling strategies, or via data-driven and machine learning techniques. Contributions that integrate experimental data, explore numerical and computational challenges, or propose innovative model validation strategies are especially encouraged.
By bringing together researchers from computational biomechanics, applied mechanics, mechanobiology, and related fields, the symposium aims to foster interdisciplinary exchange and contribute to a deeper understanding of boneâs mechanobiological behaviour in growth and adaptation. The ultimate goal is to support the development of predictive, mechanistic models of bone that are not only biologically and physically grounded, but also computationally robust and extensible across scales.
Keywords:
Bone Growth and Adaptation, Bone Mechanics, Mechanobiology, Computational Biomechanics, Multiscale Modelling
The integration of experimental and computational methods in biomechanics is critical to support the development of personalized diagnostics and treatments. In silico models serve as powerful tools for hypothesis testing and exploration of scenarios that are difficult or impossible to reproduce experimentally. To ensure clinically meaningful predictions, such models must be carefully calibrated and validated against experimental data. This process is especially critical in medical applications, where model accuracy directly impacts patient outcomes. Calibration requires determining model parameters, many of which cannot be measured directly and must be inferred through inverse analysis methods. Beyond parameter estimation, a key challenge lies in accounting for uncertainties inherent in experimental measurements, model structure, and parameter variability. These uncertainties must be quantified to ensure that predictions are robust and reliable. In the context of biological tissues, the use of advanced statistical tools and carefully designed workflows is mandatory to cope with intrinsic heterogeneities and anisotropies. However, by combining experimental rigor with computational power, researchers can enhance model fidelity, inform targeted experimentation, and expedite the translation of insights into clinical practice.
This minisymposium aims to showcase recent advances and applications at the interface of experimental and computational biomechanical modeling. We welcome contributions in theoretical developments, computational methods, experimental data analysis, and applied case studies. Topics of interest include, but are not limited to:
- Methods for parameter estimation and inverse problems in biomechanics.
- Methods for uncertainty quantification in bioengineering.
- Validation frameworks for clinical translation.
- Data assimilation and model-informed experimental design.
- Case studies integrating experimental and computational approaches.
Keywords:
Data Assimilation, Inverse Analysis, Uncertainty Quantification, Biomedical Engineering
ABSTRACT
Cancers are highly heterogeneous diseases that involve diverse biological mechanisms, interacting and evolving at various spatial and temporal scales. Multiple experimental, histopathological, clinical, and imaging methods provide a means to characterize the heterogeneous and multiscale nature of these diseases by providing a wealth of temporally and spatially resolved data on cancer development and response to therapies. These multimodal, multiscale datasets can be exploited to constrain data-driven and biophysical models of cancer dynamics in preclinical and clinical settings [1-3]. These models can then be used to test biological and clinical hypotheses, produce individualized predictions to guide clinical decision-making, and, ultimately, to design optimized monitoring and treatment plans.
The overall goal of this minisymposium is to present and discuss recent developments in computational models and methods for predicting cancer progression and treatment response, with special focus on the following areas: (i) biology-based mechanistic models of cancer growth and treatment in vitro and in vivo; (ii) computational methods for model initialization, parameterization, and patient-specific simulation; (iii) personalized optimization of monitoring plans and treatment regimens; (iv) uncertainty quantification and model selection methods; (v) hybrid strategies combining machine learning and mechanistic modelling; and (vi) digital twins in cancer research and clinical oncology.
REFERENCES
[1] G. Lorenzo, et al. (2024). Patient-specific, mechanistic models of tumor growth incorporating artificial intelligence and big data. Annu. Rev. Biomed. Eng., 26(1), 529-560.
[2] N. Cogno, et al. (2024). âAgent-based modeling in cancer biomedicine: applications and tools for calibration and validationâ, Cancer. Biol. Ther., 25, 1.
[3] M. Alber, et al. (2019). Integrating machine learning and multiscale modelling - perspectives, challenges, and opportunities in the biological, biomedical, and behavioral sciences. npj Digit. Med., 2, 115.
Keywords:
cancer forecasting, computational oncology, Digital Twins, inverse problems, machine learning, mechanistic modeling, model calibration, model selection, optimal control theory
The numerical modeling of the cardiovascular system and its pathologies has become
increasingly relevant in clinical practice, driven by significant advances in computational
biomechanics and related disciplines. The growing reliability and accuracy of developed
numerical tools grant access to new data, deepening our understanding of the cardiovascular system and its complexities. The insights gained from mathematical modeling can guide decision-making, enable the comparison of treatment strategies, and inspire the design of medical devices [1, 2].
This mini-symposium focuses on emerging topics in cardiovascular system modeling, from
methodology development to clinical translation, by means of physics-based and data-driven approaches. The scope ranges from classical continuum-mechanics-based methods to reduced order modeling including machine learning techniques, sensitivity analysis, uncertainty quantification, and other statistical methods applied to the cardiovascular system. Potential contributions may cover, but are not limited to, blood flow, tissue modeling, drug transport, electrophysiology, thrombus formation, tissue growth and remodeling, virtual treatment, hemodynamic indicators, patient-specific models, medical device modeling, stent deployment, heart valve dynamics, and numerical modeling of pathologies such as aneurysm, dissection, and in-stent restenosis.
The proposed mini-symposium aims to bring together experts across many fields, including
applied mathematics and biomedical engineering, to bridge the engineering-clinical interface through exchange of the latest results and discuss future challenges to advance personalized computational medicine. We seek to cultivate a dynamic and forward-thinking environment in the biomechanics and cardiovascular modeling community to foster discussion, encourage interdisciplinary collaboration, and inspire innovation.
REFERENCES
[1] E.L. Schwarz, et al., Biophysics Rev. 2023; 4(1):011301. doi:10.1063/5.0109400.
[2] I. Fumagalli, et al., Transl. Pediatr. 2024; 13(1):146-163. doi:10.21037/tp-23-184.
Keywords:
Cardiovascular Modeling, Computational Biomechanics, Model Order Reduction, Patient-Specific, Sensitivity Analysis, Uncertainty Quantification, Vascular Device
This mini-symposium focuses on computational models for the mechanobiology and biomechanics of cells, vesicles and biomembranes, and their structural constituents such as lipid bilayers, protein filaments, cytoplasm, cytoskeleton, organelles and nuclei. Due to the diversity of these components and due to the different length and time scales involved, a wide range of computational approaches can be considered. Examples are continuum models â like finite element and meshfree methods â structural models â like shell or beam models â and molecular models â like coarse-grained or all-atom molecular dynamics. Also, machine learning techniques can be expected to play an increasingly important role in the study of cells, vesicles and biomembranes. This session aims at bringing together researchers working on these topics and providing them with a forum for discussion.
Possible topics to be discussed in this symposium are:
⢠deformation of cells (as well as vesicles, biomembranes and their constituents)
⢠interaction, contact and adhesion of cells
⢠tethering and budding of cells
⢠cell motility
⢠diffusion into cells and across membranes
⢠protein binding to membranes
⢠virus penetration into cells
⢠cell division/rupture
⢠mechanosensing and mechanotransduction of cells
⢠multiscale modeling of cells
⢠cellular signalling pathways
⢠medical imaging of cells
Keywords:
Computational Mechanics
Intracranial aneurysms (IAs) are oblate- or prolate-shaped dilations of brain arterial vessels that are prevalent in about 3% of the adult population. If an IA ruptures, an aneurysmal subarachnoid haemorrhage develops, with a high risk for permanent disability or fatal outcome [1]. Therefore, effective treatment of unruptured IA is essential to ensure disruption of the inflow and prevent rupture. Endovascular treatments include placement of flow-modulating devices (FMDs) such as coils (CO), flow diverter (FD), woven endo-bridge (WEB), and most recently, Contour (CNT), which assures either intrasaccular flow disruption (CO, WEB, and CNT) or intravascular flow diversion (FD). Shape-memory materials, such as Nitinol, are used in these devices to enable crimping for catheter delivery and self-expansion at the aneurysm site. FMDs reduce blood flow into IA and facilitate healing. The efficiency of FMDs can be studied using in silico biomechanical structural models and homogenized poro-elastic approximations, which are used to predict the fully resolved velocity field simulations of the blood flow using advanced numerical approaches such as the Lattice Boltzmann Methods for the incompressible Navier-Stokes equations and patient-specific IA geometries. In addition, in vitro studies can be used to assess the performance of FMDs [3]. However, there is still a research gap in the interdisciplinary perspectives and joint ventures to tackle this crucial biomedical problem. This mini-symposium aims to highlight the recent developments, ongoing challenges, and prospects in the biomechanical modeling, 3D-printing, and flow dynamics of these endovascular devices for patient-specific IAs.
[1] J. L. Brisman et al., Cerebral aneurysms, The New England Journal of Medicine 355, 9 (2006) 928-939.
[2] M. Frank et al., Numerical simulation of endovascular treatment options for cerebral aneurysms, Annals of 3D Printed Medicine 47, 3 (2024) e202370007.
[3] M. S. Pravdivtseva et al., The effect of the size of the new contour neurovascular device for altering intraaneurysmal flow, Interventional Neuroradiology 31, 1 (2023) 49-56.
Keywords:
Coils, Contour, Flow Diverter, Intracranial Aneurysms, Endovascular devices
Soft robotics, characterized by compliant materials and embodied intelligence, has demonstrated significant potential across various engineering applications, from medical robotics to adaptive manufacturing systems. Computational modeling plays a pivotal role in understanding and harnessing the complex interactions inherent in soft robotic systems. This mini-symposium will showcase recent advances in computational techniques specifically tailored for modeling, simulation, and optimization of soft robotic structures and their dynamic behaviors. Key topics include numerical methods for large deformations, multiphysics modeling, dynamic simulation frameworks, control strategies, and integration of machine learning approaches. Drawing upon foundational contributions such as [1], the symposium aims to facilitate discussions that deepen the understanding of embodied intelligence through innovative computational strategies. The mini-symposium seeks to encourage interdisciplinary interaction, bridging communities from robotics to computational science, among others.
REFERENCES
[1] Mengaldo G, Renda F, Brunton SL, Bächer M, Calisti M, Duriez C, Chirikjian GS, Laschi C. A concise guide to modelling the physics of embodied intelligence in soft robotics. Nature Reviews Physics. 2022 Sep;4(9):595-610.
Keywords:
Computational Mechanics, Computational Fluid Dynamics, Soft Robotics
Computational modeling and simulation-based approaches in cardiovascular biomechanics and biomedicine have seen rapid progress in recent years. Computational approaches provide a non-invasive modality for understanding the underlying mechanics of cardiovascular diseases, as well as guiding device design and treatment planning. The future of computational cardiovascular biomechanics lies in patient-specific simulation of real disease events, enabling simulation assisted diagnostics, device design and deployment, and treatment planning decisions. The primary challenge here is that patient-specific phenomena involve a synergistic interplay of multiple underlying physical, mechanical, and chemical processes, coupled to each other across several spatial and temporal scales. Concurrently, the availability of high-resolution imaging and clinical data, and recent innovations in data-driven models and artificial intelligence/machine learning, have enabled new avenues for advancing patient-specific predictive models of cardiovascular phenomena. Together, multiphysics and data-driven modeling have thus slowly emerged as a new frontier in high-fidelity modeling of cardiovascular systems, aiming to resolve physiological and pathological phenomena in real patient-specific scenarios. Advancements in this field call for inter-disciplinary research efforts spanning beyond current multiscale computational mechanics approaches in cardiovascular biomechanics.
This minisymposium will bring together scientists working across various domains to provide a platform for discussing the state-of-the-art and future directions in multiphysics, multiscale, and data-driven modeling of cardiovascular systems. Fundamental and applied contributions from a wide range of topics focusing on theoretical and computational approaches for cardiovascular phenomena will be discussed. The term multiphysics in this context refers to coupled physical interactions including not only fundamental fluid and solid mechanics, but also multiscale transport phenomena, biological growth and remodeling, electrophysiology, biochemical interactions including drug delivery and other related aspects. Data-driven approaches include artificial intelligence, machine learning, data-augmented models, image analytics, uncertainty quantification, and related techniques.
Keywords:
Biomechanical Applications, Biomechanics, Biomedical Engineering, Cardiovascular Modeling, Computational Biomechanics, Data-driven Models, Mechanobiology, scientific machine learning
Thin material structures come in a great variety: they range from fluid to solid, can be passive or active, within linear response or strongly nonlinear, relaxed or pre-stressed, etc. They display fascinating and often unexpected mechanical and rheological properties that influence the behaviour of living objects at length scales ranging from cell membranes to tissues and organs. The peculiar properties of these structures have a geometric origin and various continuous theories have been developed to model the underlying geometric interactions. However, independent of any particular model or material under consideration, there is still confusion about how structures that have long been well-understood in a fixed flat space translate to an evolving space. Already for static structures, classical bulk elasticity leads to a hierarchy of models for plates and shells. Considering their evolution provides additional possibilities. Furthermore, there is not only a larger variety of models, compared with bulk materials, also irregularities play an even more profound role. Topological defects for example have been postulated as key players in morphogenesis [1]. We here consider thin material structures with inhomogeneities; with internal degrees of freedom, with microstructures and defects; and with singularities. Prominent examples are the arrangements of cells in epithelial tissue, and how cell shape and neighbour relations influence bending properties [2], viscous active shell theories to model the actin cortex [3] or surface liquid crystal models from which coarse-grained models for asymmetric plasma membranes can be derived [4]. The main focus of the minisymposium is on modelling aspects addressing the interplay between the internal structures (cells, filaments, lipids, ...) and the geometric interactions determining the macroscopic shape evolution. We will bring together different disciplines: geometric partial differential equations, homogenization, numerical analysis, mechanics, biophysics and scientific computing and showcase the state of the art in mathematics of thin materials structures with applications in biology. We plan with 12 presentations.
Keywords:
moving boundaries and interfaces, multiscale materials, Simulation methods for thin structures, Thin Structures
A comprehensive understanding of soft tissue biomechanics requires modelling approaches that bridge multiple spatiotemporal scales and simultaneously capture the intricate coupling between mechanical, chemical and biological processes [1]. Conventional models fail to capture this range, offering only partial insight into the complex biophysiochemical interactions governing tissue functions and material behaviour [1]. Nonetheless, advances in high-resolution imaging and physics-based simulations are on a promising trajectory to bridge this gap [2].
This minisymposium will provide a cohesive forum to showcase cutting-edge pipelines that integrate image-derived multiscale tissue structures with coupled biochemical and biophysical simulations. Key themes include: (1) modelling fluid and mass transport in complex biological systems (e.g. brains, tumours); (2) reconstructing patient-specific tissue architectures across scales from imaging data for numerical modelling; and (3) linking cellular biophysiochemical processes to tissue and organ scale functions. The objectives are: (1) present cutting-edge computational strategies that address the length- and time-scale bottlenecks in biomechanics and clinical modelling; (2) accelerate the deployment of image-based high-fidelity simulations in, e.g. tissue engineering, regenerative medicine and biomedical device design; and (3) catalyse meaningful cross-disciplinary collaboration across various disciplines.
The organisersâ complementary expertise will create a unique forum at the interface of biomechanics and hybrid-scale modelling. Dr. Tian Yuan integrates computational biomechanics with medical imaging and clinical translation; Dr. Suhaib Ardah investigates interfacial fluid-structure dynamics in soft matter; and Prof. Moran Wang is a pioneering figure in multiscale transport and porous flow modelling. Through active discussion and knowledge exchange, participants will gain valuable insights into current challenges and opportunities for translating multiscale & multiphysics biomechanics modelling to clinical applications.
References
[1] S.Y. Sun, H. Zhang, et al., and X.Q. Feng, Bio-chemo-mechanical coupling models of soft biological materials: A review. Advances in Applied Mechanics, 55, (2022) 09-392.
[2] T. Yuan, N. Pecco, et al., and D. Dini, Decoding brain interstitial transport in vivo: A fully validated bottom-up mechanistic prediction framework. BioRxiv, (2025), 2025-03.
Keywords:
Computational Biomechanics, Image-based Modelling, Multiscale Modelling, Transport Dynamics
Cell migration is a fundamental biological process that plays a critical role in diverse phenomena such as embryonic development, the immune response, wound healing, and cancer metastasis. Understanding and predicting cell migration in complex biological environments pose significant challenges that require advanced mathematical modeling and numerical simulation methods. Recent years have seen considerable advances in the development of continuum and discrete models for cell migration, including phase-field methods, finite element methods, agent-based methods, immersed boundary methods, boundary integral methods, regularized stokeslets, and various hybrid approaches. Cell migration models often need to incorporate biophysical processes such as chemotaxis, durotaxis, extracellular matrix remodeling, and mechanical feedback between cells and their environment. Coupling such complex processes with robust and efficient numerical solvers remains an active area of research with significant challenges.
This minisymposium aims to bring together researchers from across organizations and disciplines such as applied mathematics, computational mechanics, and biomedical engineering, to explore state-of-the-art numerical methods tailored to the modeling and simulation of cell migration. It will provide a platform to discuss recent developments in modeling approaches and numerical methods used to simulate cell migration, with various applications. In addition to development of new modeling and simulation methods, the session will also encourage the integration of experimental data into simulations via validation techniques and translation of computational results into biological insight.
We have had four confirmed speakers from Germany, Belgium, Canada, and the USA and two interested speakers. We are inviting more people from all over the world.
Keywords:
;Cell Migration; Modeling and Simulation, Biomechanics
Verified and validated computational models of hard tissues such as bones and teeth have been shown to be a valuable clinical tool in many applications such as: Prediction of risk of fracture in femurs and vertebrae due to osteoporosis [1], the need for prophylactic surgeries in femurs with metastatic or benign tumors, optimizing patient-specific implants, optimizing 3D printed implants and more.
This mini-symposium aims to present recent advances in the application of computational biomechanics of hard tissues with an emphasis on clinical practice, difficulties and opportunities in in-vivo validation of these methods. We aim at bringing together computational biomechanics scientists who share the same topics of interest for fruitful discussions and possible future collaborations.
Keywords:
"personalized medicine", femur, humerus, tibia, Bone
Computational biomechanics and robust numerical methods are powerful tools for supporting early disease detection and advancing modern treatment strategies. However, the complexity of living systems poses completely new challenges for mechanical models and numerical solution methods. To enable predictive simulations that are useful for clinical practice in the everyday clinical setting, it is essential to combine mechanics with biochemistry or electrophysiology via multiphysics modeling. These models can bridge the gap between metabolic processes at the subcellular level and macroscopic continuum mechanics, incorporate tissue responses to mechanical stimuli through coupling strategies, and intelligently integrate experimental data for model calibration and validation. An additional important challenge is to consider not only individual processes independently of each other, but also the interaction of different functional units in the context of an entire biological system.
This minisymposium focuses on novel approaches to address these challenges. We welcome highly interdisciplinary contributions that bring together expertise from different fields, such as mechanical modeling, numerical methods, data science, and clinical application. The goal of this minisymposium is to create valuable synergies between researchers working on different biological systems, potentially at different levels, to bring computational modeling one step closer to clinical practice. Contributions from experimental biomechanics are also welcome, since they generate data that is essential for the further development of the mechanical understanding as well as the parameterization and validation of computational simulation models.
REFERENCES
[1] Suditsch, M., Egli, F. S., Lambers, L. & Ricken, T. (2025), DOI: 10.1016/j.jocs.2025.102533
[2] Ahmadi Soufivand A., Lee SJ., JĂźngst T. & Budday S. (2025), DOI: 10.1088/2516-1091/addb19
[3] Ates, F. & RĂśhrle, O. (2024), DOI: 10.1002/gamm.202370012
Keywords:
Active Biological Systems, Continuum Biomechanics, Experimental Validation, Mechnobiology
Advances in computational modeling for oncology are revolutionizing our ability to forecast the outcomes of cancer therapies, potentially enabling their optimization and personalization. By capturing the complex multiscale dynamics of tumor growth, mass transport of therapeutic agents, microenvironment interactions, and treatment response, these models provide critical insights that bridge experimental and clinical studies. Computational models offer a powerful framework for investigating mechanisms of treatment resistance, selecting optimal combination regimens, designing more effective drug delivery systems, and tailoring interventions to individual patients through patient-specific simulations and digital twin technologies. By linking tumor biology with microenvironmental factors (i.e., vascularization, hypoxia, stromal interactions) and integrating principles of drug design and delivery, multiscale modeling supports the development of more precise and effective therapeutic strategies. Despite these advances, significant challenges remain in model fidelity, parameterization, and integration into real-world clinical decision-making. This underscores the need for close collaboration between computational scientists, experimental researchers, and clinicians.
This mini-symposium will highlight cutting-edge developments in computational approaches for cancer therapy, including:
⢠Physics-based modeling of tumor biophysics, drug delivery, and treatment effects across scales.
⢠Rational design of new molecules, drug delivery systems, and their synergistic interaction.
⢠Data-enhanced frameworks for therapy optimization and resistance prediction.
⢠Digital twins for clinical decision support and treatment personalization.
⢠Mathematical and computational challenges in model validation, parameter estimation, and clinical translation.
We welcome contributions that address these themes through novel modeling approaches, computational methods, or translational applications. The session aims to bring together researchers working at the intersection of mathematical modeling, computational science, and clinical oncology to advance patient-specific cancer care.
Keywords:
Digital twins in medicine , Drug delivery, Personalized cancer therapy, Treatment optimization, Computational oncology
The brain has an exceptionally high energy demand and can only function with proper blood supply. Vascular diseases are often linked to neurological problems, especially in ageing populations, which motivates further examination of their coupled interaction. Biomechanical testing and computational models [1] can aid this process by helping to better understand mechanics-related physiological processes, such as neurovascular coupling [2], nutrient transport, and pathological processes, such as arterial stiffening, critical stresses in aneurysms, angiogenesis in tumour growth, or vessel rupture in traumatic brain injury. However, modeling the brain vasculature is challenging for multiple reasons. First, it is difficult to determine the material properties of blood vessels and brain tissue, with the latter showing a complex poro-viscoelastic behavior and deforming under vascular pressure. Second, mapping the vasculature, including the microvasculature, can be challenging due to geometric singularities and specificities, such as the blood-brain barrier. Therefore, multidisciplinary approaches are needed to study the complex nature of the vascularized tissue. Multiphysics models enable the coupling of chemical, biological, and mechanical processes, while multi-scale approaches can help to study processes from the cell to organ scale or at longer time scales. In this regard, involving clinicians ensures that modeling efforts address real clinical needs. This minisymposium aims to give researchers working on brain vasculature the opportunity to present recent advances in the field and to connect with each other. We welcome all contributions ranging from in vivo, in vitro, and ex vivo experimental approaches to computational models and in silico studies. This includes approaches for the incorporation of in vivo data, extraction of patient-specific geometries, and the characterization of pressurised tissue under intracranial and blood pressure.
REFERENCES
[1] Belponer, C., Caiazzo, A., & Heltai, L. (2023). Reduced Lagrange multiplier approach for non-matching coupled problems in multiscale elasticity. arXiv preprint arXiv:2309.06797.
[2] Neher, C. M. et al. Perfusionâmechanics coupling of the hippocampus. Interface Focus 15, 20240051 (2025)
Keywords:
Blood pressure, Brain mechanics, Computational Biomechanics, Continuum Modelling, Intracranial pressure, Multiphysics, Neurodegenerative diseases, Neurovascular diseases, Vasculature
If a thrombus becomes lodged in a brain artery, it can lead to a stroke, the second leading cause of death worldwide. Swift and efficient removal of the thrombus is essential for optimal recovery. Thrombectomy, the mechanical extraction of the clot is the preferred treatment method. Several devices, including stent retrievers and aspiration devices, are clinically used to perform this procedure. There is significant room to improve both the procedure and the devices.
A thrombus consists of a fibrin network containing red blood cells and platelets. The mechanical properties of these thrombi are significantly influenced by the interactions among these three components. Depending on their composition, they are very soft and break down easily, while others are relatively stiff and can resist large deformations without failure. Insights into thrombus mechanics are essential for understanding how clots respond to mechanical forces during thrombectomy.
The intricate interplay between the fibrin network, flexible red blood cells, and contracting platelets need to be incorporated in microscale models, including the properties and interactions of each component. Through sophisticated multiscale approaches, we can develop microstructurally informed macroscopic models of the mechanical behaviour of thrombi6. The successful application of these models involves verification and validation through sensitivity analyses, and uncertainty quantification. Once validated, these models can be used in in silico platforms to quantify the interaction between the thrombus and a stent retriever to predict the outcome of a thrombectomy procedure.
It is obvious that advanced computational methods are key elements required to tackle the challenges described above. The proposal for this mini-symposium will focus on the computational aspects of:
⢠multiscale mechanical models for thrombi
⢠Failure mechanisms for thrombi
⢠Thrombus-device interaction
⢠Uncertainty quantification of thrombus models
⢠Development of surrogate models and digital twins of thrombi for clinical applications
Keywords:
Digital twins in medicine, Multiscale Modeling, uncertainty quantification (UQ)
Biological tissues throughout the body are dynamic and respond to mechanical stimuli
by adapting their anatomy and physiology. This process of growth and remodeling is essential for tissues to change and meet the body's shifting needs, such as a muscles growing stronger in response to exercise. However, growth and remodeling can also be pathological, and are central components of severe global health problems such as cancer and cardiovascular disease. Computational models are increasingly being used to understand the biomechanics and mechanobiology that govern the function and development of soft tissues. These models often incorporate growth laws that predict
how a tissue will grow and remodel in response to altered mechanical stimuli. Despite recent progress, many important questions remain unresolved. For example, what are the primary mechanical stimuli that drive growth and remodeling, how are these drivers sensed and processed at the cellular level, and how can computational models effectively bridge the vast time scales involved, from a single mechanical cycle to several months of tissue change?
In this minisymposium we aim to bring together scientists from various branches of soft tissue biomechanics and mechanobiology, to address fundamental challenges in computational modeling of growth and remodeling. Relevant topics include (but are not limited to):
⢠Biophysically based models of growth and remodeling processes on sub-cellular and tissue scale.
⢠Multiscale computational models of growth and remodeling, that address the fundamental challenge of disparate spatial and temporal scales in the process.
⢠Studies of growth and remodeling in specific progressive diseases.
⢠Agent-based models for cell proliferation and migration.
⢠Multiphysics and multiscale models for mechanics and mechanobiology.
⢠Numerical methods and algorithms for multiscale and multiphysics models
⢠Applications of biomechanics models in cardiovascular medical and surgical treatments
⢠Computational tools, specialized software, and databases for cardiovascular biomechanics and mechanobiology
Keywords:
biomechanics, mechanobiology, Computational Biomechanics, Growth and remodeling
Advances in data science, coupled with computational modelling, AI-based pathway identification and other areas of biomedical engineering, are enabling a new generation of cardiovascular models [1] that integrate patient-specific data with physics-based representations. Such data-driven approaches can leverage large volumes of multimodal information from medical imaging, wearable sensors, and clinical records to capture the complex, adaptive behaviour of the human cardiovascular system more accurately than traditional models alone.
Machine learning, statistical inference, and hybrid modelling frameworks now make it possible to identify disease signatures, forecast progression, and optimise treatment strategies. These techniques also support the creation of cardiovascular âdigital twinsâ that fuse continuous data streams with computational simulations for real-time monitoring, diagnosis, and personalised intervention.
This mini-symposium will bring together researchers, clinicians and industry professionals to share advances in data integration, patient-specific modelling, uncertainty quantification and case studies. Topics will include hybrid modelling, data assimilation, clinical decision support and adoption in healthcare. The session will showcase how data-driven cardiovascular modelling can enhance diagnosis, treatment planning and long-term management, advancing predictive and personalised medicine.
Keywords:
Biomedical Engineering, Cardiovascular system, data-driven modelling
Medical imaging plays a crucial role in the diagnosis and understanding of various pathological conditions, particularly in the cardiovascular domain. Increasingly, these imaging techniques are complemented by computational technologies grounded in mathematical modeling, enabling the extraction of clinically relevant insights that are not directly visible in raw images. In this minisymposium, we will discuss the latest developments in this interdisciplinary field at the intersection of imaging, mathematics, and computational science. International experts and researchers will focus on algorithmic issues and the use of both classical and newer techniques involving machine learning methods for image processing and for solving the underlying mathematical equations. We will explore both model-driven and data-driven strategies, with particular attention to hybrid methods that combine the strengths of both paradigms. The symposium will also address the clinical applicability of these approaches, along with efforts in validation and translation to real-world healthcare settings.
Keywords:
Cardiovascular Images, Computational Hemodynamics, Parallel Computing
The last 30 years have seen tremendous advances in creating patient-specific mathematical and computational models to aid in the diagnosis and treatment of human diseases. This has been driven by advances in computational mechanics as well as high performance computing, medical imaging and artificial intelligence. The sophistication of computational models and physiologic realism is remarkable. Furthermore, in the last decade there has been great progress in applying these models in pharmaceutical and medical device design as well as direct clinical application with software-enabled clinical services based on computational mechanics models. This minisymposium is focused on the development of mathematical and computational methods for biomedical digital twins with an emphasis on comprehensive models of human physiology and clinical applications.
Keywords:
Biomedical Science and Engineering, Computational Biomechanics, Digital twins in medicine
The SILICOFCM platform offers a innovative, multi-modular approach designed to optimize overall heart function and monitor the effectiveness of pharmacological treatments, aiming to decrease dependence on animal and human trials. The platform specifically targets hypertrophic (HCM) and dilated (DCM) cardiomyopathy through coupled macro- and micro-scale simulations utilizing finite element modeling of fluid-structure interaction (FSI) and molecular interactions within cardiac cells. This enables the simulation of left ventricular mechanics and the assessment of how different drugs influence electro-mechanical processes, including changes in calcium (Ca2+) handling and kinetic parameters. The overarching goal of the STRATIFYHF project is to develop and clinically validate an innovative AI-driven Decision Support System (DSS) to predict heart failure risk, support early diagnosis, and forecast disease progressionâoffering a paradigm shift in heart failure management across primary and secondary care. The DSS integrates patient-centered data from existing and emerging technologies, a digital patient library, and AI-based algorithms combined with computational modeling. Through workflows dedicated to improving heart performance and evaluating pharmacological effects, SILICOFCM and STRATIFYHF are paving new pathways to accelerate drug development and clinical testing.
Keywords:
AI, diagnostics, DSS, finite element, heart failure, therapy, cardiovascular disease
The integration of in-silico models into clinical research has advanced our understanding of tissue and fluid behaviour in the human body, enabling disease progression prediction and treatment evaluation [1,2]. This mini-symposium will focus on recent developments in modelling and analysis of forward and inverse problems in biomedicine, focusing on the linkage between novel techniques based on surrogate reduced-order modeling with applications in the context of cardiovascular and cerebral mechanics.
Forward problems address field prediction from known physiological parameters and boundary conditions, e.g., solving the NavierâStokes equations in vascular segments or simulating the poroelastic dynamics driven by CSFâbrain parenchyma interaction. These models pose computational challenges due to the complexity of geometries and of the multiphysics coupling, as well as to the need of patient-specific parameter tuning.
Inverse problems aim to reconstruct unknown parameters or full statesâsuch as pressure fields, material properties, or transport dynamicsâfrom medical data typically characterized by limited availability and quality (presence of noise and measurement artefacts), using optimization, data assimilation and scientific machine learning (SciML) strategies. Examples include physics-informed neural networks, Kalman filters, auto-encoders, and variational methods.
The symposium will highlight challenges and advances related to the application of these techniques in the context of vascular diseases (e.g., modeling of aneurysms and stenosis), CSF-related disorders (hydrocephalus, Alzheimerâs disease) and for soft tissues (e.g., liver, brain). Methodological contributions of interest include novel finite element schemes to address the coupled problems, data assimilation based on model order reduction, multiscale modelling, machine learning, surrogate and data-driven modelling and variational data assimilation.
[1] Lombardi, D. Reduced order modelling for direct and inverse problems in haemodynamics, Editor(s): Francisco Chinesta, ElĂas Cueto, Yohan Payan, Jacques Ohayon. In Biomechanics of Living Organs, Reduced Order Models for the Biomechanics of Living Organs. 2023.
[2] I. Fumagalli, S. Pagani, C. Vergara, L. Dede, A. D.A., M. Del Greco, A. Frontera, G. Luciani, G. Pontone, R. Scrofani, A. Quarteroni, The role of computational methods in cardiovascular medicine: a narrative review. Transl Pediatr. 13 (2024).
Keywords:
Biomechanics, blood flows, finite element method, scientific machine learning, soft tissue modelling
400
Geomechanics and Natural Materials
The global mean sea level is rising at an average rate of 4 millimeters per year, posing a significant threat to coastal communities and global ecosystems. The increase in ice discharge from glaciers worldwide, and in both the Greenland and Antarctic ice sheets, is now the dominant source of global mean sea level rise. However, its dynamic response to climate change remains a fundamental uncertainty in sea level rise projectionsâ .
This mini-symposium will feature presentations on studies that leverage computational mechanics, high-performance computing, and scientific machine learning along with remote sensing data to better predict mass loss from glaciers and ice sheets for global mean sea level rise projections. Relevant topics include: (i) computational strategies and software for tackling the complex, non-linear, multi-scale, multi-physics problems arising in ice sheet modeling, (ii) advanced analysis techniques, including approaches for model initialization/calibration and uncertainty quantification, (iii) performance portability of ice sheet models to advanced heterogeneous architectures, and (iv) approaches for improving the accuracy and/or efficiency of ice sheet models through the integration of AI/ML.
Keywords:
Cryosphere, High Performance Computing, ScientificMachine Learning, Computational Geomechanics
This minisymposium serves as a platform for scientists and engineers operating in the realm of computational wood mechanics, wood technology, and related biocomposite computational mechanics.
The papers submitted should reflect recent advancements and breakthroughs in the analytical and numerical exploration of the mechanical and physical properties of wood, biocomposites, and structures created from these materials. We also invite papers detailing developments in wood processing, innovative wood and biocomposites, and novel experimental investigations.
The topics that the minisymposium encompasses include:
⢠Theoretical, numerical, and experimental investigations related to computational mechanics of wood and biocomposites across different length scales.
⢠Microscale studies of wood and biocomposites, focusing on cell behavior, fibers, pulp, and paper.
⢠Macroscale investigations into solid wood, wood- and plant-based products, laminated components, and joints.
⢠Structural scale research, centering on building constructions and construction details.
Keywords:
Biocomposites, Timber, Wood, Wood-Based Products
The burgeoning field of underground engineering, including enhanced geothermal system (EGS), carbon sequestration, waste geological disposal and underground energy storage, requires a profound understanding of fracture mechanics in porous media under complex multiphysical conditions. These systems involve the interplay of thermal, hydraulic, mechanical, and chemical (THMC) processes, which significantly influence fracture initiation and propagation. Recent years have seen significant progress in the development of sophisticated constitutive models, numerical algorithms, and computational frameworks that capture the complex interactions between fracturing, fluid flow, heat transfer, and chemical reactions, such as the phase-field method, peridynamics and hybrid approaches. Additionally, machine learning and data-driven approaches are increasingly being integrated with traditional numerical methods to enhance predictive capabilities, reduce computational costs and address uncertainties in fracture modeling. However, challenges remain, including the accurate prediction of fracture nucleation under multi-axial stresses, the development of robust numerical solutions for highly heterogeneous models, the scalability of computational methods for large-scale simulations, and the validation of models against experimental and field data.
This Mini-Symposium aims to explore and showcase recent advancements in the theoretical frameworks and numerical modeling techniques for understanding fracture processes driven or influenced by coupled physical phenomena. Topics of interest include, but are not limited to: (1) Theoretical Advances: Variational fracture theory, multiscale constitutive model for fracture, analytical and semi-analytical solutions for fracture propagation, and impact of material heterogeneity and anisotropy on fracture behavior. (2) Numerical Methods: Multiphysical coupling techniques, algorithms for accelerating computations, cutting-edge computational techniques such as phase-field method, peridynamics, hybrid model and machine learning integrations. (3) Experimental and Field Validation: Laboratory experiments and field studies that provide critical insights into fracture processes and validate numerical models.
We invite contributions that advance the theoretical understanding, computational tools, and experimental validation of fracture mechanics in porous media under complex physical conditions.
Keywords:
Computational Geomechanics, Fracture Mechanics, Multiphysical Coupling, Porous Media
Geomaterialsâincluding soils, rocks, concrete, and snowâare porous, heterogeneous media that exhibit complex mechanical behavior and multiscale failure mechanisms under diverse multiphysics geological conditions. A deep and predictive understanding of deformation and failure processes in geomaterials is essential for addressing critical challenges in geophysics (e.g., fault slip, seismic rupture), geohazards (e.g., landslides, debris flows, and rock avalanches driven by climatic extremes), and geotechnical engineering (e.g., COâ storage, geothermal systems, and underground infrastructure).
With the advancement of computational science, numerical modeling has become a cornerstone in modern geomechanics, providing vital links between experimental insights and field-scale applications. This minisymposium aims to bring together researchers and practitioners to present and discuss recent developments in computational methods for geomaterials and geohazard modeling.
This minisymposium provides an open platform for interdisciplinary dialogue aimed at advancing computational tools and fundamental understanding in geomechanics, with the ultimate goal of supporting sustainable and resilient interactions with the Earthâs subsurface systems.
Keywords:
Fracture, Geohazards, Geophysics, Strain localization, Computational Geomechanics, Large deformations
Warming in the cryosphere is causing significant changes and leading to increased interest in the polar regions. In particular, Arctic sea ice loss has led to increased maritime activity in the region requiring improved sea ice forecasts. Permafrost thaw has resulted in infrastructure damage and coastal erosion. Melting and calving of ice sheets in Antarctica and Greenland lead to higher sea levels creating risks to coastal infrastructure. Accurate modeling of the thermo-mechanics and dynamics of these cryosphere systems is key to predicting and understanding future changes.
The focus of this minisymposium is on new computational methodologies for simulating cryosphere systems (land ice, sea ice, permafrost, etc.) and their interactions. The goal is to bring together researchers working on a broad range of cryosphere modeling topics to discuss recent advances and identify synergies.
Topics of interest include, but are not restricted to, the following:
⢠Novel numerical discretizations for ice and permafrost mechanics
⢠New constitutive models
⢠Mechanics-based formulations of ice fracture/calving
⢠Multiscale methods for coupling models with different spatial/temporal scales
⢠Efficient solvers and methods for improving computational performance
⢠Advanced analysis techniques including data assimilation and uncertainty quantification
⢠Data-driven surrogate modeling approaches
Keywords:
calving, fracture, land ice, modeling and simulation, permafrost, sea ice, Computational mechanics
This minisymposium aims to bring together researchers and practitioners to share recent advances in the computational modeling of geomaterials and geosystems, with a focus on both innovative methods and impactful applications. As global challenges related to infrastructure resilience, energy sustainability, and environmental protection intensify, computational geomechanics has become an essential tool for understanding and predicting the behavior of complex geomaterials such as soils, rocks, snow, and ice. These materials often exhibit highly nonlinear, coupled, and multiscale responses, necessitating the development of advanced predictive models and robust numerical approaches.
We welcome contributions on a wide range of topics, including but not limited to:
⢠Constitutive modeling and validation for geomaterials
⢠Coupled poromechanics and multiphysics processes (e.g., THMC modeling)
⢠Granular and micromechanical approaches
⢠Multiscale and multifidelity modeling techniques
⢠Large deformation and failure using meshfree, particle, or hybrid methods
⢠Fracture, damage, and localization phenomena
⢠Uncertainty quantification and probabilistic analysis
⢠Data-driven and machine learning methods in geomechanics
⢠Digital twins and computational frameworks for field applications
⢠Applications in geohazards, energy and environmental geotechnics, and geotechnical engineering
This minisymposium aims to foster discussion, promote collaboration, and explore future directions in computational geomechanics. Participation from both academia and industry is warmly encouraged.
Keywords:
Granular Mechanics, Porous Media, Geomechanics, Multiphysics
The increasing demand for sustainable and resilient geo-structures, as well as the repurposing and reuse of existing ones, presents significant challenges for geo-engineers and geo-scientists. They are tasked with designing complex projects while optimising available resources. Computational modelling and data-driven approaches have become fundamental tools in the design and back-analysis of geotechnical structures such as tunnels, deep basements, slopes, dams, retaining walls, and foundations. Recent advancements in numerical methods and data-driven techniques have revolutionised traditional approaches, enhancing the accuracy, efficiency, and reliability of ground engineering-related projects, and extending their application beyond experts to a broader range of engineers and geo-scientists. This minisymposia (MS) aims to collect advanced and stabilised/robust numerical and data-driven models dealing with geotechnical and ground engineering problems.
This MS will cover a range of cutting-edge subtopics, including but not limited to:
⢠Application of physics-enhanced machine learning in ground engineering problems
⢠Model- and data- driven techniques including model update, inverse problems, fusion of models and data and virtual control for ground engineering problems
⢠Advanced numerical methods: boundary-fitted, i.e., mesh-based and boundary-unfitted discretisation technique, i.e., CutFEM, including meshless methods; Material Point Method (MPM); Isogeometric analysis (IGA)
⢠Robust constitutive modelling techniques, e.g. sub-stepping, for tunnel and foundation engineering
⢠Advanced numerical methods and data-driven models for elastic and acoustic wave propagation problems
⢠Advanced sensing and monitoring technologies for geotechnical applications: data- and physics- driven interpretations and predictive analytics
REFERENCES
[1] NiniÄ, J. and Meschke, G., 2015. Model update and real-time steering of TBMs using simulation-based metamodels. Tunnelling and Underground Space Tech, 45, pp.138-152.
[2] Bui, H.G., NiniÄ, J., et al., 2024. Integrated BIM-based modeling & simulation of segmental tunnel lining by means of IGA. Finite Elements in Analysis and Design, 229.
[3] Liravi, H., Bui, H.G., Kaewunruen, S., Colaço, A., NiniÄ, J., 2025. Bayesian optimisation of underground railway tunnels using a surrogate model. Data-Centric Engineering 6, e32.
[4] Zhu, H. M., Huang, M. Q., and Zhang, Q. B., 2024. TunGPR: Enhancing data-driven maintenance for tunnel linings
Keywords:
Advanced Numerical Methods, Constitutive Modelling, data-driven modelling, Digital Twins for Underground Structures, Geotechnical engineering
Slope disasters, such as landslides, debris flows, slope failures, rockfalls and rock avalanches pose significant threats to human lives, infrastructure, and the environment. The damage from these disasters has become even more severe in recent years due to intensifying external forces like earthquakes and heavy rains, and mitigating the damage is a common concern for countries around the world.
In recent decades, numerical simulations have been recognized as powerful tools for predicting and assessing the risk of slope disasters. The integration of sophisticated computational techniques with high-resolution input data now allows for more accurate simulations and a deeper understanding of the mechanisms of these disasters.
This mini-symposium aims to bring together researchers and practitioners to share recent progress in advanced numerical modeling of slope disasters. Topics of interest include:
Discrete and continuum-based numerical modeling
Coupled hydrologic-geotechnical analysis
Large deformation analysis
Data-driven and machine-learning-assisted approaches
Probabilistic approach-based risk assessment
Model verification and validation (V&V) and uncertainty quantification (UQ)
Case studies on practical applications in slope disaster prevention and mitigation
Keywords:
modeling, simulation, slope sisasters
Materials such as soils, rocks, concrete, and ice are heterogeneous and complex. Advanced computational modeling techniques for capturing the mechanical and multiphysics behavior of these materials are essential in engineering applications related to energy, civil infrastructure, and the environment. This mini-symposium will provide a forum for presentations and discussions of the state-of-the-art and state-of-the-practice in computational geomechanics. Topics of interest include but are not limited to (1) development, implementation, and validation of advanced constitutive models in geomechanics; (2) particle-based numerical methods (DEM, MD, MPM, Peridynamics, etc.); (3) computational models and algorithms for multiscale and multiphysics problems; (4) numerical modeling of fracture, damage, and fragmentation processes; (5) local/nonlocal theories and models (mesh-based and mesh-free); (6) implementation and case studies of numerical methods in geo-energy, geo-hazards, and geo-environmental applications; (7) validation and verification of numerical models and algorithms in geomechanics; (8) machine learning and data-driven approaches in geomechanics; and (9) new challenges and opportunities (geomechanics in extraterrestrial exploration, nature-inspired geomechanics, metamaterials, etc.).
Keywords:
advanced numerical simulations, Advanced theories, Computational Geomechanics, Geohazards, Geophysics, Granular Mechanics, heterogeneous materials
500
Computational Applied Mathematics
Model reduction is concerned with reducing the degrees of freedom of high-dimensional state-space models of complex systems. Traditionally, model reduction is performed by approximating the dynamics on linear subspaces, building on concepts from (Petrov-) Galerkin projection, transfer function interpolation, and/or systems theory. Recent breakthroughs in scientific machine learning opened entirely new avenues for model reduction using, e.g., manifold approximations, autoencoders and online-adaptive bases, to address a longstanding approximation barrier when using linear subspaces.
This minisymposium will present a timely snapshot of the state of the art in nonlinear model reduction enabled by scientific machine learning. The themes of the minisymposium will revolve around the development of nonlinear model reduction methods with polynomial manifolds, sparse manifold constructions, data-driven nonlinear model reduction, structure-preserving nonlinear model reduction, adaptive model reduction with specific enablers for online efficiency and accelerators for integrating the the reduced-order models forward in time (such as hyper-reduction).
Applications for the developed methods may come from a wide range of computational mechanics, such as fluid and solid mechanics, plasma physics, electromagnetism, biological systems, and many more. These applications will have the shared complexities of high dimensionality paired with nonlinearity and a strong need for model reduction to enable many query applications.
Keywords:
high-dimensional systems, manifold approximations, nonlinear model reduction, scientific machine learning
The numerical simulation of poromechanics, the behavior of fluid-saturated porous media, has become of importance in several branches of natural sciences and technology for analyzing experimental data or designing theories and therapies based on mathematical concepts. In the case of fluid injection in porous media it is important for the assessment of the fluid storage capacity of the medium to accurately model and simulate the mechanical response, i.e., the deformation, of the porous media. Poroelastic waves, partial saturation, the incorporation of two- or multiphase flow, thermal processes and/or large deformations in general lead to nonlinear and even more complex mathematical models that require physically correct (structure-preserving) discretization and robust iterative solution methods. The range of physical parameters puts an additional facet of complexity on the of numerical methods.
Recently, a particular focus has been put on monolithic solvers as well as iterative coupling schemes for the coupled subsystems of fluid flow and mechanical deformation. In practice, input data often entail uncertainty and methods for its quantification are of great importance, too. There is also a need to design and analyze new models or further development in porous media with fractures, multi-compartmental, nonlinear or fully dynamic systems. Contrary to linear models, the accurate numerical solution of nonlinear poromechanics problems in general is more involved and its reliability much more difficult to guarantee. This can be due to nonlinearities of various kinds, e.g., in constitutive relations, boundary conditions, functionals in variational formulations of error control or optimization problems. Important questions to be addressed in this context concern the stability and convergence of approximations, adaptive numerical methods, a priori and a posteriori error estimates as well as robust iterative solution methods. The mini-symposium provides a forum for the presentation and discussion of progress in the above-mentioned fields.
Keywords:
Algebraic solver, Deformable porous media, Discretization schemes, Fluid flow, Splitting techniques
This mini-symposium aims to bring together new trends in the development of innovative numerical methods (finite elements, fictitious domain methods, moving interfaces, multi-scale methods, inverse problems, domain decomposition and HPC, etc.) for fluid flow simulation. Machine learning based techniques are also welcome. The focus of this mini-symposium also includes the challenges and contributions to the mathematical analysis and computational implementation of the methods, as well as their applications in the computational simulation of realistic fluid problems.
Keywords:
Fictitious Domain Methods, finite element, Moving Boundaries, Multiscale and Multilevel Methods, scientific machine learning, Stabilized methods
The European Service Network of Mathematics for Industry and Innovation (EU-MATHS-IN) links 17 national networks across Europe (https://eu-maths-in.eu/). It fosters high-impact collaborations by facilitating communication and information exchange between researchers and industrial partners.
This minisymposium showcases successful technology transfer projects achieved through EU-MATHS-IN. Researchers and industry representatives present real-life examples that demonstrate the power of mathematics to drive innovation.
In addition, we present the EU-MATHS-IN OpenDesk (https://opendesk.eu-maths-in.eu/), an innovative support platform for European companies, start-ups and public administrations. This service provides valuable resources and guidance on how to use mathematics for problem solving and technological progress.
We plan for 1 key note (40 min) and 4 regular talks (20 min).
Keywords:
European Projects, Success Stories, Industrial Mathematics
Alterations in the eye can often serve as early indicators for various vascular and systemic diseases, preceding their manifestation in the rest of the body. In the eye, hydrodynamic and biomechanical changes can be visualized and measured non-invasively in vivo. Mathematical, computational, and statistical modeling, used in synergy with clinical and experimental data, can thus be profitably used to identify possible driving mechanisms, isolate important biological factors, and test clinical hypotheses.
A broad spectrum of mathematical approaches is employed, including data-driven modeling, uncertainty quantification, fluid-structure interaction methods, free-surface modeling, and compartmental models. In addition, recent advances in hybrid modeling strategies, combining mechanistic and data-driven approaches, open new avenues for the development of patient-specific simulations and real-time predictive tools. Digital twin frameworks are particularly promising in this context, enabling adaptive and interpretable models that integrate multiscale data, from tissue-level mechanics to drug transport and metabolism.
This minisymposium aims to investigate different regions and functions of the eye, which include, proceeding from its exterior to the interior: tear film fluid mechanics, biomechanical corneal properties and contact lens interactions, and the biochemical, electrical, thermal, and vascular components of drug delivery.
We encourage contributions that explore coupling between scales and physics, surrogate modeling to reduce computational costs, and data assimilation techniques for integrating sparse or noisy measurements. Of particular interest are works that combine theoretical and numerical approaches with experimental validation, and that provide insights into personalized modeling of ocular systems. By focusing on both biological relevance and methodological innovation, this minisymposium seeks to foster interdisciplinary exchange among mathematicians, engineers, biophysicists, and clinicians working toward the modeling and simulation of the eye as a complex, dynamic, and diagnostically informative system.
Keywords:
Data-driven Models, Mechanistic
Models, Theoretical and Numerical Approaches, Experimental Validation, Eye
Linear operators like the Koopman operator have emerged as a powerful means to analyzing nonlinear dynamical systems. Among its key features is the ability to characterize complex dynamics via eigenfunctions, which can offer insightful decompositions even for highly nonlinear systems. However, the accurate and reliable computation of these eigenfunctions remains a significant challenge, particularly in the context of engineering applications where stability, convergence, and interpretability are critical.
This minisymposium will focus on recent developments in the verified and robust computation of operator eigenfunctions, with special emphasis on emerging techniques. We seek to bring together researchers from the fields of computational mathematics, dynamical systems, and engineering to explore how these computational tools can be made both mathematically rigorous and practically useful.
The scope of the minisymposium includes, but is not limited to:
- theoretical foundations of (Koopman) operator theory and spectral analysis;
- numerical methods for computing operator spectra and eigenfunctions, with an emphasis on accuracy, stability, and verification;
- integration of operator-based methods with finite element or other simulation frameworks;
- applications to engineering systems, e.g. use of operator methods in control and optimization problems in engineering.
Keywords:
Koopman operator, transfer operator
Machine learning has emerged as a powerful tool in the field of computational science, with plenty of applications in various domains such as fluid dynamics, materials science, and control systems. This minisymposium will explore recent trends in the realm of scientific machine learning that leverage data-driven approaches to enhance the solution of differential equations and to optimize control strategies. We aim to bring together researchers who are pushing the boundaries of how machine learning can be integrated with traditional numerical methods to solve complex scientific problems.
In the past few years, there has been a surge of interest in developing machine learning techniques that build upon or can be seamlessly integrated with classical numerical methods for solving partial differential equations or optimal control problems. These approaches often involve the use of neural networks or kernel methods to approximate solutions, learn operators, or even discover governing equations from data. The potential of these methods lies in their ability to handle high-dimensional problems, such as Hamilton-Jacobi-Bellman equations, and to adaptively refine solutions based on available data. Moreover, the integration of machine learning with numerical solvers can lead to significant improvements in computational efficiency and accuracy. In order to guarantee the reliability of these methods, it is of interest to investigate the theoretical foundations of machine learning techniques, including convergence properties and error estimates. This theoretical underpinning is crucial for establishing trust in machine learning approaches when applied to scientific computing tasks.
This minisymposium will feature talks focusing on different aspects of the application of machine learning to differential equations and control problems. On the one hand, theoretical results will be presented that establish the convergence and stability of machine learning methods when applied to the numerical solution of scientific computing problems. On the other hand, practical results will be showcased, demonstrating how these methods can be effectively applied to real-world problems in various scientific domains. We will also discuss the challenges and limitations of current approaches, as well as future directions for research in this exciting field.
Keywords:
Control Problems, Differential Equations, Numerical
Methods, Scientific Machine Learning
Advances in high-fidelity numerical modeling, combined with the increasing availability of observational and experimental data, are revolutionizing the integration of data and simulation in scientific and engineering analyses. However, the computational cost of high-fidelity models can be prohibitive for outer-loop tasks such as inference, uncertainty quantification, design, and control, which require repeated model evaluations. To address this challenge, recent research has focused on developing data-driven approximation methods that replace expensive models in the outer loop, enabling efficient and scalable analysis with limited evaluations of the high-fidelity model.
This session will explore innovative approaches to leveraging data-driven surrogate models to improve accuracy and reduce computational costs in simulation-aided analysis. Discussions will focus on three key categories: finite-dimensional function approximation, operator learning, and generative modeling. As advances in these areas continue to evolve rapidly, it is essential to explore the relative strengths and trade-offs of these approaches. Presentations will highlight strategies for accelerating surrogate and generative model training, theoretical error estimation, and uncertainty quantification, alongside practical applications in areas such as risk analysis of multi-scale, multi-physics systems and predictive digital twins.
Special attention will be given to experimental design for training data, the costs of surrogate construction and interrogation, and hybrid approaches that integrate high-fidelity models with surrogate models to balance the trustworthiness of high-fidelity predictions with the efficiency of low-fidelity models, enabling more outer-loop iterations. This session will showcase advances in this rapidly evolving field and their successful applications in areas ranging from fusion energy systems to materials discovery.
Keywords:
Approximation Techniques, Machine Learning, Surrogate Modeling, Uncertainty Quantification
One current challenge for continuum mechanics is the definition of a useful mathematical
setting for the prediction of motions encompassing various materials, from solids to fluids.
The quest can be motivated by many configurations and applications (fluid-structure
coupling, multi-phase flows, phase changes, etc.). One promising direction builds on a
symmetric-hyperbolic system of unified equations for the Eulerian description of various
compressible materials, from solids to fluids, that makes use of internal variables and an
updated Lagrangian description to cover various rheologies. Then the challenge precises: how
to reliably simulate multi-material flows from such a system, up to the propagation of shock
waves? How far can such a unifying system capture the physics of multi-phase mass
transport?
The symposium will try to give an overview of the questions and answers currently discussed
in the above-mentioned framework: which internal variables should be chosen? How can we
build updated Lagrangian descriptions covering a wide range of rheologies? Should the
mathematical definition start from discrete Lagrangian or Eulerian approaches? How do the
above questions and potential solutions deal with discrete models for numerical simulation?
How should we address singularities (shock waves, fracturing), especially when nonconservative systems are at stake (e.g. multi-phase flow models, compressible turbulence
models)? etc.
In addition to the organizers, speakers are expected from a well-identified, distinguished list
of scientists including S. Gavrilyuk (University of Aix-Marseilles, agreed), A. Gil and C.-H.
Lee (University of Glasgow), J. Bonet (CIMNE), M. Dumbser and I. Peshkov (University of
Trento), etc.
Keywords:
Advanced Materials, Advanced theories
Rigorous estimation of effective elastic properties of heterogeneous random materials requires a âstatistical approachâ [1], which, due to its complexity, is invariably restricted to the generation of âboundsâ, by consideration of limited statistical information [2]. Alternatively, an âintuitive approachâ [1] based on the notion of an RVE subjected to a homogeneous boundary condition [3] may be employed. In this context, use of âEshelby-likeâ solutions [4] enables, by means of intuitive procedures such as the (classical) self-consistent scheme (SCS) or the Mori-Tanaka scheme (MTS), actual estimation (rather than bounding) of effective elastic properties of polycrystals or composites, respectively. However, only exceptionally [1], do we know how reliable such SCS- and MTS-based estimates are, with for instance, available bounds for assessing SCS-based estimates for porous polycrystals, being still only up to second order and restricted to the nonporous polycrystal case [5]. This symposium aims at bringing together scientists and engineers involved in cutting-edge theoretical, computational, and experimental research on rigorous statistical estimation of effective elastic properties of heterogeneous materials. Submitted contributions should address recent theoretical advances in this area (including experimental determination of second, third, or higher order correlation functions) and comparison of corresponding estimates with SCS- and/ or MTS-based estimates.
REFERENCES
[1] J.E. Gubernatis and J.A. Krumhansl, Macroscopic engineering properties of polycrystalline materials: Elastic properties, Journal of Applied Physics 46 (1975) 1875-1883.
[2] M.J. Beran, T.A. Mason, B.L. Adams and T. Olsen, Bounding elastic constants of an orthotropic polycrystal using measurements of the microstructure, Journal of the Mechanics and Physics of Solids 44 (1996) 1543-1563.
[3] R. Hill, The essential structure of constitutive laws for metal composites and polycrystals, Journal of the Mechanics and Physics of Solids 15 (1967) 79-95.
[4] J.D. Eshelby, The determination of the elastic field of an ellipsoidal inclusion, and related problems, Proceedings of the Royal Society of London A 241 (1957) 376-396.
[5] M. Lobos FernĂĄndez and T. BĂśhlke, Representation of HashinâShtrikman bounds in terms of texture coefficients for arbitrarily anisotropic polycrystalline materials, Journal of Elasticity 134 (2019) 1-38.
Keywords:
Effective Elastic Properties, heterogeneous materials
600
Data Science, Machine Learning and Artificial Intelligence
This minisymposium provides a platform to discuss current developments in data-driven methods and scientific machine learning, which are transforming material modeling, computational mechanics, and solid mechanics more broadly. It continues the effort of bringing together researchers working on advancing these approaches to model and simulate complex mechanical problems. In that regard, several promising directions have emerged, ranging from directly exploiting data for computational mechanics without constitutive laws, applying deep learning including manifold learning and autoencoders for reduced-order modeling of nonlinear high-dimensional mechanics problems, integrating data-driven machine learning techniques with physics- and thermodynamics-based and models for various forward and inverse problems, to leveraging experimental measurements with artificial intelligence techniques towards fundamental solid mechanics models. Contributions are invited on a wide range of topics, including but not limited to: the development of machine learning-based surrogate and large-scale foundation models, physics-informed machine learning models for linear and nonlinear solid mechanics, model-free data-driven computational mechanics, data-assisted modeling of heterogeneous materials, data-driven discovery of constitutive laws and governing equations, the development of interpretable and explainable machine learning models, reduced-order real-time simulation of solids, inverse problems with machine learning, neural-enriched computational methods, as well as probabilistic and uncertainty quantification techniques. We finally invite researchers to contribute their findings in application-focused studies in solid mechanics, biomechanics, geomechanics and related disciplines. The minisymposium aims to foster an interdisciplinary discussion on the future of data-driven modeling and its potential to complement and transform traditional approaches in mechanics.
Keywords:
AI for Science, Computational Mechanics
In recent years, the growing availability of data from computational sciences has highlighted
its potential to provide valuable insights and enhance predictive accuracy. In the field of
aerodynamics, extensive research and optimization efforts generate significant amounts of
useful data, presenting opportunities to advance engineering through data-driven and datafusion models. However, the adoption of these models remains in its early stages, with best practices still emerging.
Machine learning, including techniques such as neural networks, offers powerful tools for tasks like clustering, dimensionality reduction, classification, and regression. Despite this potential, preparing and processing aerodynamic and geometric data remains challenging. These processes are often complex and tailored to specific objectives, leading to diverse
interpretations and implementations of data-driven approaches. Leveraging machine learning techniques, widely used in artificial intelligence and data mining, could significantly lower computational costs for aerodynamic analysis, optimization and uncertainty quantification [1, 2]. These innovative methods pave the way for more efficient and precise solutions in aerodynamic design, although challenges in data preparation and model refinement persist.
This minisymposium seeks to highlight novel strategies and recent advancements in applying machine learning and data-driven approaches to aerodynamic analysis and uncertainty quantification. It emphasizes practical challenges and explores new opportunities offered by scientific machine learning, the integration of advanced machine learning techniques with scientific computing, for the development of more efficient and effective methodologies for analysis and design.
Keywords:
aerodynamic analysis, machine learning, data-driven models
Nowadays, since more and more powerful heterogeneous computers are continuously emerging, scientists and engineers have been facing unprecedented challenges in adapting their workflows to the challenges posed by digital twins and scientific machine learning. This mini-symposium provides a forum for attendees to exchange information, share best practices, and keep current on the rapidly evolving information technologies impacting the convergence of simulation tools, digital twins, and scientific machine learning. The Mini-Symposium topics cover (but are not limited to):
Computational environments for advanced scientific machine learning and engineering computations
Digital prototyping techniques
Enabling software technologies
Data science in computational mechanics applications
Software libraries and applications for digital twins, model reduction, and machine learning
Supporting tools in performance evaluation, visualization, verification, and validation
Scientific workflows, theoretical frameworks, methodology, and algorithms
Keywords:
Model Order Reduction, Digital Twins, scientific machine learning
Kernel-based approximation methods are powerful and versatile tools in numerical analysis and scientific computing. They are widely used for function reconstruction from scattered data, with applications ranging from surface reconstruction and image processing to geostatistics, and machine learning. In the context of partial differential equations, kernel methods provide meshfree alternatives to traditional discretizations, enabling flexible and high-order accurate solvers that are particularly attractive in scattered or evolving domains.
Kernel methods typically rely on positive definite kernels to interpolate or approximate a target function based on a finite set of data sites and corresponding values. One of the key advantages of kernel methods is their flexibility in handling irregular data and complex geometries. However, their practical application poses significant computational challenges. The global support of commonly used kernels results in dense and often ill-conditioned linear systems of equations, which can hinder scalability and numerical stability. Furthermore, kernel methods suffer from the curse of dimensionality when applied to high-dimensional problems, with increasing computational cost and data requirements.
The aforementioned challenges have motivated the development of a wide range of techniques to improve the efficiency and robustness of kernel methods. These include localization and sparsification strategies, compactly supported and multiscale kernels, hierarchical and multilevel approaches, domain decomposition techniques, and hybrid methods that combine kernel approximation with frameworks such as sparse grids or reduced bases.
This minisymposium brings together researchers working on both the theoretical and computational aspects of kernel approximation. Topics will include stability and convergence analysis, structure-exploiting numerical algorithms, scalable solvers, and recent advances in adaptive and data-driven kernel methods. Applications to partial differential equations, inverse problems, and learning tasks will also be featured.
Keywords:
machine learning, reproducing kernel Hilbert spaces, high-dimensional approximation, Scattered data approximation
In recent literature, machine learning (ML) algorithms have made tremendous advances in accelerating computer simulations across various applications. Scientific ML (SciML) is a branch of ML that devises algorithms that are customized for scientific applications, for e.g., by imposing constraints and preserving symmetries. These are designed to bridge the gap between rigorous scientific computing, built on first-principles-based numerical modeling of various governing laws, and purely data-driven methods. In this minisymposium, we will showcase recent developments in an emergent but important application of SciML â where algorithms are designed not only to accelerate scientific computing, but also to discover some previously unknown underlying property of physical systems such as an invariant measure, a governing law, or a reward that engenders the observed behavior. Specifically, we wish to discuss how SciML algorithms can be devised to mature beyond simply reproducing observed data and simple extrapolation tasks.
Keywords:
data-driven scientific discovery, explainable AI, scientific machine learning
In recent years, rapid advances in computational science, data-driven techniques, and machine learning have profoundly reshaped the landscape of model order reduction (MOR), model discovery, and surrogate modeling. This invited session brings together cutting-edge research at the intersection of physics-based modeling and data-driven methodologies, highlighting innovative strategies for constructing efficient, reliable, and interpretable reduced-order and surrogate models. Topics include projection-based MOR for nonlinear and time-dependent problems, data-driven and data assimilation methods leveraging sparse sensing and neural networks, and novel approaches to system identification and model discovery from limited or noisy data. Applications span a wide range of domains, including fluid dynamics, structural mechanics, multi-body dynamics and multi-physics systems. Emphasis is placed on the integration of physical laws, the handling of high-dimensional parameter spaces, and the development of robust methods for uncertainty quantification. The session aims to foster dialogue between theoretical development and practical implementation, and bridging traditional modeling approaches with modern data-centric techniques, paving the way for next-generation simulation tools.
Keywords:
data-driven methods, model discovery, operator inference , Model order reduction
In the design of mechanical systems, the role of an engineer is to find the best solution by considering problem requirements, security and cost of applications. To ensure the best balance between these necessaries, the method called optimization is employed. In the process of time, world resources are decreasing and the importance of time is increasing. For that reason, optimization is a popular and improvable research area.
This mini-symposium aims to bring together all leading academicians working on the optimization of mechanical systems. The symposium is a multidisciplinary scientific meeting in order to discuss new original approaches and original applications of artificial intelligence methods. Thus, this mini-symposium will be a valuable source of information exchange between several disciplines related to mechanics. In addition to optimization studies, machine learning methods, which are used for the prediction of mechanic problems, are also included in the scope of the mini-symposium.
Keywords:
Artificial intelligence, Computational Mechanics, Structural Dynamics, Structural Mechanics, Optimization, System Dynamics
MS087 â Data-Driven Computational Mechanics and AI for Advanced Materials and Multiphysics Systems
Recent advances in machine learning (ML), artificial intelligence (AI), and data science are revolutionizing computational mechanics and materials modeling. This mini-symposium aims to provide a platform for researchers working at the intersection of AI and computational science, targeting applications in nano- and quantum materials, fluid dynamics, and complex thermophysical systems. Emphasis is placed on first-principles modeling augmented by data-driven approaches, as well as on emerging paradigms such as thermal reservoir computing, AI-accelerated multiscale modeling, and high-performance computing (HPC) for simulation-based design. Contributions involving novel algorithms, hybrid physics-AI models, and application-driven case studies are strongly encouraged.
Topics of Interest Include (but are not limited to):
ďŹ Machine learning for computational fluid dynamics (CFD)
ďŹ Thermal reservoir computing and physical neural networks
ďŹ AI-accelerated simulation of nano- and quantum materials
ďŹ High-throughput first-principles and data-driven screening
ďŹ Hybrid ML/physics-based models for multiphysics coupling
ďŹ High-performance computing with AI for large-scale simulations
ďŹ Graph neural networks and physics-informed neural networks (PINNs)
ďŹ Generative models for material structure-property design
Keywords:
Fluid Dynamics, High-Performance Computing, Thermal Reservoir Computing., Nano Materials
This mini-symposium brings together researchers from mechanics, applied mathematics, and related engineering disciplines to explore the integration of machine learning into computational mechanics. The focus is on leveraging data-driven methods to enhance modeling, simulation, and design across a wide range of scales and physical phenomena.
Particular emphasis is placed on physics-based machine learning approaches that incorporate physical principles and constraints to improve consistency, interpretability, and generalization.
We welcome theoretical developments, algorithmic innovations, and application-oriented studies that utilize machine learning to advance understanding and computation in mechanics.
The goal is to foster interdisciplinary exchange and highlight cutting-edge advances that push the boundaries of computational mechanics through intelligent, physically grounded learning strategies.
Keywords:
Artificial Intelligence, Computational Mechanics, machine learning, Multi-scale modeling, Multiphysics problems, Physical principles, Physics-Based Data-Driven Modeling
Recent advances in machine learning (ML) are transforming the field of computational methods in aerospace engineering. This mini-symposium will bring together researchers and practitioners who are at the forefront of ML methodology and its application to aerospace problems. The symposium will cover theoretical innovations and practical deployments.
We invite contributions that leverage machine learning across a broad spectrum of aerospace applications. Topics may include, but are not limited to:
- Physics-informed machine learning approaches that integrate domain knowledge and physical laws into learning frameworks to enable improved generalization, data efficiency, and robustness in complex aerospace systems.
- Anomaly detection and predictive maintenance strategies, particularly in mission-critical systems such as aircraft engines, spacecraft subsystems, and satellite constellations, where early fault detection and interpretability are crucial.
- Aerodynamic modeling and flow control, where surrogate modeling, reduced-order modeling, or reinforcement learning can speed up design and optimization processes.
- Earth observation and satellite-based sensing, including the use of ML for large-scale remote sensing data interpretation, change detection, or predictive analytics in climate and environmental monitoring.
- Space transport and autonomous navigation, where machine learning plays a role in trajectory planning, control under uncertainty, or onboard decision-making.
While these examples highlight prominent use cases, the mini-symposium encourages submissions on all aspects of machine learning for aerospace applications.
Keywords:
aerospace application, anomaly detection, Earth observation, physics-informed machine learning, scientific machine learning, space transport
The objective of this work is to illuminate the emerging research areas, technological advancements, and future directions in the fields of nonlinear dynamical systems and control. This proposal encompasses a range of disciplines, including mathematical modeling, stability analysis, bifurcation theory, and chaos theory, all of which are pertinent to the study of nonlinear dynamical systems. The text explores the dynamics of complex systems and their behavior in response to nonlinear interactions. Additionally, it delves into control theory and optimization techniques applied to nonlinear dynamical systems. The text encompasses subjects such as adaptive control, optimal control, robust control, and hybrid systems. Conversely, a compelling aspect of machine learning pertains to the data-driven identification of the governing dynamical equations. Conventionally, the advancement of nonlinear dynamics has been characterized by an approach that involves the invocation of fundamental principles and intuitions in the formulation of theoretical explanations for observed phenomena. In the contemporary era of big data, a fundamental challenge lies in the reconstruction of the underlying dynamic system through the analysis of existing data. This challenge reveals a significant gap in knowledge, particularly concerning artificial intelligence systems that accurately replicate physical systems. A further challenge pertains to the hybridization of contemporary controllers, predicated on control laws, with machine learning training algorithms. Each technique operates within its own discrete domain of control. To illustrate, aircraft control based on modern techniques is predominantly valid and safe. Nevertheless, under critical conditions, a novel data-driven controller can assume control of the aircraft based on millions of simulations of situations that are critical to classic controllers. In summary, the present work is intended for a broad audience, including researchers, academics, professionals, engineers, practitioners, students, and educators. The text addresses a range of subjects, including nonlinear dynamics, control systems, and machine learning. Consequently, it provides valuable insights, cutting-edge research, and practical applications to readers seeking to explore and advance their knowledge in these exciting fields
Keywords:
Machine Learning, Control, Mathematical Model, Nonlinear Systems
Recent breakthroughs in machine learning (ML) and artificial intelligence (AI) have sparked a paradigm shift across scientific disciplines, pushing the boundaries of computational science and redefining how physical systems are modeled and understood. No longer confined to traditional software development based on explicit task-specific coding, research is increasingly embracing optimization-driven, autonomous methodologies that demand scalability, interpretability, and memory efficiency. These requirements are critical for tackling large-scale, high-dimensional, and multiphysics problems found in modern science and engineering. In this evolving landscape, the computational mechanics communityâlong known for its rigorous numerical methods and physically grounded modelsâoffers a rich source of inspiration. Emerging ML architectures now draw from these principles to develop efficient, physics-aware algorithms capable of solving both forward simulations and complex inverse design problems. These hybrid approaches not only enable reduced-order modeling of intricate phenomena but also leverage the power of differentiable programming and modern AI hardware for optimization and control tasks.
This mini-symposium invites original contributions that explore this exciting fusion of data science, ML, and computational mechanics. Topics of interest span scalable and interpretable neural architectures, intelligent modeling and control of multiscale and multiphysics systems, computer vision, inverse design strategies, and differentiable solvers that tightly couple physics and learning. We aim to provide a vibrant forum for researchers, spanning academia and industry, to share ideas, showcase innovations, and shape the future of mechanistically inspired AI. The symposium will feature keynote lectures from distinguished thought leaders and offer a platform for early-career investigators to gain visibility and mentorship. With growing momentum in fields like advanced manufacturing, robotics, and digital twin technologies, this gathering will serve as a nexus for collaboration among scientists, engineers, and technologists working at the frontier of intelligent scientific computing.
Keywords:
AI for Science, Computational Materials Science, Computational Mechanics, Inverse Analysis, Manufacturing Process modeling
Machine learning (ML) is transforming the way computational models are developed and deployed in aeronautical engineering, enabling quicker and more flexible workflows. However, challenges remain in terms of generalization, accuracy, and model validation, especially for industrial adoption. This minisymposium aims to gather recent developments in the use of ML models in aerospace applications. Topics of interest include:
⢠Surrogate models for fast predictions, e.g., for surface aerodynamic quantities;
⢠Data-driven Physics-Informed Neural Networks (DD-PINNs) for solving fluid dynamics problems with embedded physical constraints;
⢠Deep Reinforcement Learning (DRL) for efficient aerodynamic and structural shape optimization [1];
⢠Validation and verification tools and best practices to assess ML models beyond traditional error metrics [2]
The symposium welcomes contributions showcasing methodological advances, benchmark studies, and industrial applications.
REFERENCES
[1] Ramos, David, et al. "Aerodynamic and structural airfoil shape optimisation via Transfer Learning-enhanced Deep Reinforcement Learning." arXiv preprint arXiv:2505.02634(2025).
[2] Lacasa, Lucas, et al. "Towards certification: A complete statistical validation pipeline for supervised learning in industry." Expert Systems with Applications 277 (2025): 127169.
Keywords:
aerodynamic optimization, PINNS, surrogate models, verification and validation, aerospace application, Machine learning
The rapid advancements in machine learning (ML) and artificial intelligence (AI) have significantly expanded the possibilities in computational mechanics. One of the areas most profoundly influenced by these technologies is the development of constitutive models for complex materials. ML and AI have revolutionized traditional approaches in this domain, offering innovative solutions to several challenging problems, including: Homogenizing the behavior of multiscale materials with intricate microstructures; discovering closed-form material models from large, diverse model libraries; solving inverse problems to identify heterogeneous material parameters, either by replacing the forward solver with ML surrogates or directly learning the inverse map from imaging data.
These breakthroughs have been applied across a wide spectrum of materials, ranging from biological tissues (heart, skin, arteries, brain, etc.) to metals, elastomers, and soils. Beyond single-material modeling, recent Bayesian ML tools have been employed to account for the variability in material properties. This is particularly important for materials with inherent microstructural uncertainty or, in biological contexts, inter-individual variability in soft tissues.
A key area of focus in ML-driven constitutive modeling is the integration of physics-based constraints with data-driven methods. While initial approaches have employed loss functions to enforce these constraints, more recent innovations have incorporated architectural features that ensure physics-consistent predictions across all parameter regimes. Similarly, efforts are underway to merge ML models with microstructure-informed modeling techniques, enhancing both interpretability and predictive power.
ML and AI have demonstrated remarkable versatility in capturing a wide range of material behaviors, from simple linear elasticity to more complex phenomena like hyperelasticity, viscoelasticity, plasticity, and large-deformation, non-equilibrium processes. Furthermore, to bridge the gap between theory and application, these data-driven constitutive models need to be integrated into numerical solvers, such as finite element methods, to facilitate practical use in large-scale simulations.
This symposium invites contributions in the area of machine learning-based constitutive modeling for all types of materials and behaviors (e.g., hyperelasticity, viscoelasticity, plasticity). We particularly encourage submissions that focus on probabili
Keywords:
Constitutive Modeling, Data Driven Methods, Multiscale Materials
Over the past decade, neural networks have been a revolution in the computing landscape. However, the energy demands of running deep learning models have become unsustainable on the current digital hardware. Therefore, as digital computing approaches its energy efficiency limits, especially for AI workloads, neuromorphic and analog computing have re-emerged as promising alternatives. Neuromorphic systems mimic how biological brains process information by leveraging parallelism, asynchronous signal propagation, and unified memory-computation units to optimize energy consumption. Despite neuromorphic chips can be used to speed up existing AI models, it is essential to design to develop new models and algorithms that can embrace and exploit these characteristics. On the other hand, analog computing leverages the continuous dynamics of physical systems, such as electrical currents or material properties, to perform computation at significantly lower power and latency than traditional digital systems. This makes analog approaches especially compelling for edge AI applications and energy-constrained settings.
Despite their promises, both neuromorphic and analog computing raise key research questions, such as designing algorithms that are robust to the inherent noise of analog substrates, understanding the levels of precision and reliability for different AI tasks, and co-designing models and circuits to balance benefits with scalability and programmability.
This mini-symposium explores brain-inspired and analog computing paradigms that promise transformative advances in energy-efficient AI. We discuss both theoretical foundations and practical applications of neuromorphic and analog computing, highlighting key challenges, recent breakthroughs, and future directions. The event aims to bring together machine learning researchers and hardware engineers to to promote discussion on the design of new generation AI systems for non-traditional and energy-efficient hardware platforms.
Keywords:
Analog computing, Neuromorphic computing, sustainable AI
Constitutive modeling enables the mathematical description of the behavior of different materials such as metals, polymers, composites, soft biological tissue, or active materials. Conventional modeling approaches face challenges regarding highly nonlinear, inelastic, or multiphysical behavior. This can be traced back to a lack of flexibility of conventional constitutive models. To address this, in recent years, the formulation of material models using highly flexible machine learning (ML) methods such as neural networks, Gaussian processes and symbolic regression has gained momentum. It is widely agreed that ML-based constitutive models should be formulated to fulfill mechanical conditions such as thermodynamic consistency and objectivity, which can be coined as physics-augmented, physics-enhanced, or physics-constrained ML modeling. By that, the flexibility of ML-methods is combined with a sound mechanical basis.
This minisymposium seeks to gather researchers working at the intersection of mechanics and machine learning to address current challenges and explore emerging trends. Topics of interest include, but are not limited to:
⢠Constitutive modeling based on ML-methods such as neural networks and Gaussian processes
⢠Modeling of material behavior including: (i) energy conservation and dissipation (e.g., viscoelasticity, plasticity, damage, fracture, âŚ), (ii) multiphysics (e.g., thermo-, magneto-, or electromechanics), and (iii) parametric dependencies
⢠Fulfillment of physical conditions, interpretability, sparsity, and uncertainty quantification
⢠Calibration of constitutive models to (full-field) experimental data
⢠Efficient and structure-preserving implementation in numerical schemes such as multiscale simulation and topology optimization
Keywords:
Constitutive Modeling, Inelasticity, Multiphysics, Multiscale Modeling, Parameter Identification, Physics-Augmented Machine Learning
The need for describing, predicting, and controlling the dynamics of complex systems has led to a multitude of numerical methods across several disciplines over the last decade, blending physics-based and data-driven techniques to different extents. Reduced order modeling and dimensionality reduction paradigms through, e.g., proper orthogonal decomposition, dynamic mode decomposition, as well as autoencoders and transformers, provide well-established techniques to discover low-rank spatio-temporal patterns, embed the dynamics in a reduced subspace, ultimately requiring latent dynamics modeling for reliable forecasting.
Since the introduction of the sparse identification of nonlinear dynamics (SINDy) technique ten years ago, several extensions provide nowadays a wide set of sparsity-preserving techniques, and have ultimately made data-driven modeling and discovery an extremely active research area, nowadays integrating deep learning for uncovering effective coordinates, Bayesian and kernel methods for uncertainty quantification, multi-fidelity methods for data fusion, neural networks for time series description, deep reinforcement learning for controls, to name a few.
Key to all these strategies is the suitable modeling of the latent dynamics - the aspect this minisymposium will focus on, covering the theoretical analysis, computational techniques, and practical use of data-driven methods for the model reduction and discovery of dynamical systems, all towards efficient and accurate predictions in applied sciences and engineering.
Keywords:
dimensionality reduction, Dynamical Systems, latent dynamics, reduced order modeling, sparsity-preserving techniques
Artificial Intelligence (AI) has emerged as a transformative tool for tackling complex engineering and scientific challenges, significantly impacting computational modeling, optimization, uncertainty quantification (UQ), and inverse problem-solving. Recent advancements in deep learning architectures [1] and Physics-Informed Neural Networks (PINNs) [2] have enabled powerful representations of complex physical systems, reducing reliance on extensive datasets through physics-driven regularization and constraints. Simultaneously, large language models (LLMs) have demonstrated unprecedented capabilities in reasoning, generalization, and automated knowledge integration [3], opening exciting opportunities for agent-based workflows, automated optimization strategies, and novel decision-making paradigms.
The natural progression of AI now demands interdisciplinary strategies, combining the strengths of traditional deep learning models [4,5,6], physics-based approaches [7], and the versatility of LLMs. Such integrative methodologies promise to enhance our ability to solve computationally expensive inverse problems, navigate high-dimensional optimization landscapes, and rigorously quantify uncertainties. Furthermore, by orchestrating AI frameworks with high-performance computing (HPC) platforms, the engineering and scientific communities can achieve previously unattainable computational efficiency and scale [8,9,10].
This mini-symposium invites researchers to explore cutting-edge methods and applications that integrate physics-informed machine learning, deep learning, advanced LLMs, and HPC to address fundamental and applied challenges. We encourage presentations that propose innovative agentic workflows, demonstrate novel ways to embed physics knowledge within AI models, explore the use of LLMs in multi-agent coordination, or highlight impactful case studies across diverse scientific domains.
We specifically welcome contributions that demonstrate:
⢠Novel implementations and benchmarks of Physics-Informed AI and PINNs.
⢠Application of LLMs in agent-driven design optimization workflows, simulation orchestration, and data-driven decision-making.
⢠Integration of AI-driven workflows with state-of-the-art HPC environments.
⢠Advanced optimization and uncertainty quantification methods empowered by integrated AI and physics-based approaches.
⢠Solutions to computationally challenging inverse problems leveraging combined AI and physics-driven modeling.
Keywords:
Agentic Workflows, High-Performance Computing (HPC), Inverse Problems, Large Language Models (LLMs), Uncertainty Quantification (UQ), Physics-Informed AI
Computational science and engineering applications have benefited from surrogate modeling brought by significant advances in machine learning (ML). By design, these flexible and high-capacity models are able to capture nonlinear and multiscale behavior present in complex physical phenomena. However, training ML-based surrogates for scientific applications often faces the challenges of limited, noisy, or biased data, whether from experiments or high-fidelity simulations. Thus, uncertainty quantification (UQ) becomes a necessary component in both training these models and generating trustworthy predictions, e.g. in digital twin frameworks and decision-making under uncertainty. To tackle these challenges we welcome submissions that include topics related to:
⢠Surrogate construction in and discovery of low-dimensional latent spaces
and low-rank approximations,
⢠Hierarchical surrogate models, e.g. mixture of experts,
⢠Surrogate models equipped with UQ via Bayesian approximations,
⢠Generative or stochastic surrogate models, both supervised and unsupervised,
⢠Forward propagation of surrogate uncertainties into quantities of interest,
e.g. via optimal transport or normalizing flows,
⢠Uncertainty attribution and sensitivity analysis with respect to input uncertainties and surrogate errors,
⢠Probabilistic models for AI agents.
Keywords:
Forward & Inverse UQ, High-Dimensional Problems, multifidelity surrogates, scientific machine learning, surrogate models
Artificial intelligence (AI) and machine learning (ML) are revolutionizing the way researchers address longstanding challenges in contact mechanics. From data-driven modeling and enhanced simulation techniques to novel experimental analysis via advanced imaging, the integration of AI is opening new avenues for understanding and predicting contact-related phenomena across scales, materials, and applications.
This mini-symposium aims to provide a broad and inclusive platform for recent advances at the intersection of contact mechanics and artificial intelligence. We welcome contributions exploring the use of AI in any aspect of contact problems, includingâbut not limited toâfriction, wear, adhesion, roughness, tribology, and interface mechanics. Particular attention will be given to studies leveraging AI for:
- Interpreting experimental images (e.g., DIC, infrared, SEM, or optical microscopy) for contact localization, damage evolution, or microstructure analysis;
- Accelerating computational contact models (e.g., hybrid AI-FEM, surrogate modeling, reduced order models, Physics-Informed Neural Networks, Mixture of Experts, transformers);
- Data-driven prediction of contact behaviors under complex loading and environmental conditions;
- Multiscale approaches combining physical modeling and AI to bridge micro- and macro-scales;
- Applications in engineering sectors such as transportation, manufacturing, energy, and bioengineering;
- Prediction and health monitoring: data-driven prediction of contact behavior, early fault detection, and condition monitoring of interfaces and tribological systems
- Uncertainty quantification, interpretability, and the integration of physical constraints into AI models for contact problems.
We encourage submissions from both academia and industry, including original research, benchmarks, open-source tools, and methodological advances. The symposium seeks to foster interdisciplinary dialogue and highlight the diversity of AI applications in contact mechanicsâpaving the way for robust, interpretable, and efficient next-generation solutions.
Keywords:
AI, AI for Science, Contact, Friction, hybrid approaches with machine learning, Multi-physics, multiscale mechanics
Nonlocal models have become essential tools for capturing complex physical phenomena involving long-range interactions, memory effects, and multiscale structures. By extending the classical framework of local partial differential equations to incorporate finite-range and integral interactions, nonlocal formulations are well-suited for problems in fracture mechanics, anomalous diffusion, porous media, and heterogeneous materials. These models offer increased accuracy and provide natural frameworks for coarse-graining and upscaling, especially in systems with embedded length scales or evolving microstructures.
Nonlocality also plays a central role in coarse-graining and reduced-order modeling, where eliminating fine-scale degrees of freedom give rise to nonlocal closure relations. These nonlocal closures, which incorporate memory and spatial interactions, account for unresolved dynamics and enable more accurate, interpretable surrogate models across scales.
Concurrently, the rapid advancement of machine learning (ML) and artificial intelligence (AI) has reshaped how models are constructed, calibrated, and deployed across the sciences. Recent advances suggest that there is a natural synergy between nonlocality and ML: nonlocal operators can be learned from data using kernel methods or neural networks; neural operators naturally encode nonlocal mappings and structure. Moreover, nonlocality arises not only in physical systems but also in modern learning algorithms, through nonlocal gradients, attention mechanisms, and architectures such as graph neural networks and transformers, highlighting deep parallels between physical and algorithmic nonlocality.
This minisymposium invites contributions at the interface between nonlocal modeling and machine learning/AI, that address diverse scientific and engineering problems. We believe that the following areas will be especially important for future research developments:
⢠Neural operators
⢠Learning nonlocal operators and kernels from data
⢠Nonlocal modeling for damage, diffusion, and materials with microstructure
⢠Data-driven homogenization and coarse-graining
⢠Machine learning and nonlocal gradients
⢠Data-driven nonlocal constitutive models
⢠Machine learning for diffusion problems
⢠Heterogeneities and nonlocality
⢠Probabilistic modeling of nonlocal interactions
⢠Uncertainty quantification for nonlocal and ML-augmented models
Keywords:
coarse-graining, kernel identification, physics-informed machine learning, scientific machine learning, uncertainty quantification, Nonlocal modeling
Machine and deep learning are reshaping physical simulation by enabling the development of surrogate models that either embed known constitutive laws or learn directly from high-fidelity simulation data generated by traditional numerical solvers. These data-driven and hybrid approaches offer new avenues for modeling, prediction, and control of complex systemsâ delivering faster, more efficient, and scalable alternatives to conventional computational methods.
This mini-symposium explores recent advances at the intersection of machine learning and scientific computing for PDEs, and their applications. Topics of interest include deep learning and generative models for surrogate modeling, neural operators, reduced-order modeling, physics-informed and structure-preserving networks, data-driven discovery of system dynamics, and the mathematical foundations underpinning these techniques.
We aim to bring together researchers exploring both theory and applications, working toward robust, generalizable, and physically consistent models for next-generation simulation tools in areas such as structural mechanics, fluid dynamics, materials modeling, and multi-physics simulation.
Keywords:
Deep Learning, neural operators, Physics-Aware, surrogate modeling
In the past few years machine learning techniques have increasingly been employed for solving problems in scientific computing, i.e., the approximation of differential equations. Such efforts are now commonly referred to as âscientific machine learningâ [1]. Just as with classical numerical methods, it has been observed that scientific machine learning benefits from preserving structural properties of the differential equations, e.g., symplecticity [2] and symmetries [3].
This minisymposium invites speakers to present work on preserving geometric structure in machine learning models for applications like system discovery from experimental [4] data and reduced order modeling [5, 6]. Presentations can encompass novel algorithms as well as software implementations and comparisons of existing approaches.
REFERENCES
[1] Baker N, et al. Workshop report on basic research needs for scientific machine learning: Core technologies for artificial intelligence. USDOE Office of Science (SC), Washington, DC (United States), 2019.
[2] Jin P, et al. SympNets: Intrinsic structure-preserving symplectic networks for identifying Hamiltonian systems. Elsevier Neural Networks 132, 2020.
[3] Lishkova Y, et al. Discrete Lagrangian neural networks with automatic symmetry discovery. Elsevier IFAC-PapersOnLine, 2023.
[4] Greydanus S, Dzamba M, Yosinski J. Hamiltonian neural networks. Advances in neural information processing systems 32, 2019.
[5] Buchfink P, Glas S, and Haasdonk B. Symplectic model reduction of Hamiltonian systems on nonlinear manifolds and approximation with weakly symplectic autoencoder. SIAM Journal on Scientific Computing 45.2, 2023.
[6] Brantner B, Kraus M. Symplectic autoencoders for model reduction of Hamiltonian systems. arXiv preprint arXiv:2312.10004, 2023.
Keywords:
geometric mechanics, neural networks, structure preservation
The increasing integration of artificial intelligence (AI) into engineering and applied sciences is reshaping our ability to better understand and potentially control complex systems. Nevertheless, the inherent opacity of many AI methodologies, notably deep learning, frequently restricts their deployment in many applications. Recent advances in explainable AI (XAI) have begun addressing these issues by enabling interpretable insights into AI-driven predictions and decisions. This mini symposium brings together leading researchers and practitioners to showcase state-of-the-art developments in XAI for complex systems arising in engineering and applied sciences. We encourage submissions on topics including mechanistic interpretability, post-hoc interpretability, and causal inference, to show how these methods are accelerating scientific discovery and enabling the control of complex systems. Examples of work in this space include [1] for mechanistic interpretability, [2] for post-hoc interpretability, [3] for XAI in scientific discovery, and [4] for XAI in control. The mini symposium aims to bring the community together, around the central topic of XAI, encouraging interdisciplinary collaboration across different applied disciplines.
REFERENCES
[1] Discovering governing equations from data by sparse identification of nonlinear dynamical systems SL Brunton, JL Proctor, JN Kutz. Proceedings of the national academy of sciences 113 (15), 3932-3937.
[2] Evaluation of post-hoc interpretability methods in time-series classification H TurbĂŠ, M Bjelogrlic, C Lovis, G Mengaldo. Nature Machine Intelligence 5 (3), 250-260.
[3] Explain the Black Box for the Sake of Science: the Scientific Method in the Era of Generative Artificial Intelligence G Mengaldo arXiv preprint arXiv:2406.10557.
[4] Improving turbulence control through explainable deep learning M Beneitez, A Cremades, L Guastoni, R Vinuesa. arXiv preprint arXiv:2504.02354.
Keywords:
Interpretability, Scientific Discovery, Control, Explainable AI
All models in engineering and the applied sciences rely on approximations and uncertain parameters, making their predictions inherently uncertain. Rather than ignoring this, it is essential to quantify, analyze, and incorporate uncertainty into decision-making to ensure safer, more reliable, and robust computational results.
Bayesian methods gather a collection of techniques for learning from data, improving models and predictions, and performing optimization, explicitly accounting for uncertainty. With a solid mathematical foundation, advances in computational algorithms, and increased data availability, Bayesian approaches are now widely used for model calibration, machine learning, surrogate modeling, optimization, and control. These techniques are being applied across diverse domains, from computational mechanics and climate modeling to materials design, biomedical engineering, aerospace, and energy systems.
This minisymposium invites contributions on the development and application of Bayesian methods in computational science and engineering. We welcome new algorithms, theoretical advances, and practical implementations in areas including but not limited to computational solid, fluid, and biomechanics. We also encourage contributions from researchers developing software tools or frameworks that facilitate Bayesian analysis in complex environments.
Our aim is to foster cross-disciplinary exchange among those applying Bayesian thinking to improve the safety, reliability, and predictive power of computational models across a wide range of scientific and engineering applications.
Keywords:
Bayesian methods, Computational modeling, Computational Science and Engineering, Data-driven modeling, Uncertainty Quantification
Engineering decision-making increasingly relies on integrating sensing data with computational models that are both accurate and efficient, enabling real-time insight into system performance and condition. This is particularly vital for the operation and maintenance of structural infrastructure assets throughout their life cycle.
This minisymposium will focus on adaptivityâthe ability of computational models to account for uncertain and evolving environmentsâpositioning them as essential components of intelligent, high-level decision-making ecosystems. Central to this discussion are physics-aware modeling frameworks, including Reduced-Order Models (ROMs), Physics-Informed Neural Networks (PINNs), and Physics-Guided Machine Learning (PGML) approaches. These methods aim to embed physical knowledge into data-driven models, balancing fidelity with computational efficiency. While these approaches have demonstrated success in tasks such as response prediction and surrogate modeling, significant challenges arise when dealing with high-dimensional systems subject to stochastic loading, nonlinearities, damage evolution, and environmental variability. Static models that fail to respond to changes in operational context or rely solely on passive data assimilation limit their effectiveness and robustness in real-world deployments. The objective of this minisymposium is to explore recent advances and open questions in the adaptive integration of sensing data with physics-aware models for real-time decision support. We welcome contributions that address key challenges in this space, including but not limited to:
⢠Model adaptability under environmental and operational variability, including system degradation and damage evolution
⢠Real-time and online data assimilation for continuous model updating
⢠Active data acquisition and learning strategies for improved observability and control
⢠Scalable modeling approaches for complex, high-dimensional, and real-world systems
⢠Applications in digital twins, structural health monitoring, and predictive maintenance
By bringing together researchers from computational mechanics, control, data science, and structural engineering, this minisymposium aims to foster cross-disciplinary dialogue and catalyze future research directions in adaptive modeling for engineering decision support.
Keywords:
adaptive methods, Data Assimilation, Model Order Reduction/Reduced Order Modeling, Physics-Aware, physics-informed machine learning
Time-dependent systems, governed by discrete dynamical models or evolutionary partial differential equations (PDEs), describe various scientific and engineering applications. The combination of handling numerical approximation via time-advancing schemes and, at the same time, discovering differential equations from time series measurements poses significant challenges to scientific machine learning methods (SciML). This minisymposium aims at exploring recent advances in data-driven and hybrid end-to-end SciML approaches for enhancing the modeling and simulation of time-dependent phenomena.
Specifically, our scope is to present recent advances and emerging trends in operator learning approaches designed to model temporal dynamics, including architectures such as recurrent neural networks (RNNs), neural ordinary differential equations (NODEs), attention-based architectures, and other discrete- or continuous-time formulations.
Particular attention will be devoted to techniques and applications that integrate machine learning with classical numerical solvers, accelerate time integration, and improve solution strategies for learning stiff or multiscale surrogate models in multi-query regimes.
In this context, key challenges to be addressed include balancing computational efficiency with numerical fidelity, enhancing generalization and long-term prediction, tuning high-dimensional models with limited data, and preserving physical consistency.
We especially encourage submissions that apply these methods in computational mechanics and biomedical applications, including areas such as computational cardiology, tissue modeling, and poro-mechanics.
Keywords:
dynamical models, evolutionary partial differential equations, scientific machine learning, time-dependent systems
The pursuit of fully virtualized product design workflowsâencompassing initial design, manufacturing processes, performance prediction, and iterative redesignâhas fueled rapid developments in multiscale and materials modeling. Concurrently, emerging machine learning (ML) techniques are being actively explored to accelerate these modeling workflows and reduce the time-to-solution for predicting complex material behaviors. Among these, deep material networks (DMNs) have emerged as a promising framework, offering fast and accurate nonlinear predictions across a broad class of material systemsâincluding particle-reinforced composites, fiber-reinforced structures, and polycrystalline aggregatesâeven when trained solely on linear elastic data. Looking forward, the continued integration of mechanistic insights with data-driven learning will be essential for advancing the frontiers of multiscale modeling and enabling robust, generalizable simulations across engineering applications.
This minisymposium aims to bring together researchers and practitioners at the forefront of computational mechanics, machine learning, and materials modeling. We welcome contributions that span fundamental developments, algorithmic innovations, and industrial case studies, with the goal of identifying key challenges and opportunities in deploying ML-enhanced multiscale methods in practical engineering workflows.
Keywords:
Multiscale Modeling, Machine Learning, Material modeling
The accelerating demand for design of sustainable, efficient, and high-performance vehicles is driving a transformation in engineering methodologies. Among the most promising enablers of this transformation is machine learning (ML), which offers powerful tools for reducing development time, accelerating design space exploration, and enabling real-time decision-making across multiple fidelity levels (Serani et al. 2024, Di Fiore et al. 2024). This Minisymposium aims to gather contributions that explore the integration of ML into vehicle design workflows, with particular focus on early-stage and multidisciplinary processes. Targeted applications include air, sea, and ground vehicles, spanning a broad range of mobility systems. Topics of interest include, but are not limited to: (i) ML-based surrogate modeling and dimensionality reduction for complex, high-dimensional design spaces; (ii) Reduced-order modeling integrated with ML to replace or augment high-fidelity solvers; (iii) Adaptive sampling and optimization strategies across fidelity levels; (iv) Multi-fidelity modeling and data fusion approaches; (v) Explainable and physics-informed ML methods to ensure design transparency and accountability; (vi) ML-driven design methodologies with a focus on sustainability metrics; (vii) Benchmarking and best practices for ML-enhanced design frameworks. The Minisymposium is motivated by active international collaborations investigating how ML can improve performance, robustness, and sustainability of vehicle design processes. By bridging communities working on ML, reduced-order modeling, optimization, and computational mechanics, the session seeks to foster critical dialogue, highlight emerging trends, and build synergies across disciplines and application domains (Mendez et al., 2025).
Keywords:
Reduced-Order Models, Sustainability, Vehicle Design
Equation discovery from data (EQD) or symbolic regression (SR) is a task for artificial intelligence (AI) methods rooted in machine learning and symbolic computation and has garnered significant attention within the realm of computational mechanics and material science. It enables data-driven modelling that is inherently interpretable, i.e., producing equations that best describe a dataset. Because computational mechanics is built around mathematical models, the equations produced by these AI methods are readily incorporated into a variety of preexisting workflows, including analytical derivations used in computational applications. This minisymposium seeks to illuminate the latest breakthroughs and applications of EQD and SR techniques in advancing the simulation, analysis, and optimization of complex mechanical systems.
ââââ
By harnessing the potential of AI for equation discovery, researchers are discovering novel avenues to enhance accuracy, model interpretability, and computational efficiency in the domain of computational mechanics. This minisymposium will explore a comprehensive array of topics that include, but are not limited to:
⢠Equation Discovery and System Identification: AI techniques and SR enable the discovery of governing equations from experimental or simulated data, facilitating the identification of system dynamics and behaviors without a priori assumptions.
⢠Physics-Informed Machine Learning: Integrating domain-specific physical insights into EQD methods may yield hybrid models that combine data-driven learning with established physical laws to enable identification of the physical significance of individual model components while also ensuring greater interpretability, model generalization and robustness.
⢠Uncertainty Quantification, Uncertainty Propagation and Sensitivity Analysis: Techniques for uncertainty quantification and propagation of uncertainties and sensitivities for equation discovery methods and SR within computational mechanics models, providing a deeper understanding of system behavior through traceable uncertainty propagation.
⢠Equation Discovery and SR in Multiscale Modelling: Computational mechanics and material sciences require modelling effects across multiple scales. EQD and SR are capable of producing efficient and accurate surrogates or emulators for computationally expensive numerical models.
⢠Optimization, High-throughput Screening, and Design: The application of EQD and SR techniques in optimization sce
Artificial Intelligence (AI) is becoming a foundational enabler in computational mechanics, offering transformative capabilities for the design, analysis, and optimization of materials and structures across multiple length and time scales. This symposium aims to highlight cutting- edge developments in integrating AI into computational workflows, from atomistic modeling and mesoscale simulations to continuum mechanics and system-level behavior. By leveraging data-driven models, machine learning, and physics-informed approaches, researchers are accelerating the discovery of novel materials, enabling the inverse design of complex structures, and bridging gaps between simulation and experiment. The rise of autonomous design loops, uncertainty-aware models, and real-time data assimilation is also paving the way for the development of digital twins, virtual replicas of materials and structures that evolve with their physical counterparts. Contributions are invited across a broad range of topics including surrogate modeling, multiscale and multiphysics learning, high-throughput simulation, generative design, and AI-enhanced optimization. The symposium welcomes theoretical advances, application-driven studies, and infrastructure tools that push the boundaries of how AI can inform, accelerate, and transform computational solid/structural mechanics and materials design. Interdisciplinary work that connects materials science, mechanics, data science, and applied mathematics is particularly encouraged.
Keywords:
Computational Mechanics, Digital twin, Multi-scale analysis, Optimisation
The application of Artificial Intelligence (AI) to metal forming has shown significant potential in addressing long-standing challenges related to simulation time, process optimization, and material modelling [1, 2]. This Minisymposium invites contributions that explore the integration of data-driven methods with conventional numerical techniques to improve prediction, control, and robustness in metal forming processes.
A key area of interest is the use of machine learning and surrogate models to accelerate computationally expensive simulations, particularly those involving non-linear constitutive behaviour and complex boundary conditions. Emphasis is also placed on methods that incorporate uncertainty quantification, enabling more reliable predictions in the presence of variability in material properties, process parameters, or operating conditions.
Relevant topics include:
⢠Data-driven material parameter identification and model calibration
⢠Surrogate and reduced-order modeLling techniques for forming simulations
⢠Hybrid approaches combining physics-based models with AI algorithms
⢠Sensitivity analysis and uncertainty propagation
⢠Transfer learning and domain adaptation across forming scenarios
⢠Robust and multi-objective optimization under uncertainty
⢠AI-based methods for real-time process monitoring and control
References:
[1] A.M. Habraken, T.A. Aksen, J.L. Alves et al., Analysis of ESAFORM 2021 cup drawing benchmark of an Al alloy, critical factors for accuracy and efficiency of FE simulations. International Journal of Material Forming 15 (2022) 61.
[2] A.E. Marques, T.G. Parreira, A.F.G. Pereira, B.M. Ribeiro, and P.A. Prates, Machine learning applications in sheet metal constitutive Modelling: A review, International Journal of Solids and Structures 303 (2024) 113024.
Keywords:
Machine Learning, Parameter Identification, Surrogate Models, Uncertainty Quantification, Metal Forming, Process Optimization
Generative artificial intelligence is transforming multiple scientific fields. However, in engineering design applications, such as generating component geometries from desired outcomes, significant challenges remain before reaching practical, functional tools. For instance, Engineering design must comply with multiple physical constraints, like continuity of the solution, conservation of mass and energy, stress convexity with respect to dimensions⌠When disposing of a direct problem surrogate, countless scenarios must often be evaluated in brute-force optimization to solve inverse problems, resulting in high computational and time costs. Identifying the optimal inputs for a known desired outcome is referred to as the inverse problem [1]. Inverse problems are known to be intractable, ill-posed [2], and frequently falls into the category of Cauchy problems. Consequently, training a surrogate to predict the solution of an inverse problem is often impractical. Moreover, this approach often lacks a mechanism to enforce the problemâs physical constraints.
This minisymposium aims to showcase recent advances in optimal generative design and foster collaboration within the scientific community. Topics include, but are not limited to: optimal geometrical design generation, constrained inverse-problem solution methods, artificial intelligence approaches to inverse problems, reinforcement-learning-based inverse-problem solutions, generative adversarial networks and autoencoder generation methodologies.
[1] D. Di Lorenzo, V. Champaney, C. Ghnatios, E. Cueto, and F. Chinesta. Physics-informed and graph neural networks for enhanced inverse analysis. Engineering Computation, Nov. 2024. https://doi.org/10.1108/EC-12-2023-0958.
[2] P. Jayaraman, J. Desman, M. Sabounchi, G. N. Nadkarni, and A. Sakhuja. A primer on reinforcement learning in medicine for clinicians. NPJ Digit. Med., 7(1):337, Nov. 2024.
Keywords:
Autoencoder, generative design, Physical constraints, Transformer
Generative models such as diffusion models, flow-matching models, stochastic interpolants, and normalizing flows have shown remarkable success in generating realistic images and videos. More recently, these models have been applied to inverse problems and data assimilation tasks in science and engineering. Notable achievements include the discovery of three-dimensional protein structures and the generation of highly accurate weather forecasts. These advances have spurred a surge of interest in applying generative models across a wide range of scientific and engineering domains. In turn, these applications have driven new research directions focused on method development. Key areas of ongoing investigation include: enforcing physical constraints during sample generation; analyzing convergence properties; developing models capable of handling both sparse observations and continuous streaming data; improving efficiency and scalability; and extending these methods to address multiscale and multiphysics problems.
Motivated by these developments, this minisymposium invites submissions focused on the development, extension, and application of state-of-the-art generative models for inverse problems and data assimilation. Topics of interest include, but are not limited to:
⢠Incorporating physical constraints into generative models
⢠Analyzing generative models as tools for probabilistic inference
⢠Extending generative models to settings with limited training data
⢠Applying generative models for state and parameter inference in dynamical systems
⢠Evaluating the performance of generative models for probabilistic inference
⢠Developing novel formulations for faster and more scalable inference
Keywords:
data assimilation, diffusion and flow matching, inverse problems, generative models
The application of artificial intelligence (AI) technologies in computational mechanics has a long and rich history. However, the integration of recent AI advancesâparticularly deep learning, physics-informed neural networks (PINNs), generative models, and multi-fidelity learningâinto computational mechanics is still in its early stages and rapidly evolving. The objective of this mini-symposium is to explore how AI techniques, including deep learning and other machine learning approaches, can be effectively applied to address fundamental and applied problems in computational mechanics. We warmly welcome contributions that advance the synergy between these two fields, aiming to develop impactful and innovative methodologies. Of particular interest are studies where AI enables the simulation of previously intractable physical phenomena, accelerates large-scale or multi-physics simulations, supports data-driven scientific discovery, or significantly improves the accuracy and efficiency of existing computational models.
Keywords:
Data-driven modeling, Multi-fidelity Learning, surrogate models
Computational Fluid Dynamics has been a cornerstone of scientific computing for decades, enabling the simulation and analysis of complex flows in engineering, geosciences, and biomedical applications. While high-fidelity CFD methods continue to grow in accuracy and capability, their computational cost often limits exploration of large design spaces, real-time decision making, and uncertainty quantification.
In recent years, Machine Learning (ML) and Artificial Intelligence (AI) have emerged as novel, transformative tools for CFD, offering avenues to reduce computational costs [2], improve predictive accuracy [1], extract physical insight from data, and to build reduced-order or hybrid models [3]. However, despite rapid progress, the integration of ML/AI in CFD still faces fundamental open questions: ensuring physical consistency, generalization to out-of-distribution conditions, and effective use of limited or noisy data.
This workshop aims to bring together leading experts from academia, industry, and research labs to: (i) Showcase state-of-the-art ML/AI approaches in CFD, from predictive surrogates to hybrid solver architectures. (ii) Discuss emerging applications in turbulence modeling, multi-phase flows, aerodynamics, and environmental flows. (iii) Address challenges and open questions such as model validation, uncertainty quantification, and reproducibility.
In addition, our aim is to foster discussion, critical evaluation and collaboration across disciplines to advance the adoption of ML and AI methods in CFD.
REFERENCES
[1] D.A. Bezgin, A.B. Buhendwa, N. Adams: JAX-Fluids: A Fully-Differentiable High-order Computational Fluid Dynamics Solver for Compressible Two-phase Flows; Computer Physics Communications, 2023.
[2] N.A.K. Doan, W. Polifke, L. Magri: Physics-informed Echo State Networks; Journal of Computational Science 47, 2020.
[3] B. List, L.W. Chen, N. Thuerey: Learned Turbulence Modelling with Differentiable Fluid Solvers; Journal of Fluid Mechanics, 2022.
Keywords:
Computational Fluid Dynamics, Deep Learning, Turbulence Modeling
With increasing prevalence of data-driven computational tools, novel methods for uncovering complex nonlinear relationships between otherwise-disparate data have achieved tremendous improvements on a range of tasks, including but not limited to problems in computational mechanics. However, measuring causality, and not merely correlation, among these relationships remains a difficult task: many theoretical, practical, and computational issues persist, particularly concerning graphical model recovery, causal attribution, confounding relationships, measurement error, causal time-series models, and complex causal mechanisms. This minisymposium will convene world-class researchers in a forum to present advances in graphical modeling, causal inference, causal discovery, and structured causal models, drawing upon expertise in machine learning, statistics, scientific computing, and specific domain applications in mechanics and materials modeling.
Keywords:
applications, causal, graphs, model discovery, structural model
In the face of increasingly complex natural disasters, particularly the severe challenges posed by earthquakes, the development of Machine Learning (ML) has opened new avenues for enhancing the resilience and safety of urban structures. This mini-symposium aims to bring together researchers and engineers to discuss Machine Learning's latest applications and advancements in Computational Mechanics and Earthquake Engineering.
We cordially welcome contributions on the following topics: machine learning-based structural damage diagnosis and prediction, intelligent structural analysis and design optimization, earthquake early warning and rapid response systems, as well as AI applications in building resilience assessment, smart retrofitting and strengthening techniques for existing structures, disaster response and recovery planning for buildings, intelligent and adaptive building systems, and AI-enhanced hazard modeling and impact assessment.
Contributions may explore a wide range of techniques, including but not limited to deep learning, reinforcement learning, data-driven models, neural networks, probabilistic inference, optimization algorithms, intelligent structural health monitoring, and hybrid applications of AI with traditional computational mechanics models. The focus is on applying these advanced AI methods to significantly improve the safety and performance of urban structures under various disruptive conditions, such as earthquakes. This symposium seeks to foster in-depth exchange and collaboration among specialists in Machine Learning, Computational Mechanics, and Earthquake Engineering. We encourage the development of integrated and cutting-edge AI solutions to mitigate seismic risks and comprehensively enhance overall urban resilience effectively. By sharing knowledge and innovations, we strive to contribute to developing safer and more reliable urban infrastructure.
Keywords:
Earthquake Engineering, Machine Learning
In recent years, Scientific Machine Learning (SciML) has established itself as a powerful, diverse, and rapidly evolving field of research, driving transformative changes in computational mechanics and the sciences more broadly. By enabling fast surrogate models based on deep learning, SciML has opened new possibilities for replacing parts or, in some cases, the entirety of traditional, computationally expensive numerical solvers.
However, most real-world applications require large-scale analyses, whether due to the spatial extent of the domain, the multi-scale nature of the phenomena, the long time horizons, or the small time step and mesh sizes required to capture the phenomena. This mini-symposium will explore innovative approaches to extend and enhance established SciML methods to meet the demands of large-scale applications.
These challenges include the collection and curation of appropriate datasets, particularly in contexts where data acquisition is costly or limited. They also call for the development of efficient deep learning models capable of scaling to inputs involving millions of particles or grid points and perform long-time extrapolation. Additionally, the symposium will address the emerging debate around the use of large, specialized deep learning models versus general foundation models that can be applied across a variety of downstream tasks and potentially amortize training costs.
Keywords:
large scale simulations, Machine Learning, neural operators, scientific machine learning, surrogate models
With increasing prevalence of data-driven computational tools, there is a growing need for multiscale machine-learned methods that can exploit different data scales. Traditional numerical methods exploit scale separation in scientific applications to develop hierarchical or multilevel methods yielding solvers with (near) optimal runtime scaling. However, machine learning methods and architectures have been resistant to multilevel analysis despite the existence of scale separation (e.g. spectral bias) in many scientific training tasks. Thus creating efficient multilevel schemes remains a difficult task, all the while training times for machine learning optimizers are dominating many computational cycles due to quadratic scaling of training costs with respect to the number of network parameters.  However, applying multilevel methods to training will require innovation to overcome the theoretical, practical, and computational issues that persist. For scientific machine learning tasks innovations are needed at the intersection of multiscale representation, efficient spatial decompositions, training dynamics, deep learning architectures, and approximation theory. This minisymposium will convene world-class researchers in a forum to present advances in multilevel methods, approximation, optimization, and deep learning, drawing upon expertise in machine learning, statistics, scientific computing, and specific domain applications in mechanics and materials modeling.
Keywords:
AI for Science, Multigrid, Multiscale and Multilevel Methods
Data-driven methods have achieved tremendous improvements on a range of tasks, including but not limited to problems in computational mechanics. Many of such improvements require high-quality scientific data, and the acquisition of such data is often costly, regardless of whether data is experimental, historical, or simulation-driven. This minisymposium will convene world-class researchers to provide a forum to discuss challenges and opportunities in data curation, statistical machine learning, scientific machine learning, with a specific emphasis on issues concerning data diversity, data generation, multi-use datasets, generative learning, and feature representations. Due to the increasingly interdisciplinary nature of data-driven methods, this session will draw upon expertise in machine learning, statistics, scientific computing, and specific domain applications in mechanics and materials modeling, highlighting practical achievements alongside theoretical results.
Keywords:
data curation, data generation, data science, representation learning, scientific machine learning
Advances in science and engineering increasingly depend on machine learning models to accelerate discovery, design, and decision-making. Yet, many applications operate in low-data regimes, where sample acquisition is costly, time-intensive, or experimentally constrained [1]. Some representative cases include generating high-fidelity datasets for turbulence closure model calibration, geometrical design optimization, and the development of digital twins of biological systems with inherently inaccessible empirical data. In such settings, the ability to extract maximal information per sample becomes critical.
This mini symposium will address the central challenge of how to learn effectively when data is scarce. Topics include adaptive experiment design, active learning, uncertainty quantification, and optimization, as well as surrogate modeling architectures that integrate known (or partially known) physics, use multi-level and multi-fidelity data or employ reduced-order approaches for efficient preprocessing [2]. A particular focus will be on models that adapt their sampling strategies as they learn, iteratively refining surrogates to improve accuracy and robustness even under sparse or noisy measurement conditions.
By combining intelligent sampling with cost-aware model design, these approaches enable fast-to-train, generalizable surrogates suitable for deployment in environments where reducing model uncertainty and computational efficiency are essential.
REFERENCES:
[1] U. Fasel, J. N. Kutz, B. W. Brunton, and S. L. Brunton. Ensemble-sindy: Robust sparse model discovery in the low-data, high-noise limit, with active learning and control. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, 478(2260):20210904, 2022.
[2] Fu, Y., Zhu, X. & Li, B. A survey on instance selection for active learning. Knowl Inf Syst 35, 249â283, 2013.
Keywords:
active learning, adaptive sampling, physics-informed machine learning, multifidelity surrogates, multilevel surrogates, optimization
The optimization of energy consumption and heat recovery in industrial processes is a priority challenge in modern engineering, particularly in systems that employ mixing tanks for thermal operations. These units, widely used in sectors such as agro-industry, food engineering, water treatment, and the chemical industry, require precise control of temperature, mixing homogeneity, and the utilization of residual heat. Artificial intelligence (AI), particularly predictive models based on machine learning, offers advanced tools to model and optimize the thermodynamic behaviour of these systems under variable operating conditions.This study proposes the development and implementation of predictive models trained with both experimental data and numerical simulations to estimate in real time the key parameters that determine the energy efficiency of mixing tanks. The approach integrates variables such as inlet and outlet temperatures, mixing flow rate, fluid properties (viscosity, density, specific heat capacity), supplied power, thermal losses, and heat recovery through associated exchangers. The AI architecture combines deep neural networks for nonlinear variable estimation with optimization algorithms based on metaheuristics (such as Particle Swarm Optimization or Genetic Algorithms) for dynamic adjustment of operating parameters. The proposed model not only predicts thermal behaviour but also recommends operating strategies to maximize residual heat recovery, thereby reducing primary energy demand. For validation, data obtained from experimental prototypes and Computational Fluid Dynamics (CFD) simulationsâconsidering mixing phenomena, heat transfer, and temperature gradientsâare employed. The accuracy of the predictions is evaluated using metrics such as Root Mean Square Error (RMSE) and the coefficient of determination (R²), achieving over 95% accuracy in test scenarios. The practical implementation of the model is integrated into a real-time monitoring and control platform, enabling operators to automatically adjust agitation conditions, flow rates, and target temperatures. This approach directly contributes to reducing energy consumption and improving the sustainability of industrial processes, aligning with circular economy strategies and carbon footprint reduction goals. Preliminary results demonstrate that combining AI with advanced predictive modelling in mixing tanks increases thermal efficiency.
Keywords:
Energy Efficiency, Heat Recovery, Artificial Intelligence
Scientific Machine Learning (ML) and Artificial Intelligence (AI) now impact many diverse fields of engineering and science that include applications in sustainability, medicine and biology, and other physical sciences. These applications have spurred great interest and work on this topic and have led to development of novel algorithms and architectures. However, a more rigorous analysis of these algorithms, which is necessary for understanding their performance and pitfalls, is lacking. With this as motivation, this MS will focus on the topics that are tied to developing a more fundamental understanding of the algorithms of Scientific ML and AI. Topics of interest include analysis of the convergence of these algorithms with increasing data and complexity, and the development of new algorithms with quantifiable measures of performance. The domains of interest include, but are not limited to operator learning, optimization algorithms, transport maps, generative algorithms including transformer architectures, structure preserving frameworks, and novel function representation models.
Keywords:
data science, scientific machine learning
Recent advances in Generative AI are transforming how we model, analyze, and predict complex scientific and engineering systems. Traditional modeling workflowsâprogressing from observational data to governing equations and then to simulation and analysisâoften struggle with the high dimensionality, multi-physics coupling, and nonlinearity inherent in real-world systems. Moreover, such systems are frequently affected by both aleatory (inherent randomness) and epistemic (model-form) uncertainties, which pose significant challenges for accurate prediction and robust decision-making. To address these issues, generative modelsâincluding diffusion models, variational autoencoders, and score-based generative approachesâoffer powerful new tools for learning latent structures, sampling from complex distributions and constructing surrogate or reduced-order models. These models can enable efficient modeling, even in regimes where traditional simulation is prohibitively expensive. For instance, in multiscale dynamical systems, generative models can learn to represent and sample from slow manifolds, capturing long-term behavior without requiring full-scale simulation. However, the application of generative AI in scientific domains is still constrained by challenges such as limited training data, high-fidelity modeling costs, and the need for physical interpretability and generalization. To unlock the full potential of these methods, further advances are needed in geometry-aware generation, data-efficient learning, and physics-informed model design. This MS aims to bring together leading researchers in generative AI, uncertainty quantification, and scientific machine learning to explore how next-generation generative models can accelerate and improve scientific discovery and engineering design. We welcome contributions that bridge theory and application across disciplines, demonstrating how generative models can be tailored to respect physical laws, reduce computational cost, and enhance predictive reliability in complex systems.
Keywords:
Manifold Learning, Score-based Diffusion Models, Generative AI, Machine/Probabilistic Learning
Particle-based processes play a pivotal role in a wide range of engineering applications, from powder synthesis and granular flow handling to advanced manufacturing of energy storage materials. The increasing complexity of these processes, especially in high-value applications such as lithium-ion battery electrode production, demands modeling frameworks (e.g., DEM, FEM, CFD, SPH, LBM) that can integrate physical understanding with data-driven insights. Hybrid modeling, which synergistically combines first-principles simulations with machine learning and artificial intelligence, has emerged as a promising paradigm to address these challenges.
This minisymposium will focus on recent advances in computational methods for hybrid modeling of particle-laden and particulate-based processes across multiple scales, with a special emphasis on their application to the handling of particulate materials. Topics of interest include the integration of discrete and continuum models with AI/ML for predictive accuracy and generalization; multiscale and multiphysics particulate system simulation from nano to macro scale; hybrid digital twins for process optimization and control; data-driven discovery of governing laws in particle technology using symbolic regression and physics-informed neural networks; simulation-based design and optimization of particle processes in battery manufacturing, additive manufacturing, and powder metallurgy; and uncertainty quantification and validation frameworks for hybrid models , including emerging AI approaches such as Large Language Models (LLM) and Mixture of Experts in bulk solids processing.
This minisymposium will bring together experts from computational mechanics, process engineering, particle technology, and data science to develop robust and interpretable modeling frameworks for complex particulate processes. We particularly welcome contributions on novel computational strategies, experimentalâcomputational integration, and industrial applications.
Keywords:
Battery Manufacturing, Machine Learning, Particle Processes, Hybrid Modeling
Interfaces govern critical behaviors in systems ranging from heat transfer, material failure and fluid flow in heterogeneous microstructures to fault slip in geophysical systems. Such problems frequently involve discontinuous solutions and moving boundaries, presenting significant challenges for traditional mesh-based numerical methods such as the finite element method.
Recent advances in physics-informed machine learning (PIML) offer powerful tools to complement and enhance conventional computational approaches by embedding measured data into models grounded in first-principles physics. PIML methods have demonstrated potential in both forward and inverse interface problems, although important challenges remain. This minisymposium will highlight recent developments in PIML methods and their applications to interface problems in mechanics, with the goal of fostering collaboration among researchers in computational mechanics, applied mathematics, and machine learning.
Topics of Interest include, but are not limited to:
⢠Methodological advances in PIML for interface problems
(e.g., physics-informed neural networks (PINNs), physics-informed neural operators (PINO), Fourier neural operators (FNO), deep operator networks (DeepONet))
⢠Applications to multiphase flow, heterogeneous material systems, poromechanics, and fracture mechanics (including crack propagation)
⢠Handling discontinuities, moving boundaries, and complex geometries with PIML
⢠Inverse problems and system identification involving interfaces
⢠Hybrid computational frameworks combining PIML with traditional FEM or other numerical methods
⢠Benchmarking, validation, optimization and uncertainty quantification in PIML for interface problems
Keywords:
hybrid modeling, interface problems, neural operators, Physics-informed machine learning, uncertainty quantification (UQ)
Digital twins have received significant attention and development in recent years due to the promise of enabling real-time design and analysis of complex systems. However, realizing this potential will require supplementing high-fidelity numerical representations with reduced-order and data-driven surrogate models that can be queried in real-time. Meanwhile, scientific machine learning is an emerging discipline that merges scientific computing and machine learning. Whilst scientific computing focuses on large-scale models that are derived from scientific laws describing physical phenomena, machine learning, including deep learning, focuses on developing data-driven models which require minimal knowledge and prior assumptions. With the contrast between these two approaches follows different advantages: scientific models are effective at extrapolation and can be fitted with small data and few parameters whereas deep learning models require a significant amount of data and a large number of parameters but are not biased by the validity of prior assumptions. Scientific machine learning endeavors to combine the two disciplines in order to develop models that retain the advantages from their respective disciplines. This mini-symposium collects recent works on scientific machine learning methods in the context of enabling real-time inference for digital twins. We anticipate contributions covering theories and algorithms for both forward and inverse problems with applications in engineering, sciences, and scientific computing.
Keywords:
Inverse Problems, Surrogate Models, digital twins, Machine Learning
There is a growing demand in modeling and simulation to capture and predict increasingly intricate details of complex physical systems including fluid dynamics, structural mechanics, or multiphysics phenomena. This growing demand, coupled with the use of fine spatial and temporal resolutions, generally leads to the development of large-scale high-fidelity models that exhibit superior predictive capability of a systemâs behavior. However, the increased fidelity comes with significant trade-offs, including higher computational demands and data storage limitations. These challenges are further aggravated in outer-loop applications, such as optimization, design, or uncertainty quantification, where multiple model evaluations are required. A promising approach to mitigate these challenges is the development of lower fidelity surrogate models which approximate the high-fidelity model and incur a lower computational cost. The focus of this minisymposium is to showcase novel surrogate modelling methods that accelerate large-scale high-fidelity simulations, and that address the critical challenges associated with their effective use in real-world applications.
Keywords:
Reduced Order Modeling, scientific machine learning, surrogate modeling
Simulations and experimental processes in scientific and engineering applications often generate large amount of data. Data can be leveraged in a variety of Scientific Machine Learning (SciML) workflows, but data/dimension reduction strategies need to be first employed to reduce/compress, curate, and process these data.
In this minisymposium, we welcome contributions focused on state-of-the-art strategies for dimension reduction and data compression with a particular emphasis on approaches relevant to SciML. In particular, we invite contributions that address one or more of the following topics: i) data compression and dimension reduction strategies that account for the preservation of relevant physical features; ii) approaches for the identification of low-dimensional structures and manifolds that can scale to large and/or distributed datasets; iii) data-driven approaches to data and dimension reduction that can leverage information sources characterized by varying degrees of accuracy and fidelity.
We welcome contributions focused on algorithmic and/or theoretical advancements, but advanced deployments of these techniques to large scale scientific and engineering applications are also in scope.
SNL is managed and operated by NTESS under DOE NNSA contract DE-NA0003525
Keywords:
data reduction, dimensionality reduction, scientific machine learning
Artificial intelligence, machine learning, and data science are transforming predictive capabilities in multi-physics systems. These technologies offer unprecedented advances in accuracy and uncertainty quantification for complex applications spanning energy, chemical processing, environmental science, healthcare, and transportation. This symposium will address the critical challenges of dynamic complexity, physics coupling, non-linear behaviors, and computational scalability through next-generation AI-driven approaches.
The symposium will showcase cutting-edge advances at the intersection of computational methods, data sciences and intelligent systems. It will foster discussion and collaboration among scientists and professionals on cutting-edge novel algorithms, data-driven methodologies, and hardware implementations that push the boundaries of predictive multi-physics modelling. In particular, it will cover fundamental research areas on computational methods, machine learning algorithms, data sciences, optimisation and data assimilation.
All application areas in structural mechanics, single/multi-phase flow dynamics, radiation and robotics are welcome. Contributions are sought on, but not limited to, the following topics:
- Computational structural/fluid/radiation dynamics;
- Predictive methods for sensitivity and uncertainty quantification;
- Coupling models for multi-physics problems, and;
- Data integration and assimilation.
Keywords:
AI-Augmented Simulation, Computational Intelligence in Engineering, Data-Driven Physics Modeling, Machine Learning for Multi-Physics, Multi-Fidelity Data Integration
This minisymposium focuses on recent advances in integrating Machine Learning (ML) techniques with classical numerical methods to enhance the solution of Partial Differential Equations (PDEs), which are at the core of computational science and engineering. Traditional PDE solvers often face significant challenges in handling nonlinear, high-dimensional, multi-scale, and multi-physics problems, particularly in extreme-scale applications. Machine learning provides novel opportunities to overcome these limitations, offering new pathways to improve efficiency, scalability, and predictive power.
The session will explore a wide spectrum of approaches, including physics-informed neural networks (PINNs) and operator learning methods for directly approximating PDE solutions, ML-accelerated iterative solvers for large-scale systems, and hybrid frameworks where ML components are seamlessly embedded into finite element, finite volume, or spectral methods. Contributions addressing the role of ML in model reduction, adaptivity, preconditioning, and multi-grid methods are particularly welcome. Special attention will be given to methods that preserve physical consistency, generalize across problem classes, and integrate naturally with high-performance computing environments.
By bringing together experts in numerical analysis, computational science, and machine learning, this minisymposium aims to foster interdisciplinary dialogue and highlight emerging methodologies that will shape the future of PDE-based simulations across science and engineering. Topics of interest include, but are not limited to:
⢠Physics-informed neural networks (PINNs) and variants
⢠Operator learning approaches (DeepONets, Fourier neural operators, etc.)
⢠ML-accelerated iterative solvers and preconditioners
⢠Hybrid PDE solvers combining ML and classical methods
⢠Reduced-order modeling with ML
⢠ML-based multigrid and domain decomposition methods
⢠Uncertainty quantification with ML-enhanced PDE solvers
⢠Integration of ML-enhanced solvers with HPC
Keywords:
Hybrid Machine Learning - Numerical Methods, Operator Learning, PDE solvers, scientific machine learning, surrogate modeling
Machine learning (ML) is revolutionizing molecular dynamics (MD) simulations and opening new potential applications in mechanics. Recent advancements such as machine learning interatomic potentials (MLIPs), geometric deep learning, and generative modeling have enabled simulations of unprecedented scale, fidelity, and efficiency. This mini-symposium will provide a platform for researchers across disciplines to present and discuss cutting-edge developments at the interface of ML and molecular simulations, with an emphasis on both methodological innovations and impactful applications in mechanics, materials, and structural design.
Keywords:
Machine Learning, Molecular Dynamics, Sampling, Simulations
700
Digital Twins
Computational modeling plays a crucial role in biomedical science and engineering. These mathematical models and robust numerical methods contribute significantly to understanding physiological and pathological processes, optimizing medical device design, and enhancing decision-making in clinical trials. Furthermore, to enable clinical translation, digital twinning, and in-silico trials, the development of highly efficient and cost-effective techniques for model personalization is essential.
Recent advancements in scientific machine learning offer promising avenues for this purpose. Data-driven approaches, such as deep learning, kernel methods, and reduced-order modeling, provide reliable surrogates for complex biological phenomena by leveraging large datasets while reducing computational costs compared to traditional physics-based models. Additionally, novel hybrid methodologies that integrate physics-based and data-driven techniques, such as physics-informed machine learning, can further mitigate nonphysical artifacts and enhance the generalization of predictive performance. These approaches have also proven effective in solving inverse problems across various biomedical domains. Moreover, probabilistic techniques such as Bayesian optimization and generative approaches are widely applied in model calibration, uncertainty quantification, and virtual population generation, enhancing the reliability and generality of computational estimates.
This mini-symposium will bring together applied mathematicians, computational scientists, biomedical engineers, and clinicians to discuss state-of-the-art developments in physics-based and data-driven methods for computational biomedicine. Topics of interest include, but are not limited to, patient-specific model parameterization and optimization, advanced machine learning methods for real-time modeling and many-query scenarios, robust Bayesian methods for model calibration and uncertainty quantification, the integration of multi-scale models with clinical data for personalized medicine, in-silico trials for regulatory science and therapy evaluation, and the industrial applications of medical digital twins in device design and testing. By fostering discussions on these cutting-edge topics, we aim to encourage future collaborations and expand the international research network in the emerging field of digital health and medical digital twins.
Keywords:
Biomedical Science and Engineering, Digital Twin, In-silico Trial, Physics-based modeling
Agent-based models have proven to be useful tools in supporting decision-making processes in different application domains. The advent of modern computers and supercomputers has enabled these bottom-up approaches to realistically model human mobility and contact behavior.
Taking the next step and combining realistic agent-based models with up-to-date or real-time data yields powerful agent-based digital twins that can help to save lifes or protect human health endangered in situations of disaster or catastrophes.
In [1], the authors have recently presented a city-scale agent-based model that was capable to evaluate advanced testing and isolation strategies during the COVID-19 pandemic. Bi-directional coupling of simulations as proposed in [2], at the example of an agent-based pedestrian simulation with an electrical simulation in an airport terminal, allows to test contingency plans in saboteur scenarios.
However, for agent-based digital twins to be useful in different situations of disaster very different requirements need to be fulfilled. For large-scale catastrophes such as pandemics or country-wide network shutdowns, the corresponding agent-based models need to scale well on supercomputing infrastructure to allow the simulation of millions of agents in a fraction of second. For small-scale perils such as the local leakage of a chemical hazard, fine-granular pedestrian dynamics are needed. Moreover, future integration of real-time data with agent-based simulations could leverage its applications in adaptive emergency planning.
By attending this symposium, participants will gain a deeper understanding of agent-based digital twins that can help decision makers and domain experts to protect humans from danger caused by disasters of different nature.
The symposium will explore different approaches to agent-based digital twins with its domain specific challenges, including:
⢠Fine-granular pedestrian movement models to protect humans in situations of chemical, thermic or mechanical dangers or catastrophes
⢠Small- to large-scale models to react to network or cyber catastrophes
⢠Large-scale models to protect human health in times of pandemics
⢠Impact estimation of disasters in the context of contingency planning and risk analysis
⢠Integration of real-time data with agent-based models for improved emergency management
Keywords:
digital twins, disaster, emergency management, pandemics, risk analysis, sabotage, Agent-based modeling
Digital twins are dynamic digital counterparts of physical assets that enable simulation-driven decision-making by bidirectionally coupling physics-based models (physical twin) with real-world data [1]. Originally emerging from the manufacturing sector within the framework of Industry 4.0, digital twins are now being applied across diverse fields such as healthcare, education, meteorology, and construction [2], as well as critical infrastructures, such as road networks, bridges, wastewater treatment plants, and energy systems â systems that are costly, long-lived, and safety-critical. The digital twinsâ predictive power and value in operational settings hinge on advanced computational mechanics, real-time simulation capabilities, and robust data-model integration.
Developing reliable digital twins for infrastructures requires extensive interdisciplinary collaboration â bridging domain-specific modeling, numerical simulation, data assimilation, and machine learning to effectively develop and connect sub-models, datasets, and interfaces. Unlike traditional simulation pipelines, digital twins must address real-time performance constraints, online data integration, and dynamic model updating â particularly challenging in high-dimensional, multi-physics, or nonlinear systems. Ensuring robustness and interpretability in safety-critical contexts introduces additional demands on numerical stability, uncertainty quantification, and model hierarchy design.
Topics of interest include, but are not limited to:
⢠Efficient numerical models (e.g., FEM, ROMs) and scientific machine learning approaches (e.g., PINNs, Neural Operators) for simulating the complex physical asset
⢠Acquisition, preprocessing, and assimilation of sensor data from real objects and experiments as data source for model calibration and real-time updating
⢠Coupling of heterogenous models and data into a unified digital representation (e.g., model hierarchies, hybrid physics-data models)
⢠Twinning approaches to keep the real object and its digital representation consistent
⢠Digital twin system architectures and domain-specific deployment strategies,
⢠Quantification and propagation of uncertainties within digital twins.
REFERENCES
[1] M. Asch, A Toolbox for Digital Twins: From Model-Based to Data-Driven, Philadelphia, PA: Society for Industrial and Applied Mathematics, 2022.
[2] A. Rasheed, O. San and T. Kvamsdal, âDigital Twin: Values, Challenges and Enablers From a Modeling Perspective
Keywords:
Digital Twins, Experiments and Sensors as Data Source, Infrastructure, Physical Models
Digital twins are transforming engineering and applied sciences by enabling real-time monitoring, simulation, and predictive analysis of physical systems and processes. As outlined in the 2024 report by the National Academies of Engineering, Science, and Medicine [1], digital twins differ from both forward digital models and digital shadows [2]. A digital twin is a tailored virtual representation that captures key attributes of a physical system or process. This digital representation synchronizes with its physical counterpart by assimilating sensor data and refining predictive capabilities. Through this continuous updating, digital twins can simulate what-if scenarios, supporting predictive decision-making aimed at maximizing value.
Conventional digital twins primarily rely on fixed computational models and passive data assimilation, which limit their adaptability in uncertain and dynamic environments. Both research and industry recognize the need for a new level of autonomy and resilience in digital twins by closing the loop between perception and action â fostering sentient digital twins equipped with learning and adaptation capabilities that actively seek information to improve situational awareness and manage the evolution of their environment. At the same time, new forms of computation can emerge within the physical counterpart itself, enabling embedded computational capabilities for processing external stimuli or assimilating sensor data.
This session aims to gather contributions highlighting the impact of self-learning, adaptation, and information seeking strategies on the ability of digital twins to react to uncertain and dynamic environments. Contributors are invited to discuss topics including (but not limited to) self-learning and self-adaptivity in smart structures; uncertainty quantification, propagation, and resolution in digital twins; parameter and state estimation for decision support; computationally efficient signal interpretation; hybrid physics-data approaches; multi-agent control of physical systems; and (physical) reservoir computing.
[1] National Academy of Engineering and National Academies of Sciences, Engineering, and Medicine, Foundational Research Gaps and Future Directions for Digital Twins, 2024.
[2] W. Kritzinger, M. Karner, G. Traar, J. Henjes, W. Sihn, Digital Twin in manufacturing: A categorical literature review and classification, 2018.
Keywords:
Decision-making, Learning, Perception, Reservoir computing, Scientific computing
Digital Twins are transforming industrial systems by enabling real-time monitoring, predictive analytics, and intelligent control through the integration of physical assets with their digital counterparts. As industrial applications become more complex and data-rich, scalable computational methodsâoften augmented by artificial intelligence and machine learningâare extremely important for the realization of digital twins with guaranteed performance, adaptability, and reliability.
This minisymposium focuses on recent advances in scalable computational methods that support the development and deployment of Digital Twins in industrial settings. We invite contributions that address the algorithmic, methodological, and implementation challenges associated with high-fidelity, real-time, and large-scale digital twin applications.
Topics of interest include, but are not limited to:
⢠Model order reduction and surrogate modelling for real-time simulation in industrial-scale applications
⢠Uncertainty quantification and robust model calibration for reliability and trustworthiness
⢠AI- and ML-based data-driven modelling techniques for enhanced autonomy and adaptability
We particularly encourage submissions that demonstrate the application of these or similar methods to real-world industrial systems, highlighting the synergy between computational innovation and practical impact.
Keywords:
AI-driven modelling, digital twins, reduced order models, Scientific computing, scientific machine learning, simulation and optimization, Uncertainty Quantification
Throughout their life cycle, physical systems experience evolving behaviors and properties due to factors such as wear, environmental exposure, and operational stress. With the advancement of sensor technologies and computational modeling, the creation of High-Fidelity Digital Twins (HFDTs), virtual replicas of physical systems, has become increasingly feasible across a wide range of engineering applications. A cornerstone of HFDT development is system identification, which involves assessing the current state of a system and detecting areas of degradation or vulnerability. This typically requires solving inverse problems through effective parametrization, often framed as complex optimization tasks. Strategic sensor placement further enhances this process, making optimal sensor deployment a critical component in the development of HFDTs.
This mini-symposium will explore cutting-edge techniques in system identification, optimization, sensor placement, feedback, as well as hybrid approaches that integrate machine learning with traditional physics-based numerical methods in the creation of HFDTs. Applications can be from (but are not limited to) civil engineering, bio-medical engineering, aerospace engineering, and mechanical engineering. The topics will include advanced optimization strategies such as gradient-based algorithms, genetic algorithms, Bayesian optimization, and data-driven methods that improve the accuracy, robustness, and computational efficiency of digital twin approaches.
We also encourage discussions on case studies that demonstrate the practical application of these methodologies. Our goal is to foster interdisciplinary collaboration and drive innovation in the use of HFDTs to enhance system performance, safety, and sustainability. By examining current challenges, best practices, and emerging technologies, including AI-enhanced predictive maintenance and decision-making tools, we aim to empower participants with actionable insights for building more resilient and intelligent engineering systems.
Keywords:
Digital Twins, high-fidelity models, hybrid approaches with machine learning, inverse problems, optimal sensor placement
The design and deployment of Digital Twins (DTs) of a physical asset require seamless integration of multi-scale / multi-physics numerical models and observational data, which can have various degrees of fidelity. Despite significant advances in hardware, computing, and data analytics, complex systems of practical significance defy their robust and reliable high-fidelity, DT-like representations. The successful deployment of DT technology calls for a new developmental paradigm, requiring new mathematical and computational methods that integrate physics-based modeling with data-driven methods, operating across different spatial and temporal scales and supporting real-time decision making under uncertainty.
This minisymposium brings together researchers in mathematics, computational science, and engineering to explore recent advances and open challenges in the development of mathematically grounded DTs. Topics of interest include:
1) methods for bidirectional data flow between digital and physical systems,
2) model reduction methods that preserve essential physics while enabling real-time performance,
3) operator learning of partial differential equations,
4) scalable methods for uncertainty quantification and probabilistic inference,
5) optimization and control strategies tailored for live environments,
6) hybrid methods integrating scientific machine learning with traditional simulation methods,
7) complex systems representations,
8) continual learning systems.
â¨By focusing on the mathematical foundations and computational methods that underpin DT technology, this MS aims to create new collaborations across disciplines and foster the development of rigorous, reliable, and scalable DTs.
Keywords:
complex systems, control theory, real-time computing
Digital twins represent a transformative approach for integrating computational models, sensor data, machine learning, and real-time analysis to support engineering decisions for dynamical systems. However, a system -- whether linear or non-linear -- is subject to uncertainties arising from system parameters, environmental conditions, experimental setups, and complex non-linear phenomena (e.g., large deformations, material non-linearities, contact, and multi-physics couplings). It opens many questions on how those uncertainties must be integrated in the context of digital twins.
This mini symposium aims to bring together advances in digital twin methodologies, numerical optimisation, inverse problems, and uncertainty quantification for dynamical systems, with a particular focus on non-linear behaviours. Contributions may address theoretical developments, computational methods, or experimental validations, including but not limited to:
- Physics-based models combined with machine learning
- Real-time model updating and parameter identification
- Sensor placement and optimal experimental design
- Stochastic modelling in structural dynamics
- Multi-scale and multi-fidelity uncertainty propagation
- Reliability analysis and robust design under uncertainty
- Parametric and structural optimisation in non-linear dynamics
- Model calibration and inverse methods under uncertainty
- Surrogate modelling for high-dimensional problems
We welcome contributions that bridge gaps between digital twin technologies, computational optimisation, and uncertainty modelling for improved prediction and decision-making. Example of applications: structural dynamics, wind turbines, aircraft engines, rotordynamics, automotive/aerospace industry.
REFERENCES
[1] Wagg, D.J. et al., The philosophical foundations of digital twinning, Data Centric Engineering, 6, e12, 2025. doi:10.1017/dce.2025.4
[2] Ritto, T.G., Rochinha, F.A. Digital twin, physics-based model, and machine learning applied to damage detection in structures. Mechanical Systems and Signal Processing, 155, 107614, 2021
[3] Wagg, D.J., Worden, K., Barthorpe, R.J., Gardner, P. ASCE ASME Journal of Risk and Uncertainty in Engineering Systems Part B Mechanical Engineering, 6(3), 030901, 2020
Keywords:
inverse problems, Nonlinear Systems, Optimisation, Physics-Based Data-Driven Modeling, Uncertainty Quantification
Digital twins are emerging as powerful tools for real-time simulation, monitoring, and
predictive control of oceanic and atmospheric systems. Creating effective digital twins for
such complex environments requires computational efficiency without sacrificing
interpretability and accuracy. This mini-symposium focuses on advanced Reduced Order
Models (ROMs) and hybrid, physics-informed machine learning approaches, offering real-
time performance while preserving the critical physical characteristics of environmental
processes.
Oceanic and atmospheric phenomena, characterized by multiscale and multiphysics
interactions, pose substantial computational challenges, particularly for high-resolution and
real-time applications. Hybrid approaches leveraging physics-informed neural networks
(PINNs), dynamic mode decomposition (DMD), proper orthogonal decomposition (POD),
and generative AI techniques bridge the gap between computational feasibility and physical
fidelity. These methods embed domain-specific knowledge into data-driven frameworks,
enhancing reliability, interpretability, and predictive capabilities of digital twins.
Target applications of interest include digital twins for offshore energy systems (e.g., wind
farms, wave energy converters, oil/gas platforms), aerosol-cloud interactions in climate
modeling, ocean-atmosphere coupling dynamics (e.g., hurricane forecasting, air-sea
interactions), and atmospheric pollution transport. Contributions highlighting advancements
in multi-fidelity modeling, data assimilation techniques, uncertainty quantification, and
computational efficiency improvements are particularly encouraged.
Aligned closely with WCCM-ECCOMAS 2026 themesâDigital Twins (700), Multiscale and
Multiphysics Systems (1600), and Scientific Computing (1800)âthis symposium will foster
interdisciplinary exchange among researchers from computational mechanics, environmental
engineering, climate science, and applied mathematics communities. Participants will discuss recent developments, challenges, and future directions to advance real-time, scalable, and
interpretable digital twins for oceanic and atmospheric systems.
Keywords:
atmosphere, earth sciences, multi-scale simulations, ocean, reduced order modeling
Digital twins have emerged as transformative tools, enabling real-time simulation, analysis, and control of complex physical systems by integrating computational models, sensor data, and advanced analytics. This mini-symposium will highlight recent advances and emerging computational technologies driving the development, refinement, and application of digital twins across engineering and applied sciences. Topics covered include machine learning, high-fidelity modeling, data assimilation, uncertainty quantification, real-time analytics, and scalable computational platforms. The symposium seeks to foster discussions on innovative methodologies and computational frameworks, aiming to enhance the predictive capabilities, accuracy, and reliability of digital twins. By bringing together leading researchers and practitioners, the event will promote interdisciplinary collaboration, accelerating the adoption and integration of digital twins for improved decision-making and operational efficiency in engineering and applied scientific domains.
Keywords:
Computational Engineering, Reduced Order Modeling , Digital Twins
The global shift toward fossil-free energy hinges critically on advanced computational methods that enable the design of novel structures by predicting chemo-thermo-mechanical system response. Computational methods are the unseen force propelling the fossil-free energy revolution, enabling the creation of advanced technologies and their seamless integration into a sustainable energy ecosystem.
Keywords:
Battery, Energy storage, Fuel cell, Multiphysics simulations, Hydrogen
As engineering applications face escalating demands for sustainability, efficiency, and resilience, there is a pressing need for advanced digital frameworks that support design innovation, continuous monitoring, and informed decision-making. Digital Twins provide a transformative solution: serving as a continuously evolving virtual counterpart of physical systems that integrates physics-based models, multisource data, and uncertainty quantification to enable real-time insight, predictive analytics, and lifecycle optimization for next-generation greener technologies [1-3]. This minisymposium will showcase recent advances in digital twin methodologies for sustainable engineering applications across air, land, and sea. Topics of interest include the integration of scientific machine learning, surrogate modeling, and model reduction techniques for real-time performance, as well as Bayesian approaches to inverse problems. Additional themes will cover data assimilation and continuous model updating, interpretable machine learning, multisource and multifidelity active learning, along with uncertainty quantification and propagation. Contributions demonstrating applications to predictive maintenance, energy efficiency, emission reduction, and lifecycle design optimization are particularly encouraged. The minisymposium will highlight both methodological innovations and cross-domain applications, illustrating the power and versatility of the digital twin paradigm in advancing sustainable engineering solutions.
REFERENCES
[1] AIAA Digital Engineering Outreach and Integration Committee, Digital twin: Definition & value, AIAA and AIA Position Paper, 2020.
[2] L. Mainini, M. Diez, Digital Twins and their Mathematical Souls, In STO-MP-AVT369 Research Symposium on Digital Twin Technology Development and Application for Tri-Service Platforms and Systems, 2023.
[3] The National Academies Collection: Reports funded by National Institutes of Health, Foundational Research Gaps and Future Directions for Digital Twins, Washington DC, National Academies Press (US), 2024 Mar 28. PMID: 39088664.
Keywords:
Lifecycle Design Optimization, Predictive Maintenance, Real-time Monitoring, Digital Twins
Digital models (DMs) are designed to be replicas of systems and processes. At the core of a digital model (DM) is a physical/mathematical model that captures the behavior of the real system across temporal and spatial scales. One of the key roles of DMs is enabling âwhat ifâ scenario testing of hypothetical simulations to understand the implications at any point throughout the life cycle of the process, to monitor the process, to calibrate parameters to match the actual process, and to quantify the uncertainties. This mini-symposium presents the latest developments in real-time forecast and calibration approaches for digital twins. Approaches that are equipped with uncertainty quantification are especially highlighted in the mini-symposium.
Keywords:
digital models, digital twins, forecast, calbration, scientific machine learning
A digital twin (DT) is a computational model that evolves over time to persistently represent the structure, behavior, and context of a unique physical system or process. DTs are characterized by a dynamic and continuous two-way flow of information between the computational model and the physical system. Data streams from the physical system are assimilated into the computational model to reduce uncertainties and improve model predictions, which in turn are used as a basis for controlling the physical system, optimizing data acquisition, and providing decision support. The DT must execute rapidly enough to support decisions and controls in time scales relevant to the physical system, and must manage and quantify uncertainties across its lifecycle. Often this necessitates DT-aware reduced order or surrogate models, i.e. those that map uncertain parameters, decision variables, and current states to quantities of interest. This minisymposium will focus on mathematical, statistical, and computational foundations underlying DTs, in particular addressing challenges in (1) data assimilation and statistical inverse problems, (2) optimal control and decision making, (3) optimal experimental design, and (4) model reduction and surrogates, all in the context of DTs of complex systems.
Keywords:
Bayesian inverse problems, Decision-making under uncertainty, Model reduction, Optimal control, Optimal experimental design, Surrogates
The digital twin paradigm represents a significant shift in engineering and science, establishing high-fidelity virtual models that operate in real-time. However, the immense computational cost of complex simulations and the uncertainties inherent in the connection between physical systems and their virtual models continue to pose significant challenges.
This minisymposium addresses those challenges by focusing on recent advances in key enabling technologies. The scope includes the incorporation of AI-powered methods for integrating operational data and managing uncertainties, with topics including data-driven approaches and physics-informed neural networks. Concurrently, the minisymposium will cover advanced large-scale computational techniques for the efficient and scalable simulation of complex systems. These techniques include, but are not limited to, recent developments in domain decomposition methods, reduced-order models, and advanced interface mechanics for multi-physics analysis.
The primary objective is to provide a collaborative platform for researchers from academia and industry to discuss both fundamental and applied aspects. This forum will facilitate the dissemination of recent advancements and foster new collaborations for creating the next generation of digital twins.
Keywords:
AI-Powered Methods, Data-Driven Methods, Large-Scale Computation, Digital Twin
Many of todayâs societal needs such as mitigation of natural hazards, energy and environmental sustainability, development of resilient civil infrastructure, and accessing natural resources require studying the physical properties and processes of the Earth and geophysical systems across all scales from both scientific and technological perspectives. Moreover, natural hazards such as landslides, earthquakes, floods, and wildfires pose major threats to human lives and the critical infrastructure. Machine learning techniques and digital twin technology offer great potential for advanced management of geosystems and natural hazards.
This mini-symposium aims to provide a forum to discuss recent advances in applications of Artificial Intelligence and Digital Twins to enhance monitoring and assessment of geosystems and infrastructure, as well as prediction and mitigation of natural hazards. The topics of interest include, but are not limited to:
- Data-driven and physics-informed modeling of geosystems and natural hazards across scales
- Advanced numerical modeling techniques for geosystems and natural hazards
- Advances in sensing and monitoring techniques
- Geohazards prediction and assessment
- Data analytics in geosystems application
- High-performance computing for digital twins
- Reduced order modeling
- Inverse modeling techniques
- Probabilistic forecasting of natural hazards
- Real-time assessment and monitoring of structures and infrastructure
- Infrastructure maintenance and retrofitting
Keywords:
Artificial intelligence, Computational Geoscience, Digital twin, disaster, Geohazards, geomechanics, Geophysics, Geotechnical engineering, Machine Learning, Natural Hazard
This minisymposium invites contributions that explore how advances in computational engineering, robotics, and mechatronics can address some of the most pressing global challenges. As societies face growing pressures from climate change, energy transitions, population growth, and technological disruption, engineering innovation is increasingly expected to deliver sustainable, scalable, and equitable solutions across multiple sectors.
We welcome original work focused on the development and application of computational methods, including numerical simulations, multiphysics modelling, materials simulation, and data-driven approaches, to tackle real-world engineering problems. Particular emphasis is placed on high-integrity systems, smart actuators, reconfigurable robotic platforms, smart materials, autonomous robotics, and resilient infrastructure, which serve as critical enablers of societal transformation. Advances in integrated hardware-virtual environments, model-based design, and intelligent control systems are particularly encouraged.
Cross-disciplinary perspectives are strongly encouraged. We welcome studies at the intersection of mechanical, electrical, and control systems, particularly those integrating AI, digital twins, or advanced materials to support adaptive and intelligent decision-making. Submissions addressing verification, validation, and certification challenges in safety-critical systems are also of interest, especially those bridging theoretical development with experimental implementation.
Researchers working in aerospace, automotive, energy, healthcare, and sustainable urban systems are particularly invited to contribute. This minisymposium aims to provide a platform for forward-looking, computation-driven research that addresses engineering challenges and supports transformative societal outcomes through novel, scalable, and digitally integrated solutions.
Keywords:
Computational Mechanics, Digital twin, multiscale and multiphysics modelling, Sustainability
Digital twins have emerged as transformative tools that allow the coupling of computational and physical assets. Their role can be of critical importance in both design and optimisation, as well as real-time decision making and control. A digital twin is typically enabled by two core components, data and simulators, which need to accommodate its requirement for efficiency, accuracy and in various cases adaptability.
The simulators can be physics-based, data-driven or hybrids. They can be fully coupled to the physical system, adaptive, or updatable. In any case, they need to be robust and reliable since they are critical for the successful development, application and deployment of a digital twin in an industrial setting.
This session aims to gather contributions highlighting the use of physics-, data-, or hybrid-based simulators and their impact on digital twin applications. Contributors are invited to discuss topics ranging from, but not limited to physics-driven modelling, reduced order modelling, scientific machine learning, model adaptivity and updating, predictive maintenance, optimal sensing, monitoring and optimal control in energy systems.
Keywords:
Data-driven modeling, Digital twins, Hybrid Analysis and Modeling, Industrial Applications, Physics-based modeling
800
Fluid Dynamics and Transport Phenomena
This minisymposium is devoted to recent developments in numerical methods for the approximation of hyperbolic partial differential equations (PDEs). A key difficulty in the study of such problems lies in the formation of discontinuities and the presence of multiple scales, requiring the use of advanced numerical schemes [1,2].
Hyperbolic PDEs and their numerical treatment remain an active area of research in applied mathematics due to their theoretical challenges and wide range of applications. Particular relevance is found in fluid dynamics, where hyperbolic models arise in compressible flows, shallow water equations, and multiphase systems, among others. This session brings together contributions addressing these issues from different methodological perspectives.
The session highlights recent progress in the construction and mathematical analysis of numerical schemes specifically designed for hyperbolic systems. Special emphasis is placed on high-order methods, structure-preserving techniques, and relaxation-based formulations, among others. In addition to the development of numerical schemes, the session is open to contributions on theoretical aspects such as consistency, stability, and convergence, as well as numerical tests that assess their performance. Both real-world applications of hyperbolic systemsâespecially in fluid dynamicsâand more theoretical research are welcome.
REFERENCES
[1] E.F. Toro, Riemann Solvers and Numerical Methods for Fluid Dynamics: A Practical Introduction, 3rd Edition, Springer, 2009.
[2] R.J. LeVeque, Numerical Methods for Conservation Laws, 2nd Edition, Birkhäuser, 1992.
Keywords:
fluid dynamics, hyperbolic systems, partial differential equations, numerical methods
The study of multiphase flow phenomena has been a major research focus in recent decades due to their importance in various natural and technical processes and their inherent complexity. Microfluidic multiphase flows are critical in numerous engineering applications, such as Lab-On-a-Chip devices, inkjet printing, fuel cells, microreactors, oil-gas/water transport, and COâ sequestration in porous media. In these often geometrically complex systems, interfacial forces âparticularly surface tensionâ dominate fluid behavior. Additionally, liquid-gas systems form a contact line with solid surfaces. As the three-phase system strives to reach its equilibrium configuration, it exhibits dynamic behavior determined by its physicochemical properties.
Broadly speaking, theoretical models fall into three main categories: those focused on the microscopic scale, particularly molecular dynamics, mesoscopic descriptions like phase field models, and those developed at the continuum level. Combining the outcomes of these three categories can improve modeling and physical understanding. Recently, microscopic (molecular dynamics) simulations have provided a practical way to deepen our understanding of fundamental physics. On the other hand, strong effort has been put into adapting continuum-based modeling high-fidelity Computational Fluid Dynamics methods to the application at hand.
Some recent works strive to enrich the existing modeling approaches with the use of data-driven methods. Development of the Machine Learning approaches has opened new research avenues for tackling challenges such as curvature approximation, model discovery for contact line dynamics, and optimization of complex microfluidic systems. Data-driven methods can also improve the efficiency of the numerical modeling of multiscale phenomena.
The mini-symposium will bring together researchers working on the different types of micro/meso and macroscopic modeling and simulation of multiphase flows in microfluidic applications as well as the emerging application of data-driven methodologies in multiphase microfluidics. We will discuss the latest research on how data-driven methodologies can improve both microfluidic system designs and the understanding of complex multiphase flow dynamics.
Keywords:
Droplet dynamics, Interfacial flow, machine learning, Wetting dynamics
The modeling of multiphase flows with liquid-vapor transition such as cavitating flows, sprays, evaporating and boiling flows has applications in numerous fields of science, for instance meteorology, geophysics, and many sectors of engineering such as aerospace and nuclear technologies. These flows are characterized by complex multiscale phenomena involving the dynamic formation of phase interfaces and inter-phase transfers. Important advances have been made in computational models for the simulation of multiphase flows with phase change, based on various mathematical and physical models and different numerical approaches, e.g. [1,2,3]. Yet there are many open challenges towards the accurate description of the hydro-thermodynamics of these flows in realistic configurations. Some difficulties concern for instance the modeling of non-equilibrium phenomena and metastability in heat and mass transfer processes, and the description of nucleation mechanisms. In some problems it is crucial to take into account the multiscale nature of the process, for example in the description of the atomization of liquid jets into droplets or of surface wettability effects in boiling and evaporating flows. Further difficulties arise in problems involving complex multicomponent flows described by realistic equations of state. The rising complexity of newly developed mathematical and physical models entails numerous challenges in the design of accurate and efficient numerical algorithms and the development of computational tools applicable to complex geometries and to a large range of Mach number regimes. The aim of this minisymposium is to bring together scientists working on computational models for multiphase flows with liquid-vapor phase change to share and exchange ideas, discuss challenges and innovative methods in the field. The minisymposium will be open to a broad spectrum of modeling techniques and numerical approaches.
Keywords:
boiling, cavitation, Computational Fluid Dynamics
This minisymposium covers applications of state-of-the-art computational fluid dynamics (CFD) simulations for multi-physics problems in science and engineering. Topics of interest include, but are not limited to, reactive flows, multiphase/multiscale flows, Newtonian/non-Newtonian fluid flows, and turbulent flows. It serves as a forum to exchange ideas for future developments in the field. Emphasis is placed on novel computational methods, advanced simulations, and innovative uses of deep machine learning. Recent advances in data-driven analytics and AI are expanding the boundaries of traditional disciplines such as fluid science and engineering. CFD integrated with machine learning offers promising approaches to address complex flow problems across multiple spatial and temporal scales, while enabling efficient surrogate modeling to reduce computational costs. Contributions from students and early-career researchers are especially welcome, particularly those tackling unsolved or insufficiently addressed problems with creative strategies.
Keywords:
Multi-physics Problem, Turbulence, Fluid Dynamics, machine learning
Meshless Particle Methods are a relatively new approach in the field of computational fluid dynamics (CFD), which has attracted significant attention over the last two decades [1]. Their popularity is largely due to their ability to circumvent the mesh tangling problem, which provides some unique advantages in modeling multi-physics flows and associated transport phenomena, albeit at the cost of computational power. In this class of methods, the grid is completely abandoned, and the discrete viscous flow is represented by replacing the conventional mesh with a finite number of particles that carry the fluid's characteristic properties, such as density, velocity, pressure, and other hydrodynamic properties in a Lagrangian manner. These particles essentially substitute the nodes in conventional mesh-based techniques. The fluid system's evolution is governed by interactions between these particles. Some examples are, but not limited to [2]: Smoothed particle hydrodynamics, Dissipative particle dynamics, Discrete Element Method, Reproducing kernel particle method, Moving particle semi-implicit, Particle-in-cell, Moving particle finite element method, Cracking particles method, Immersed particle method, Lattice Boltzmann Method, etc.
On the other hand, as these techniques are still developing CFD methods, it is crucial to identify their advantages and limitations in modeling realistic multi-physics flow problems of real-life applications and of industrial interest [3]. Toward this end, this session aims at presenting motivations, current state, and challenges behind solving the relevant partial differential equations utilizing these methods, advancing their state-of-the-art application for addressing industrial problems, as well as benchmarking and deriving general conclusions regarding their benefits and limitations and stressing the remaining challenges to make them hand-on computational tools. Regarding the methodologies, we also look for novel developments and the extension of current numerical models/ algorithms that make the simulations of complex processes using such methods more reliable and/or faster. Some examples include novel methodologies for high-performance computing, neural network techniques, and Lagrangian tracking, among others. Additionally, step-by-step benchmarking as well as developing the capabilities of Meshless Particle Methods for new multi-phase, multi-scale, and multi-physics problems are of interest at the current symposium.
Keywords:
CFD, Meshfree Methods, Multi-physics, Multi-Scale, multiphase flows, Particle Methods
The methods emerging from the field of computational topology are gaining increasing relevance in various aspects of materials science. Examples include porous media flow, flow through membranes, network-based transport, dense suspensions, and granular matter, among others. Topology-based methods correlate structure and performance, help explain and understand experimental and computational results, and propose new materials with superior properties for the applications considered.
We envision an interdisciplinary mini-symposium involving materials scientists and engineers, mathematicians focusing on various applied aspects of computational topology, computational scientists developing new numerical methods, data scientists analyzing various aspects of data from experiments and simulations, and machine learning and AI developers, who may find the data emerging from computational topology-based approaches useful for developing novel methods.
Keywords:
computational topology, fluid dynamics, transport phenomena
The numerical simulation of time-dependent compressible flow fields presents significant challenges due to the presence of dynamic flow structures and evolving geometries. Accurate and efficient resolution of such phenomena requires advanced spatial and temporal discretization strategies. These flows often involve large boundary deformations, moving solid bodies, or dynamic internal flow features that demand localized grid refinement or motion. Moreover, compressible flows frequently exhibit multiple time scales, e.g. in low-Mach, reacting or multi-phase flows, or even viscous and inter-phase drag forces.
In body-fitted methods, mesh deformation or adaptation is crucial to accurately and efficiently resolve complex flow patterns. The Arbitrarily LagrangianâEulerian (ALE) formulation has proven to be an effective strategy to handle relative motion between the mesh and the flow, since, in principle, it allows a complete separation of the mesh velocity from the material velocity. However, different characteristic time scales make the time integration challenging. Fully explicit schemes are often limited by stringent stability constraints requiring an excessively small time step, while fully implicit schemes can be computationally expensive and might lead to global nonlinear systems to be solved. Semi-implicit or IMEX (Implicit-Explicit) time integration methods provide a promising alternative by treating fast and slow dynamics differently.
This minisymposium focuses on body-fitted finite-volume and discontinuous Galerkin methods that specifically address these challenges. Topics include Lagrangian and ALE schemes, with their high-order extension, over moving, adaptive, or overset grids, and time integration strategies such as IMEX or semi-implicit formulations. Contributions exploring key numerical features such as conservation, stability, moving boundary conditions, computational efficiency, as well as practical implementation aspects, are welcome. Target applications may involve single- and multi-phase flows, supercritical and non-ideal fluids, multispecies and reacting flows.
Keywords:
Body-fitted mesh, Compressible Fluid Dynamics, Moving grids, Semi-implicit time integration, Arbitrarily-LagrangianâEulerian (ALE) scheme
Many problems in geophysical and environmental fluid mechanics exhibit a wide range of scales and must be solved over large, geometrically complex spatial domains, often for long periods of time. Computational methods for these types of problems have matured considerably in recent years. This minisymposium will examine the latest developments in solving geophysical and environmental fluid mechanics problems. Topics of interest include:
⢠Model development and application.
⢠Coupling of flow and transport processes and models.
⢠High-performance computing and parallelization strategies.
⢠Error analysis, verification and validation.
⢠Unstructured mesh generation algorithms and criteria.
⢠Fluid-structure interactions.
⢠Novel discretization methods.
Keywords:
Numerical Methods for PDEs, Scientific computing, shallow water equations
The aim of this minisymposium is to discuss and to share recent advances regarding mathematical modelling, numerical discretization and solution techniques, computational algorithms and software optimization, validation and benchmarking configurations and high performance computing simulations in the case of real-life applications for the efficient treatment of viscoelastic flow problems. Typical topics may cover new ideas regarding numerical techniques for a broad range of Weissenberg numbers, but also new approaches w.r.t. differential and integral models to describe the complex rheological behaviour, up to future high performance computing environments which will be in the exascale range and which will include massively parallel, heterogeneous architectures together with specific accelerator hardware (GPUs) including reduced arithmetic precision, are in the focus of this minisymposium. The minisymposium will mainly concentrate on methods and their foundations, and will highlight the interplay of these aspects with computational and algorithmic tools and particularly their realization in software to address current and future challenges in the field.
Keywords:
Computational Rheology, High Performance Computing, Numerical Simulation, Viscoelastic Flow Problems
Computational Fluid Dynamics (CFD) has become an essential tool for analyzing and designing complex systems in areas such as renewable energy, transportation, and biomedicine. This progress has been enabled by advances in high-performance computing, structure-preserving numerical methods, and physics-based turbulence models. As the field is transitioning from RANS to eddy-resolving techniques, the demand for predictive simulations with fast turnaround times highlights the need to balance numerical robustness, physical fidelity, and computational efficiency. This MS, aligned with the activities of the ERCOFTAC Special Interest Group SIG55, aims to bring together researchers and end-users working in the broad area of scale-resolving CFD simulations, with particular emphasis on:
-Structure-preserving discretizations with secondary conservation properties (e.g., kinetic energy, enstrophy, helicity, entropy) and physics-inspired LES models.
-Robust and flexible algorithms for unstructured grid, conservative immersed-boundary methods, dissipation-free schemes and stable multiphysics coupling.
-Improvement of algorithmic efficiency, with critical appraisal of low- vs. high-order schemes, low-dissipative time-integration methods and cost-effective approaches.
This MS will contribute to defining a shared vision and laying the groundwork for a next-generation roadmap in CFD research and development. As a concrete outcome, it is expected to define a representative set of test cases for the numerical assessment of scale-resolving techniques.
Keywords:
Computational Efficiency, Physics-compatible schemes, High-fidelity CFD
Multiphase flows are found in many mechanical, process, chemical, maritime, civil, and biomedical applications. Their features include transport phenomena in bubbles, droplets, films, and sprays, potentially in a reacting environment; turbulence modulation and drag reduction in bubbly flows, fluid-structure interactions in free surfaces, or cavitation in rotating machinery, among others [1]. The assessment of such critical features requires describing fundamental physical phenomena including bubble growth, detachment, dispersion, deformation, coalescence, and collapse; film instability and breakage; jet atomization, phase change, Marangoni convection, or electro-wetting, among others. All these physical phenomena share the core role of the two fluids interface in their underlying mechanisms.
To gain a deeper understanding in multiphase flow physics, numerical simulation is an invaluable tool. Nonetheless, it requires resolving a moving, deformable, interface; treating potentially huge differences in physical properties, and including interfacial phenomena itself, like surfactants surface diffusion. Consequently, the numerical simulation of multiphase flows is a rich field of research with several open questions, as proven by the same coexistence of several numerical techniques.
The numerical treatment of multiphase flow physics is then challenged to develop better numerical schemes [2] that improve conservation (mass, momentum, energy), surface tension, interface reconstruction (surface area, normal vector, curvature), interface transport; and computational techniques for time-stepping, variable coefficient Poisson equation, Adaptive Mesh Refinement, or numerical instabilities, among others.
In this mini-symposium, we want to gather developers of interface-resolved multiphase flows simulations (e.g.: F-T, VOF, (C)LS, PF, etc) and practitioners of multiphase flows simulations, from different application areas, to exchange their experiences in solving the numerical challenges highlighted above and the common challenges experienced in industry.
REFERENCES
[1] D. Lohse, Bubble puzzles: From fundamentals to applications. Physical Review Fluids. 3 (2018)
[2] S. Popinet, Numerical Models of Surface Tension. Annual Review of Fluid Mechanics. 50 (2018)
Keywords:
Adaptive Mesh Refinement, Conservation, Conservative Level Set, Front Tracking, Immersed Boundary Method, Interface Capturing, Interface Tracking, Phase Field, Surface Reconstruction, Surface Tension, Volume Of Fluid, Multiphase Flow
The presence of solid walls in many flow systems strongly influences the flow, heat and mass transfer in the adjacent fluid layer, especially when phase changes and chemical reactions take place. The development of relevant computational methods (including both theoretical models and numerical algorithmic approaches) and experimental techniques for model validation is driven primarily by the demands of engineering practice and advances in various application systems, which require a better understanding of the underlying multiscale and multiphysics near-wall processes.
High-temperature materials synthesis and processing, engine heat transfer and combustion, chemical engineering (chemical vapour deposition and infiltration, catalytic processes, etc.) and boundary layer flames with relevance for fire safety are just a few representative examples. Accordingly, processes such as material deposition, film growth and etching, surface reactions and their coupling with chemically reacting flows, wall-flame interaction - all together with the presence of conjugated processes of heat and mass transfer - need to be addressed.
This mini symposium aims to highlight some of the achievements made in this area. Since both the modelling approaches and the near wall validation measurements are challenging, the mini-symposium will help by providing the state of the art with respect to (a) reliable modelling approaches for the simulation of multiscale and multiphysics near wall processes in combustion systems, (b) appropriate experimental data required for comprehensive model validation, and (c) validation/uncertainty quantification issues for LES and relevant scale-resolving turbulence models in the near-wall region. The mini symposium will provide an opportunity for participants to present their latest results, and also to develop and initiate new collaborations in this field.
Keywords:
Computational Fluid Dynamics, Experimental investigation and validation, fluid mechanics, Modeling and Simulation of Complex Engineering Systems, Multiphysics;
This minisymposium focuses on recent approaches in constitutive modelling of fluids with special emphasis on non-Newtonian flows which occur, for instance, in biological systems (blood) or lubricants (synthetic or natural). It is well-known that linear relations between the velocity gradient and the stress fail to accurately describe the material behaviour in such cases. Consequently, novel approaches are required for the constitutive modelling of fluids.
A particular emphasis of this minisymposium is the advancement of scale-bridging material models for fluids as the vision is to overcome the burden of fine numerical discretizations which are still required today. To this end, temporal and/or spatial homogenization techniques might be a promising strategy as they have already proven beneficial for the modelling of solids.
Possible topics for contributions of the minisymposium are:
- Homogenization approaches
- Modelling based on internal variables
- Surrogate modelling
- AI-based strategies
- Adaptive parameter adjustment
- Experimental investigations
- Mathematical analysis
Keywords:
Multiscale Modeling, Non-Newtonian Fluids, Numerical Experiments
In this mini-symposium, we invite world-leading researchers to present and discuss recent advances in the numerical modelling of granular and multi-phase flows. The focus will encompass a broad spectrum of numerical techniques, including but not limited to the Discrete Element Method (DEM), Smoothed Particle Hydrodynamics (SPH), Moving Particle Simulation (MPS), Volume-of-Fluid (VOF), and the Lattice Boltzmann Method (LBM). We also encourage contributions that integrate data-driven approaches, such as surrogate modelling, reduced-order modelling, and machine learning frameworks, to enhance, accelerate, or augment these numerical simulations. The symposium aims to foster interdisciplinary discussions that bridge fundamental modelling, computational techniques, and applications across science and engineering, thereby highlighting both methodological innovations and practical impacts.
Keywords:
coarse graining, Discrete Element Method, Free surface fluid flow, Lattice Boltzmann Method, Multiphase Flow, Reduced Order Modelling (ROM), Smoothed Particle Hydrodynamics, Volume Of Fluid
The ability to control flow enables a wide range of technical applications, from drag reduction with the potential to achieve significant reductions of CO2 emissions to the stabilization or destabilization of existing flow topologies. These approaches can be particularly beneficial in the air mobility sector, where manipulating the flow around air vehicles can enhance both environmental sustainability and economic efficiency. Passive flow control methods are widely used; however, while they perform efficiently and reliably at design points, they cannot adapt to changing flow conditions. Active flow control (AFC) devices may bridge this gap offering adaptability to specific flow applications while minimizing control efforts.
In this mini-symposium, special attention is given to design strategies and flow analysis procedures aimed at identifying the best flow control device types, the best locations and, finally, the best operational modes of flow control devices tailored to specific flow applications. We are particularly interested in the theoretical, analytical, and numerical frameworks supporting these objectives, as well as the flow physics that govern the operation of specific AFC devices. The impact of AFC on associated flow phenomena and the change in topology of the targeted flow are also of interest.
The aim of the mini-symposium is to bring together researchers, scientists and professional practitioners working in the field of AFC, with a focus on technologies that enable objectives such as reducing drag, increasing lift, or even improving flight control, among others. The exchange of ideas, new concepts and innovative approaches is desired. We invite contributions from all areas of analytical, experimental and computational fluid mechanics for AFC including reduced-order modelling, stability analysis and decomposition techniques and energy-based methods. Further topics (experimental, analytical, numerical and other related areas) may include AI-based or AI-supported AFC strategies, device design and operation.
Keywords:
Active Flow Control, global sensitivity analysis, Mode Decomposition, Reasoning
Cavitation and microbubble dynamics are important in a wide range of fluid and solid dynamics contexts including naval hydrodynamics, focused ultrasound therapies, and soft material rheometry. The dynamics involve inertial volume oscillations and/or non-spherical bubble surface perturbations of single or population of bubbles in liquid or viscoelastic surroundings. For example, ultrasound therapies can be used to induce inertial cavitation in soft tissues and non-spherical bubble surface perturbations to induce microjets for drug delivery applications. During inertial collapse and upon reaching minimum volume, the kinetic energy is concentrated into the bubble. The energy is then released as a radial shock that propagates into the surroundings. The implosion may also lead to local damage (e.g., cavitation erosion on surfaces, soft tissue ablation, and encapsulating shells buckling for diagnostic ultrasound).
Open challenges remain to accurately predict bubble dynamics and surrounding material behavior [1]. Chief among the challenges is converging modeling and experimental efforts given multiscale modeling difficulties and optical and time resolution limitations [2], respectively. The aim of this minisymposium is to bring together modeling and experimental bubble dynamics scholars to share the latest developments and ideas. Submissions from experiments, modeling, and theory across length scales with applications in fluids and solid mechanics, biomedicine, and energy sciences are welcomed.
Keywords:
Cavitation, Fluid and Solid Mechanics, Bubble dynamics
Kinetic models have emerged as a comprehensive and versatile multiscale modeling paradigm in many applications in science and engineering to simulate transport in and out of equilibrium, e.g. rarefied-gas dynamics, radiative transfer, plasma physics, dispersed-particle flows, galactic dynamics, etc. Kinetic theories typically lead to high-dimensional equations, in which interactions manifest themselves as complicated non-linear integro-differential operators. The high dimensional setting of kinetic problems implies that the corresponding computational complexity is prohibitive, particularly when attempting to simultaneously resolve both equilibrium and nonequilibrium flow phenomena.
Recent advances in the numerical approximations for kinetic theories exploit fundamental structural properties of kinetic equations related to trends toward equilibrium and corresponding hydrodynamic (continuum) limits. Exploiting such properties enables the derivation of eďŹicient and robust methods that are capable of adaptive resolution of equilibrium and non-equilibrium effects. However, computationally scaling such methods to complex geometries remains an outstanding challenge.
The aim of this mini-symposium is to assemble researchers in the area of multiscale modeling, analysis and simulation of kinetic theories, in order to discuss recent developments, to explore open issues, and to foster cross-fertilization. The envisaged range of topics spans (but is not limited to):
⢠Moment theories for kinetic equations
⢠Hydrodynamics and continuum limits
⢠Scale-bridging and adaptive methods
⢠Structure-preserving discretization techniques
REFERENCES
[1] Sone, Yoshio, ed. Molecular gas dynamics: theory, techniques, and applications. Boston, MA: Birkhäuser Boston, 2007.
[2] Cercignani, Carlo, and Carlo Cercignani. The boltzmann equation. Springer New York, 1988.
Keywords:
Direct Simulation Monte Carlo, Moment Methods, Multiscale Fluid Dynamics, Multiscale Methods, Plasma Physics, Radiative Transfer, Rarefied Gases, Computational Kinetic Theory, Hyperbolic Systems
The mini-symposium targets researchers developing or applying moment-based continuum and kinetic models for non-equilibrium phenomena in gases, liquids, plasmas, and particulate systems. Contributions are welcome on theoretical advances (regularised moment closures, hyperbolicity, entropy consistency), numerical schemes (high-order DG/FV/FE, method of fundamental solutions, time integrators), data-assisted or machine-learned closures, and validation against DSMC, Lattice Boltzmann, or experiment. Application areas include micro/nano-scale gas dynamics, MEMS, vacuum technology, porous-media transport, hypersonic & rarefied aerothermodynamics, plasma processing, and population-balance or radiative-transfer problems. Talks that present high-fidelity benchmarking, reproducible workflows, or that quantify model uncertainty are especially encouraged, as well as recent software advances across these areas (e.g., open-source releases, HPC/GPU implementations, verification & benchmarking suites, and reproducibility best practices).
Keywords:
Grad Closure, Moment Methods, Multiscale Modelling, Non-Equilibrium Transport, Rarefied Gases
900
Fluid-structure Interaction
This symposium will bring together researchers from the engineering community to discuss their work on computational fluid dynamics (CFD) and fluid-structure interaction (CFSI). The symposium will cover both computational methods and engineering applications. Topics will include but not limited to theoretical developments, novel computational frameworks, new discretization methods, high-order approaches, moving-mesh methods such as the arbitrary Lagrangian-Eulerian (ALE) and space-time (ST) methods, isogeometric analysis (IGA), FSI coupling strategies, Eulerian and ALE hydrocodes, high-performance computing, and applications to complex problems in engineering, science, and medicine. Recent trends in machine learning techniques for CFD and CFSI will also be of interest in this symposium. The symposium will provide a venue for researchers from both academia and industry to discuss the most recent advances and emerging research directions in this field.
Keywords:
Computational Fluid Dynamics, Computational Fluid-Structure Interaction, High-Performance Computing, Multiphysics problems
The objective of this Mini Symposium is to discuss progress and recent advancements in the numerical computation of fluid-structure-interaction problems, with an emphasis on new innovative formulations, (numerical) analysis, and methods and algorithms leading to faster, more accurate predictions and improved software design. The envisaged range of applications spans (but is not limited to) aero-elasticity, hydro-elasticity, biomechanical FSI and noise/structural acoustics. In particular, we welcome contributions in the vanguard of:
⢠error estimation;
⢠adaptive methods;
⢠immersed and unfitted methods;
⢠multiscale models;
⢠reduced order models and methods;
⢠artificial intelligence and machine learning;
⢠novel iterative solution techniques;
⢠shape optimization and inverse methods;
⢠software engineering.
In addition, this Mini Symposium is intended as a platform for other state-of-the-art developments in FSI, such as those pertaining to FSI with auxiliary-field interactions, e.g. FSI problems with (massive) self contact, FSI problems with fracture (e.g. hydraulic fracturing, blast-induced FSI, etc.), elasto-capillary FSI, etc.
Keywords:
adaptive methods, auxiliary-field interactions, error estimation, immersed methods, iterative solution methods, multiscale models, reduced-order modeling, Fluid-structure interaction
The proposed session brings together the latest research and advances in fluid-structure interaction (FSI), uniting complementary perspectives from both detailed local-scale modeling and large-scale engineering applications. The aim is to gather a diverse community to exchange knowledge on a broad spectrum of approaches â from refined methods that explicitly represent fluid-structure interfaces (e.g., Arbitrary Lagrangian-Eulerian formulations) to more robust strategies such as immersed boundaries or porous media models, as well as hybrid couplings involving particles, finite elements, finite volumes, or Lattice Boltzmann methods.
The scope encompasses a wide range of physical regimes â incompressible turbulent flows, multiphase flows, and compressible flows â interacting with various structural models, including cases with strong nonlinearities such as contact, impacts, and fluid-elastic instabilities. The session will address applications from multiple industrial fields, including power plants, biological systems, and offshore engineering.
A distinctive focus will be placed on multi-scale modeling and upscaling strategies to deliver certified computational mechanics tools with quantified confidence levels for industrial end-users. This includes rigorous uncertainty quantification and propagation from the local to the industrial scale, as well as validation at every relevant scale through combined experimental and numerical studies.
Key topics of interest include:
- Efficient and robust partitioned coupling methods, extended to extreme-scale computing
- Local-scale advanced modeling and its connection to homogenized or porous representations at large scale
- Strategies for managing and validating nonlinear dynamic responses of immersed structures
- Effects of turbulence, compressibility, wave propagation, and added mass/damping coefficients
- Cross-community knowledge sharing on state-of-the-art solutions for complex FSI problems
By bridging fine-scale analysis with industrial-scale applications, the session seeks to provide a platform where methodological diversity meets engineering reliability, fostering collaboration across theoretical, computational, and experimental domains.
Keywords:
Modelling, Numerical methods, Simulation and experiments, Fluid-structure interaction, validation and certification
Recent developments in computational wind engineering, numerical methods, and high-performance computing have matured to model and compute wind effects on structures accurately and efficiently via fully coupled wind-structure interaction. Advancements in high-fidelity wind simulations have resulted in more accurate modelling of complex aero-dynamic phenomena. These accurate simulations play a critical role in the performance and safety evaluation of buildings, bridges, flexible structures, membranes, etc.
Complementing these developments, new data-driven techniques - including machine learning, surrogate modelling, and digital twins - are also interesting for how wind loads and structural response to wind loads are evaluated and monitored. Combining high-fidelity wind simulations with wind tunnel and real-world measurement data offers reliable and fast structural response prediction. The intersection of high-fidelity simulation with data-driven methods provides a chance for efficient preliminary design, structural design optimization, uncertainty quantification, real-time structural assessment, etc.
In this context, this mini symposium aims to bring together experts from academia and industry to discuss recent progress in high-fidelity simulations, wind structure interaction methods, validation studies, and applications. Topics include (but are not limited to) high-fidelity wind simulations, structural response prediction, reduced-order modelling, real-time response prediction of structures, digital twins under wind, hybrid simulation frameworks, data-driven FSI, and case studies on tall buildings, long-span bridges, wind turbine towers, membrane structures, etc. Interdisciplinary submissions from both academia and industry are encouraged.
Keywords:
computational wind engineering, data-driven methods, digital twin, fluid-structure interaction, high-fidelity simulation
1000
Fracture, Damage and Failure Mechanics
This is a long-standing interdisciplinary Mini-symposium, held at WCCM 8 (Venice, Italy, 2008), WCCM 9 (Sydney, Australia, 2010), WCCM 10 (Sao Paulo, Brazil, 2012), WCCM 11 (Barcelona, Spain, 2014), WCCM 12 (Seoul, South Korea, 2016), WCCM 13 (New York, USA, 2018), WCCM 14 (Paris, France/virtual edition, 2021), WCCM 15 (Yokohama, Japan/ virtual edition) and WCCM 17 (Vancouver, Canada). Its aim is to bring together specialists in mechanics and micromechanics of materials, applied mathematics, continuum mechanics, materials science, physics, biomechanics as well as mechanical, automotive, aerospace and medical engineering to discuss the latest developments and trends in computational analysis of relationships between the microstructural features of advanced engineering and natural materials and their local and global behaviours as well as their effects on performance of components and structures.
The topics of the Mini-symposium include, but are not limited to, the following:
⢠computational mechanics of advanced materials and structures;
⢠effect of microstructure on properties and performance of advanced materials;
⢠prediction of deformational behaviour and life-in-service of structures and components made of advanced materials;
⢠computational models of biological and biomedical materials;
⢠computational methods for analysis of modern composite and nanocomposite materials;
⢠mechanics of composite materials with relaxation and phase transitions;
⢠simulation of failure mechanisms and damage accumulation processes in advanced materials;
⢠computational analysis of cutting of advanced materials;
⢠numerical simulation of mechanical behaviour of materials in technological processes;
⢠optimization problems in mechanics of advanced materials and structures.
Keywords:
Biomaterials, Microstructure, Properties and Performance, Advanced Materials
The accurate and realistic modeling of inelastic behavior, damage and fracture processes of different materials is extensively discussed in the literature. Many continuum approaches have been presented and their efficiency has been demonstrated in different applications in a wide range of engineering fields. Due to intensive research activities during the last years, damage and fracture models have now reached a high level of quality based on sophisticated experiments with uniaxially and biaxially loaded specimens and corresponding numerical simulations as well as on numerical results taken from calculations on the micro-level. To be able to numerically analyze damage and fracture processes in an efficient and accurate manner, different modified and new techniques in computational mechanics as well as many robust numerical algorithms have been recently developed. A variety of continuum models and corresponding numerical aspects as well as current and future trends in computational damage and fracture mechanics will be discussed in the proposed mini-symposium.
Keywords:
continuum model, damage, fracture, numerical techniques
Damage and fracture modeling in materials and structures remains a complex challenge due to the highly nonlinear mechanisms involved. Phenomena such as strain-softening behavior, the loss of ellipticity in governing equations, and localization effects significantly complicate the search for an objective solution to the mechanical problem. Although various conventional and advanced numerical approaches have been developed, researchers continue to focus extensively on these issues to devise efficient strategies for accurately capturing the initiation and evolution of damage in traditional and innovative materials.
This minisymposium is dedicated to exploring recent developments in computational methods, with a particular emphasis on simulating damage and fracture processes. We invite contributions on topics including, but not limited to advanced discretization strategies for phase-field fracture modelling; enriched finite element techniques; virtual element methods for simulating crack propagation; eigenerosion and eigenfracture approaches; enhanced cohesive zone models; meshfree techniques for fracture analysis, peridynamics-based fracture modelling.
Keywords:
Brittle and cohesive fracture, Crack nucleation and propagation, Damage Mechanics, Numerical approaches
Cementitious materials are the foundation of modern civil engineering infrastructure. As design standards evolve and demands for longevity and sustainability grow, a deep understanding of their fracture, damage, and failure mechanisms becomes increasingly crucial. This symposium seeks to offer a focused yet comprehensive overview of current advances and future directions in the mechanics of cement-based materials. We welcome contributions on the following topics:
1. Fundamentals of Fracture Mechanics
Explore stress and strain relationships, crack initiation and propagation, Griffithâs criterion, and practical approaches for determining fracture toughness in cementitious systems.
2. Damage Mechanics and Material Heterogeneity
Clarify the distinction between damage and fracture. Examine the initiation and growth of microcracks, the impact of heterogeneity, and the role of aggregates and fibers in damage development.
3. Multi-Scale and Numerical Modeling
Bridge the gap from microstructural behavior to macroscopic performance. Highlight advanced modeling techniques including finite element (FEM) and discrete element methods (DEM), with an emphasis on capturing inhomogeneities.
4. Durability under Environmental Stressors
Investigate the long-term effects of mechanical loading and environmental factors such as freeze-thaw cycles, moisture ingress, and chemical attack. Present mitigation strategies and design implications.
5. Fiber-Reinforced Concrete (FRC)
Analyze the mechanics and performance of FRC under various loading conditions. Discuss different types of fibers (e.g., steel, polymer, glass) and their contribution to strength, ductility, and crack control.
6. Self-Healing Cementitious Systems
Introduce emerging technologies in self-healing, from autogenous recovery to biological methods. Explore their potential to enhance service life and resilience of concrete structures.
7. Non-Destructive Testing (NDT) and Monitoring
Emphasize the importance of early damage detection. Present advanced NDT techniques and their integration with predictive maintenance and structural health monitoring frameworks.
Keywords:
Computational Mechanics, Microstructure, Multiscale Modeling
The phase-field approach is a very powerful technique to model and simulate complex fracture phenomena under various loading conditions in multi-field settings and across the scales. Due to its flexibility, this methodology has gained wide interest in the engineering and applied mathematics communities, especially in the past decade.
This mini-symposium provides a forum for the discussion and exchange of ideas related to new advances and applications of the phase-field approach to fracture, corrosion, and fatigue in engineering. It welcomes contributions on phase-field modeling of fracture, including brittle, cohesive, and ductile fracture in solid and structural mechanics. Research results on basic aspects of phase-field formulations and of their numerical implementation, experimental validation as well as extensions to novel and/or more complex settings and relevant applications are all welcome.
Keywords:
Corrosion, Fatigue, fracture and failure, Multi-physics Problem, Phase-Field Modeling
In the absence of specialized treatments, local models based on classical continuum mechanics often yield inaccurate results when applied to problems involving discontinuities, such as crack initiation and propagation. On the other hand, nonlocal models offer a robust alternative to deal with such challenges in fracture mechanics. For example, peridynamics is a nonlocal theory mainly used to model elasticity and fracture, whereas the phase field model employs a nonlocal formulation specifically for modeling fracture. Both these approaches incorporate an intrinsic length scale into the formulation, which serves to prevent stress singularities at crack tips. Additionally, this characteristic length scale enables the modeling of physical phenomena that inherently involve finite-length effects. Despite their advantages, numerical methods based on nonlocal formulations tend to be less computationally efficient than their local counterparts.
This minisymposium aims to foster insightful discussions among researchers in mathematics, computational methods, and engineering applications. Its primary objective is to deepen the understanding of the role of nonlocality in fracture mechanics and related fields, emphasizing both its advantages and limitations. By bringing together experts from diverse disciplines, the minisymposium seeks to promote the exchange of ideas, stimulate interdisciplinary collaboration, and lay the groundwork for future research partnerships.
Topics of interest include (but are not limited to):
1. Mathematical and numerical analysis of nonlocal models
2. Nucleation and propagation of damage in nonlocal models (impacts, fatigue, etc.)
3. Nonlocal constitutive models of heterogeneous, anisotropic, and/or nonlinear materials
4. Peridynamics: nonlocal elasticity, wave dispersion, material stability, etc.
5. Phase field: variational formulations, stress regularization, phase transformations, etc.
6. Nonlocality in other fields: diffusion, heat transfer, fluid dynamics, etc.
7. Multiphysics nonlocal modeling: stress-corrosion, electro-thermo-mechanics, etc.
8. Multiscale modeling: local/nonlocal and nonlocal/atomistic coupling
9. Contact and interface mechanics with nonlocal models
10. Applications in science and industry: material failure, image processing, etc.
11. Numerical methods, solvers, and machine learning applications for nonlocal models
Keywords:
crack propagation, engineering applications, length scale effects, multiphysics phenomena, multiscale methods, peridynamics, phase field model, Nonlocal mechanics
Understanding and predicting material failure in natural and engineered structures subjected to multiphysics loading conditions and under the influence of multiscale effects is of paramount importance. Traditional approaches to fracture and damage modelling often fall short when faced with real-world scenarios involving coupled physical phenomena - such as thermal gradients, chemical reactions, fluid-structure interactions, electromagnetic effects, to name a few, that interact with mechanical response. Furthermore, underlying microstructure plays a crucial role. As a result, multi-physics & multi-scale driven simulations have emerged as a critical framework to capture the interplay of these diverse processes. This evolving field aims to unify mechanical fracture mechanics with other physical systems, enabling more accurate and predictive models. Recent advances in computational power, numerical methods (XFEM, VEM, peridynamics, thick level set, lipfield, localised gradient damage model, phase-field models) and data-driven techniques are fuelling the progress in this domain. Moreover, integration of multiscale modeling allows researchers to link microstructural evolution with observable macroscopic failure, enhancing material design and preventive failures. The aim of this mini-symposium is twofold: (a) bring together experts from computational mechanics, material science and applied physics to explore the frontiers of fracture and damage modeling in multiphysics and multiscale environments and (b) identify key challenges, share novel computational methods and foster collaboration towards developing robust, reliable predictive numerical frameworks capable of addressing the complexity of real-world material and structures.
Keywords:
Damage, Multiphysics, Fracture, Multiscale;
Understanding and predicting the failure and fracture behavior of engineering systems is a fundamental requirement for the safe and efficient design of modern infrastructure. In particular, components involving anchoring, fastening, and bonding applications represent some of the most challenging cases in this context. These systems are often exposed to highly nonuniform stress distributions, complex multi-axial loading paths, and varying environmental conditions. Furthermore, they frequently involve heterogeneous material compositions, such as combinations of metals, concrete, rock, polymers, and adhesives. As a result, accurately predicting their nonlinear mechanical response, including damage initiation, evolution, and ultimate failure, is an ongoing challenge in computational mechanics and material modeling.
In recent decades, significant progress has been made in the development of advanced constitutive models that are capable of describing such complex material behavior. These models incorporate sophisticated descriptions of material degradation, inelasticity, and fracture processes, often extending beyond classical continuum theories. Higher-order continuum models, such as nonlocal, micropolar, or micromorphic formulations, allow for the incorporation of intrinsic length scales that are essential to capture size effects and damage localization. Moreover, phase-field models for fracture provide a thermodynamically consistent framework for simulating crack initiation and propagation.
This minisymposium is dedicated to the modeling, simulation, and practical application of advanced constitutive models for failure and fracture. Contributions are particularly encouraged in areas including, but not limited to:
⢠Generalized continuum models (nonlocal, micropolar, micromorphic)
⢠Phase-field approaches to brittle and ductile fracture
⢠Strain gradient formulations
⢠Particle methods, e.g. Material Point Method (MPM) or Lattice Discrete Particle Model (LDPM)
⢠Discrete crack models, including embedded discontinuities and XFEM
⢠Robust, efficient, and scalable numerical methods and their implementation
⢠Applications to real-world engineering problems involving anchoring, fastening, and bonding technologies
We particularly welcome contributions that demonstrate the practical applicability of advanced models to real engineering challenges, bridging the gap between theoretical developments and industrial or structural applications.
Keywords:
Brittle and cohesive fracture, Computational Mechanics, Continuum Modelling, Methods for fracture and failure
A wide range of computational approaches have been developed over recent decades to model failure mechanisms in materials, including fracture, damage, strain localization, buckling, and erosion. Despite this progress, major challenges remainâparticularly in achieving accurate, efficient, and predictive models. These include the calibration and validation of models under complex loading, the representation of material instabilities and evolving microstructures, and the development of robust and scalable numerical schemes. Moreover, many modern applications increasingly demand modeling of material failure under multiphysics conditionsâsuch as thermal, chemical, fluid, or electric stimulationâarising in natural systems and engineered devices. These coupled environments introduce significant complexity but are essential for realistic and high-fidelity simulations.
This mini-symposium aims to bring together advances in computational modeling of damage and fracture, with a special emphasis on multiphysics scenarios. Topics of interest include, but are not limited to:
⢠Innovative techniques for representing failure (e.g., phase-field, regularized damage, XFEM, CZM, meshless methods, peridynamics, high-order continua)
⢠Multiphysics-informed constitutive and phenomenological models for fracture initiation and growth
⢠Mixed finite element and multiscale formulations for coupled physical processes
⢠Failure mechanisms under coupled fields (e.g., thermomechanical, electrochemical, fluid-driven, hygrothermal)
⢠Solver and algorithmic developments for nonlinear, multiphysics systems (e.g., monolithic/staggered schemes, preconditioning)
⢠Advances in discretization and modeling of instabilities (e.g., buckling, localization, softening)
⢠Model calibration and validation with experiments targeting fracture, shear bands, and microstructural evolution
⢠Machine learning and data-driven approaches for accelerated, physics-informed modeling of failure under multiphysics conditions
⢠Application-focused studies including hydraulic fracturing, corrosion, thermo-plasticity, soft matter rupture, and fluid-structure interactions
Keywords:
fracture, multiphysiscs, numerical methods, plasticity
Advanced materials and Structures have an increasing role in engineering, in various industrial applications [1-3]. These structures operate in severe environments, withstand complex multi-axial loading conditions. Fracture of advanced materials is also a major problem that may occur inside the structures consisting of different materials and at the interfaces between different advanced materials.
Topics of interest include, but are not limited to the computational analysis of composite structures made from advanced materials, Failure of composite structures, Comparison of computational and experimental methods in composite structures from advanced materials, Computational analysis of interface problems in composite structures, Computer aided design in composite structures, Computational study of constructions made of advanced materials.
REFERENCES
[1] A. M Nikolarakis and E. E Theotokoglou, âThermal shock problem of a threeâlayered functionally graded zirconia/titanium alloy strip based on a unified generalized thermoelasticity theoryâ, Journal of Thermal Stresses, Vol. 40, pp. 583-602 (2017).
[2] I. K. Giannopoulos, E. E. Theotokoglou and X. Zhang , âImpact damage and CAI strength of a woven CFRP material with fire retardant propertiesâ, Composites Part B: Engineering, Vol. 91, pp.8-17 (2016).
[3] O.C. Zienkiewicz and R.C. Taylor, The Finite Element Method, 4th Edition, Vol. 1, Mcgraw Hill, 1989
Keywords:
Advanced Materials, Computational Mechanics, Failure of composite structures, Fracture mechanics
Recent advances in computational mechanics have greatly improved our capacity to model, analyse, and predict damage phenomena in engineering materials and structures. These developments are driven by progress in numerical algorithms, high-performance computing, and the integration of data-driven techniques with physics-based formulations. The resulting hybrid methodologies offer enhanced accuracy, robustness, and scalability for simulating complex failure mechanisms under diverse material behaviours and loading scenarios.
The purpose of this mini symposium is to provide a forum for discussion of challenges and advances in computational modeling and prognosis of damage, placing special emphasis on efficient, robust, physically informed, and data-driven approaches. We particularly welcome contributions that bridge mechanics with innovations from applied mathematics, materials science, and data science.
The topics of interest include, but are not limited to the following:
1. Multi-scale and multi-physics approaches for damage and failure prediction
2. Physics-enhanced and data-driven machine learning techniques for fracture mechanics
3. Structural health monitoring and damage prognosis
4. Inverse problems involving fracture, such as fracture parameter identification
5. Vibration- and impact-induced fracture processes
6. Advanced discretization techniques and solution algorithms
7. Verification, validation, and benchmarking of computational methods
We encourage both theoretical developments and application-oriented studies. Of special interest are contributions relevant to domains such as civil, mechanical, aerospace, automotive, and energy engineering.
Keywords:
computational fracture mechanics, data-driven modelling, multi-physics simulations, physics enhanced machine learning
Nonlocal models offer a robust framework for the study of materials and structures where classical continuum theories prove inadequate, particularly for problems involving discontinuities, singularities, and long-range interactions [1]. Peridynamics, a reformulation of continuum mechanics based on integral equations, provides a mathematically robust framework that avoids spatial derivatives and naturally enables the modelling of crack initiation, propagation, and other discontinuous phenomena [2].
This mini-symposium addresses recent advances in both the theoretical and computational aspects of Peridynamic theory and related nonlocal methods. Emphasis is placed on their application to multi-scale and multi-physics problems involving fracture, damage evolution, and material discontinuities. The scope also includes numerical methods for nonlocal operators, convergence analysis, and alternative nonlocal theories that address similar physical and computational challenges.
Contributions focusing on validation against experimental data, comparative studies between classical and nonlocal models, and applications in fracture, fatigue, impact, or damage evolution are encouraged. The goal is to promote interdisciplinary discussion among mechanics, materials science, applied mathematics, and computational communities to advance both understanding and practical capabilities in nonlocal modelling. Contributions are also welcome on a broad range of emerging and interdisciplinary topics, including but not limited to additive manufacturing, artificial intelligence and machine learning, composite materials, fatigue, functionally graded materials, impact, reduced order modelling, structural health monitoring, topology optimization, and related areas.
REFERENCES
[1] S.A. Silling, Reformulation of elasticity theory for discontinuities and long-range forces, Journal of the Mechanics and Physics of Solids 48(1) (2000) 175â209. https://doi.org/10.1016/S0022-5096(99)00029-0
[2] E. Madenci and E. Oterkus, Peridynamic Theory and Its Applications, 4th Edition, Vol. 9781461484, Springer New York, 2014.
Keywords:
Nonlocal Models, Damage, fracture, Peridynamics
Soft materialsâincluding polymers, gels, elastomers, and biological tissuesâexhibit rich and complex behaviours that challenge conventional theories of solid mechanics. Their highly deformable and often inelastic nature gives rise to strongly non-linear, dissipative mechanisms associated with fracture, cavitation, cutting, and puncture, among others. Owing to their anisotropic, hierarchical, and adaptive microstructures, deformation, damage and failure processes are tightly coupled to micromechanical mechanisms such as crosslink rupture, fibre reorientation, and degradation. Therefore, the mechanics of soft materials frequently involves interactions across multiple scales, where microscopic properties and macrostructural effects jointly govern material response. Understanding and predicting such mechanisms is central to applications ranging from soft robotics and medical devices to tissue and foodstuff engineering. Computational approaches must capture large elastic and inelastic deformations, instabilities, and failure patterns, often with evolving discontinuities. At the same time, state-of-the-art experimental methods, such as digital image correlation, rheology, and nanoindentation enable high-resolution data that inform and validate sophisticated models.
This mini-symposium aims to bring together researchers working at the intersection of solid mechanics, materials science, bioengineering, and applied physics and mathematics to discuss current advances and identify future directions. We particularly welcome contributions that address both fundamental and applied aspects of fracture, inelasticity and failure, from either computational or experimental perspectives. Topics include, but are not limited to:
⢠Constitutive modellingâcontinuum and micromechanicalâfor soft materials, including fracture, damage, rate effects, and instabilities;
⢠Fracture- and contact-driven phenomena, including cutting and cavitation;
⢠Coupled problems and multiphysics failure in soft materials;
⢠Data-driven and hybrid modelling approaches for soft material behaviour and failure;
⢠Experimental characterization of soft materials and tissues, including fracture;
⢠Numerical methods for simulating discontinuities and complex failure patterns at large deformations, e.g., cohesive zone, phase-field, and configurational forces models;
⢠Applications, e.g., in biomedical engineering, surgery, adhesives, and food industry.
Keywords:
Computational solid mechanics, Experimental solid mechanics, Soft materials
This mini-symposium aims to highlight recent advances in computational methods for fracture mechanics and structural failure analysis. Contributions utilizing state-of-the-art numerical techniques such as FEM, X-FEM, G-FEM, S-FEM, BEM, IGA, Peridynamics, and other emerging approaches are strongly encouraged. In addition, the integration of machine learning (ML) and artificial intelligence (AI) for predictive modeling, surrogate modeling, and model calibration is welcomed. The symposium provides a platform for discussing innovative strategies to model crack initiation, crack propagation, and fracture, as well as structural failure under complex loading and environmental conditions. Applications span a wide spectrum, including aerospace, automotive, naval, nuclear, civil, and mechanical engineering. Both fundamental and applied studies are invited to foster knowledge exchange and promote the development of novel computational frameworks within the mini-symposium.
Keywords:
Numerical Simulation, Structural Analysis, Artificial intelligence, Fracture mechanics
Fracture initiation and propagation remain central challenges in engineering, with significant implications across mechanical, civil, naval, aerospace, and other disciplines. Over the years, extensive attention has been given to the modeling of brittle [1] and ductile fracture to describe the ultimate behavior of materials. More recently, increased focus has been directed toward the initiation and growth of cracks under fatigue loading conditions.
From a computational standpoint, modeling fatigue crack propagation offers the potential to significantly reduce the need for extensive laboratory testing, which is often time-consuming and complex. However, a key challenge lies in the simulation of a large number of load cycles required to accurately capture fatigue behavior.
The aim of this mini-symposium is to bridge the gap between fatigue fracture research and computational fracture mechanics. It invites contributions on numerical models capable of capturing fatigue-driven crack evolution, including but not limited to phase-field methods [2] and peridynamics. Both academic and industrial applications are welcome.
REFERENCES
[1] Griffith, Alan Arnold. "VI. The phenomena of rupture and flow in solids." Philosophical transactions of the royal society of london. Series A, containing papers of a mathematical or physical character 221.582-593 (1921): 163-198.
[2] Carrara, Pietro, et al. "A framework to model the fatigue behavior of brittle materials based on a variational phase-field approach." Computer Methods in Applied Mechanics and Engineering 361 (2020): 112731.
Keywords:
Damage modeling, Fatigue crack initiation, Fracture mechanics
The integration of Artificial Intelligence (AI) with fracture simulation is revolutionizing predictive modeling, enabling unprecedented accuracy and efficiency in computational mechanics. This mini-symposium focuses on advancing AI-driven methods for fracture simulation in engineering applications, addressing key challenges in crack propagation, material failure, and simulation optimization.
We invite contributions on topics including but not limited to:
⢠AI-enhanced fracture propagation models for accelerated and high-fidelity simulations in engineering systems.
⢠Physics-informed neural networks/operators for predictive modeling of brittle and ductile fracture in structural components.
⢠Deep learning for inverse problems in fracture mechanics and damage assessment of engineered materials.
⢠Surrogate modeling with neural networks to enable real-time fracture analysis in industrial applications.
⢠Uncertainty quantification in AI-augmented fracture prediction for reliable engineering solutions.
The mini-symposium aims to enhance AI methodologies specifically for fracture simulation in engineering contexts, bridging the gap between fundamental research and industrial applications. Researchers from computational mechanics, materials science, and applied AI are encouraged to participate, sharing insights on next-generation simulation tools that combine physics-based modeling with data-driven approaches for engineering challenges.
Keywords:
Design Optimization, Fracture Mechanics, Artificial Intelligence, engineering applications
Dynamic fracture mechanics is concerned with fracture phenomena for which the effects of material inertia cannot be ignored. The advent of ultra-high-speed imaging and advanced computational tools has enabled researchers to observe phenomena, such as crack branching instabilities, that classical theories could not adequately explain. This has led to a paradigm shift, reframing the dynamic crack tip not merely as a mathematical stress singularity but as a ânatural laboratoryâ for probing material behavior under the most extreme conditions of strain, strain rate, and energy dissipation. Consequently, the central focus has evolved from predicting failure to understanding the complex, non-equilibrium dynamics that govern it. This broader perspective has attracted interdisciplinary interest, with applications now spanning the design of advanced materials for aerospace and defense, and the modeling of geological events like earthquakes.
This symposium aims to provide a forum for the exchange of knowledge and the promotion of innovation in the dynamic fracture of materials and structures. By bringing together leading researchers from the fields of experimental mechanics, theoretical modeling, and computational simulation, this event will facilitate fruitful discussions on the pressing, unresolved questions that define the frontiers of this discipline.
Specifically, the following topics are of interest:
⢠Strain-Rate Dependency of Material Properties.
⢠The Role of Rate-Dependent Constitutive Laws, both phenomenological and physically-based.
⢠Disentangling the Multifaceted Role of Inertia. in the context of continuum damage and phase-field models, which represent a crack by âsmearingâ the damage over a finite region. This raises a fundamental question: should this region also lose its inertial mass?
⢠Terminal Crack Velocity in Materials.
⢠Crack Branching Instability.
⢠High-Fidelity Measurement and Visualization (high-speed photography, dynamic photoelasticity, digital image correlation).
⢠Uncertainty Quantification (UQ).
⢠Advanced Numerical Formulations: cohesive zone models, extended finite element methods, phase-field models, material sink theory, peridynamics, meshfree methods.
⢠Multiscale modeling.
⢠Multiphysics. specifically coupled systems relevant for industrial applications.
Keywords:
Finite Element Methods, fracture, Fracture mechanics, Phase Field
The phase-field approach has emerged as a powerful and versatile framework for modeling and simulating complex fracture phenomena under diverse loading conditions, across scales, and in multifield settings. Owing to its flexibility, it has attracted wide interest in the engineering community, particularly in the past decade. Recent advances include the study of crack nucleation, establishing the connection to strength theories, and the extension to fracture of complex materials and structures, as well as to fracture in coupled multi-field settings.
This mini-symposium provides a platform for the discussion and exchange of ideas on recent advances and applications of the phase-field approach to fracture. Contributions are welcome on topics including brittle, cohesive, and ductile fracture in solids and structures; fundamental aspects of phase-field formulations; numerical implementation strategies; experimental validation; as well as case studies and extensions to novel or more complex scenarios and applications.
Keywords:
complex materials, damage, fracture, multiphysics phenomena, nucleation, phase-field
Over the past 15 years, computational models of fracture have been increasingly utilized and pursued to describe and predict the nucleation and propagation of fracture in solids, hard and soft the same, under realistic quasi-static and dynamic loading conditions.
In this context, this Mini-Symposium aims at providing a forum to present the state of the art on computational models of fracture in hard â such as ceramics, glasses, and metals â and soft solids â such as elastomers, hydrogels, and biological tissues â as well as one to present and discuss open problems in the field. Contributions focusing on computational, theoretical, and/or experimental aspects are all welcome.
Keywords:
Crack nucleation and propagation, dynamic fracture, Soft materials
1200
Industrial Applications
For the future defossilization of the energy and transport sector, hydrogen as an energy storage and carrier will play a key role. Important energy conversion paths will be the formation of green hydrogen with electrolyzers (EL) and its conversion to electric energy in power plants and automotive applications with the help of fuel cells (FC). The most widespread current technologies are based on polymer electrolyte membranes exchanging membranes and operate below 100 °C.
It is well known that there is a risk of thermal, chemical and mechanical material degradation if non-ideal operating conditions prevail, which has a harmful impact on the FCs or ELs performance and lifetime. Membranes, electrodes, and gas diffusion layers may undergo mechanical damages like cracks and ruptures, delamination of the active layers from the membrane and gas diffusion layer, rarefaction, agglomeration, and deactivation of active sites. Sources of these degradations are manifold. Temperatures, material humidification and electric potentials that are out of the recommended range contribute to material downgrading. Switching on and off processes and transient operation, as is common for fuel cells in the automotive sector, for example, lead to an uneven distribution of reactive species and humidification, which also has an influence on the durability and performance of the materials.
Considering the manifold challenges and degradation scenarios, a wide range of numerical approaches has been applied to FCs and ELs to model the performance and material degradation, ranging from detailed 3D CFD simulation to represent to flow field accurately to 0D mathematical modelling of the different layers to allow numerical studies in real-time or faster in order to cover long physical time durations, which are usually involved if material degradation is to be analyzed.
With this minisymposium, we invite researchers to present their results in the field of numerical material modelling in fuel cells and electrolyzers and to evaluate material ageing. Beside the numerical simulations, experimental results that allow model validation are welcome. Though polymer exchange membranes are currently the most widely spread technology, investigations of other chemistries like solid oxide or alkaline may be submitted and will allow a holistic discussion of this very relevant issue of the energy and mobility transformation.
Keywords:
chemical degradation, finite elements, finite volume, mechanical degradation, PEM electrolyzer, PEM fuel cell, thermal degradation, Hydrogen
Supercomputers have provided researchers with an unprecedented level of computational power. However, "power without grip is useless": the mere availability of thousands of processors must be accompanied by significant advances in software development and HPC techniques in order to effectively tackle the most complex simulation problems in computational physics and engineering. In particular, in application domains such as industry, energy, the environment, and biomechanics, the simulation of complex and often coupled fluidâsolid phenomena remains a major challenge, demanding the full utilization of available computational resources. One distinctive feature of such large-scale simulations is their ability to generate massive volumes of data. In recent years, artificial intelligenceâespecially neural networksâhas been employed to enhance simulations through surrogate and reduced-order modeling, to reconstruct flow fields from incomplete data, and to explore high-dimensional design spaces using reinforcement learning and other advanced methods. The objective of this Mini-Symposium is to foster discussion and exchange on current challenges and future directions in HPC simulations and AI techniques, with a particular focus on real-world applications spanning a broad range of fields, including biomechanics, automotive engineering, aerospace, pharmacology, energy, and environmental science. Relevant topics include algorithms, simulation strategies, and programming techniques for complex multiphysics simulations requiring massive HPC environments. Contributions addressing related aspects such as performance optimization, robustness analysis, and integration with pre- and post-processing tools (e.g., CAD, mesh generation, visualization) are also highly welcome.
Keywords:
AI, CFD, Computational Mechanics, engineering science, High-Performance Computing
Computational Fluid Dynamics (CFD) is a cornerstone technology in ensuring the performance, reliability, and safety of critical systems in both civilian and defense sectors. Its applications span a wide range from sub- to hypersonic aerodynamics and thermodynamics of defense vehicles and re-entry capsules to simulating explosion dynamics and mitigation or fire suppression, to name only a few applications. Despite its maturity, CFD continues to face major challenges, including the accurate prediction of complex, multi-scale phenomena, the integration of multi-physics models and the need for robust, scalable solvers suitable for real-time or large-scale simulations. This mini-symposium invites contributions that address these challenges and highlight innovations in numerical methods, physical modeling, high-performance computing, and data-driven approaches within the scope of safety, security and defense related applications. The goal is to foster dialogue among researchers and practitioners working to advance CFD for safer and more effective systems.
Keywords:
Computational Fluid Dynamics, Safety Security and Defense
Isogeometric analysis (IGA) was originally introduced to achieve seamless integration of
computer-aided design (CAD), computer-aided engineering (CAE), and computer-aided
manufacturing (CAM). Many IGA technologies have undergone significant advancements
since their introduction, including the development of splines that are simultaneously suitable for CAD, CAE, and CAM and the use of spline-based immersed approaches. IGA and its extensive applications continue to evolve as these methods transition from academia into industry. This minisymposium will feature a broad representation of industrial results and IGA research projects, including presentations from academics consulting on industry projects, software vendors, end users, and academics working on large-scale parallel implementations of IGA.
Keywords:
CAD/CAE Integration, Industrial Applications, isogeometric analysis, Spline-Based Methods
⢠TRACK Number 1200 Industry Applications
â˘
ALLAN ZHONG*, AMANDINE BATTENTIERâ , HAITAO ZHANGâ
* Halliburton
11 Tuas South Ave 12, Singapore 637131
allan.zhong@halliburton.com
â SLB
Sugar Land, TX 77478, USA
abattentier@slb.com, hzhang25@slb.com
Key words: finite element analysis, computational fluid dynamics, AI/ML, fluid structure interactions, optimization, failure analysis, metals, polymers, geomechanics, oil and gas industry
ABSTRACT
This mini symposium aims to provide a single platform for numerical analysts, data scientists, engineers and scientists in oil and gas industry to exchange ideas, showcase best practices on how they use computational tools such as finite element analysis, computational fluid dynamics to understand fundamental issues in designs, tests and field operation issues; highlight the solutions, especially the use cases of AI/ML, to challenging real world engineering problems in every segments of oil and gas industry, from upstream, to midstream, to downstream.
We are interested in the following broad topics:
1) Numerical methodologies to solve complex problems in oil and gas industries - from tools, wellbores to reservoirs;
2) Applications of AI/ML in oil/gas industry
3) Constitutive modeling for metals, polymers, geomechanical materials and their applications;
4) Fluid-structure interactions and other multi-physics simulation;
5) Reliability, uncertainty quantifications and application of statistical FEA/CFD in real world problems;
5) Verification and validation of analysis of complex engineering designs;
6) Engineering simulation, AI/ML driven designs and innovations, and design optimizations;
7) Please contact the mini symposium organizers if you have suggestions for other topics.
Experts and practicing engineers in all disciplines in oil and gas industries are invited to submit their abstracts to this symposium.
Keywords:
AI/ML, computational fluid dynamics, failure analysis, fluid structure interactions, geomechanics, metals, optimization, polymers, finite element analysis
The efficient and reliable modeling of battery cells is a key enabler for improving performance, lifetime and safety in various industrial applications, especially under the increasing demand for advanced energy storage solutions. This mini-symposium addresses micro-macro simulation approaches, focusing on the individual constitutive modeling of anodes, cathodes, and separators, along with the challenges of homogenization to model the mechanical, electrical, and thermal behavior of battery cells and whole battery systems. Therefore, all possible aspects of multi-physical modeling, that could typically include the solution and coupling of mechanical, electrical, chemical, thermal and fluid fields, are targeted.
Coated anodes, typically composed of graphite or silicon, exhibit distinct material properties that significantly influence battery performance. Graphite anodes, while stable and well-understood, are limited by their capacity and lithium-ion diffusion rates, which can lead to capacity fade. On the cathode side, materials such as lithium nickel manganese cobalt oxide (NMC) and lithium iron phosphate (LFP) present varying performance profiles. Separators, which serve as barriers to prevent short circuits while allowing ionic transport, must also be carefully considered. Commonly used materials like polyethylene (PE) and polypropylene (PP) must balance mechanical strength and ionic conductivity. The separatorâs properties can vary based on the design of the cellâwhether it is a prismatic stack or a cylindrical windingâimpacting the overall performance and safety of the battery.
A significant challenge in the industrial application of these technologies is the homogenization of the complex behaviors exhibited by these materials, since the size of high-fidelity models is prohibitive in many industrial applications. Various interaction effects due to charging, cooling, swelling and even mechanical impact on battery systems are of utmost importance for the industrial application of battery systems. This mini-symposium should highlight cutting-edge coupled simulation methods, including AI-driven approaches, that facilitate the integration of detailed microstructural data into macroscopic models. By leveraging machine learning techniques, it is expected to enhance the predictive accuracy of models, enabling real-time optimization of battery designs and performance under various operational conditions.
Keywords:
homogenization, impact, microâmacro coupling, multi-physics, structural mechanics, Battery modelling
This minisymposium is focused on computational methods for solids and structures subjected to severe loads, such as high-speed impact, explosive detonation and blast. Its purpose is to provide a forum for technical presentation and exchange, and to establish communication and collaboration between academic, government and industrial researchers who use computational mechanics for simulations involving these loads. A broad area of contributions is sought, to include numerical modeling of both the severe loads and the subsequent dynamic responses, which may include the coupling of multiple areas of computational mechanics. Typical contributions to this forum come from defense, construction, petroleum, mining, space, counterterrorism or law-enforcement applications. The nature of these applications typically involves some of the most challenging aspects in structural mechanics: nonlinear material behavior under large strains and/or high strain rates; failure and dynamic fracture; initiation, burning, and detonation of energetic materials; phase change and transition; and high-velocity contact and friction. Presentations concerning theoretical developments, multi-spectral physics coupling, new higher-order and isogeometric element technologies, meshfree modeling, algorithms and numerical methods, implementation and parallel computational issues, exploitation of GPU programming, new constitutive modeling, experimental validation, and practical applications are all welcome.
Keywords:
blast, detonation, impact, shock, solid mechanics, Structural Dynamics
The design and verification of structural systems for experimental prototype facilities represent one of the most demanding frontiers of modern structural engineering. Non-conventional structures, such as prototype facilities for physical experiments, large-scale detectors, large-scale vacuum and cryogenic chambers, low-frequency seismic isolation systems, large-scale infrastructure for industrial applications (e.g. [1-4])âoften require highly customized solutions.
Their design requires advanced engineering solutions and computational modelling to capture the interaction of complex geometries, hybrid materials, multi-material assemblies, extreme boundary conditions, and strict operational tolerances.
This mini-symposium will focus on structural engineering challenges and computational mechanics methods applied to non-standard prototype facilities. Topics of interest include the development of advanced finite element models (1D/2D/3D and multiphysics), the implementation of material constitutive laws tailored to non-conventional components and the coupling between thermal, mechanical, and dynamic effects. Special attention will be given to the validation of numerical models against experimental tests, highlighting both laboratory-scale campaigns and full-scale measurements performed on working infrastructures.
Another key aspect is the interaction between numerical simulations and design codes. While Eurocodes, AISC, or other standards provide a general framework, prototype facilities often operate outside traditional regulatory domains. This requires engineers to adapt and extend code-based approaches, integrating them with computational evidence and probabilistic safety evaluations.
From an industrial perspective, the mini-symposium aims to bridge the gap between academic research and practical engineering applications. Industrial partners are increasingly involved in the fabrication and assembly of such facilities, where tolerances, connections, and construction sequences play a crucial role in the global structural response. Computational-experimental frameworks capable of anticipating these effects are therefore of paramount importance to reduce costs, improve safety, and accelerate installation schedules.
Overall, the goal of this mini-symposium is to bring together researchers, engineers, and industry professionals working at the intersection of structural engineering, computational mechanics, and experimental validation.
Keywords:
Industrial Application, Non-conventional infrastructure, Numerical Simulation, Computational Structural Mechanics
1300
Inverse Problems, Optimization and Design
The design process in engineering applications is currently experiencing a change in paradigm, away from experience-based design to numerical design. In many such engineering applications, flows of complex fluids are encountered, posing the challenge of understanding, describing, computing, and controlling these flows. In this spirit, this mini-symposium aims at providing a forum for questions concerning both numerical and optimization methods specific to fluid flow. On the modeling side, it covers the issues related to complex, non-Newtonian flow phenomena, such as the choice of model or appropriate stabilization. Furthermore, in the area of simulation, novel numerical methods, ranging from discretization methods to both free-boundary problems and deforming domain problems, are considered. In all cases, the flow solution may serve as the forward solution of a shape optimization problem. To this end, this mini-symposium will cover novel techniques for shape representation as well as new methods for an efficient evaluation of the design.
Topics of this mini-symposium include, but are not limited to:
⢠Non-Newtonian fluid models describing shear-thinning or viscoelastic properties.
⢠Simulation methods, including stabilization schemes, interface capturing, and interface tracking.
⢠Methods related to shape optimization in fluid flow, in particular geometry representation, reduced order models, and development of objective functions.
⢠Methods particular to specific applications.
Keywords:
Model Order Reduction, Moving Boundaries, Non-Newtonian Fluids, Shape Optimization
In recent years, the performance requirements for mechanical structures have been increasing, and expectations for the industrial application of topology optimization, one of the optimal design methods, have been rising. However, to apply it to actual product design, it is essential to consider complex physical phenomena such as multi-physics, transient response, and material nonlinearity. Furthermore, optimization methods that incorporate manufacturability and economic feasibility are required, making ităindispensable to take a multidisciplinary design perspective rather than focusing on a single design domain.
This mini-symposium focuses on topology optimization for multidisciplinary design and aims to discuss cutting-edge methodologies with industrial applications in mind. We broadly invite research presentations on methods incorporating advanced analyses such as multi-physics, transient response, and material nonlinearity, as well as designs that consider manufacturing constraints and economic feasibility, and optimization through interdisciplinary integration.
Keywords:
Manufacturability, Material non-linearity, Multi-physics, Multidisciplinary design, Topology optimization
Identification of model parameters through calibration processes is critical to achieve realistic results in computational mechanics models. Identifying these parameters from observed experimental data represents an inverse problem. With advances in measurement technology, more experimental data is available to calibrate numerical models (e.g. digital image correlation [DIC]); at the same time, computational models tend to increase in complexity over time, making a rigorous calibration process more difficult. Identifying model parameters is further complicated in the presence of highly nonlinear, unstable, and stochastic physical phenomena considered by computational mechanics models, such as turbulence or fracture. Model calibration is particularly critical for constitutive modeling in solid mechanics but retains application to diverse physics.
A range of techniques have been applied to perform model calibration in computational mechanics, such as gradient-based optimization methods, non-gradient and genetic optimization methods, adjoint methods, and probabilistic (e.g. Bayesian) methods. A new class of models employ machine-learning techniques, such as surrogate models or physical-informed neural networks, for model parametrization. Yet, significant challenges remain for model calibration: these methods can be prohibitively time consuming when applied to high-frequency or full-field experimental data, or to models with many parameters; calibrated parameter sets must be confined to physical spaces and achieve stable solution (e.g. Courant-Friedrichs-Lewy [CFL] condition) when applied to problems distinct from the characterization data.
For this minisymposium, we solicit contributions that address these challenges to advance the state of the field, namely those that (1) propose novel methods for calibration that overcome these shortcomings, (2) demonstrate rigorous application of calibration processes on complex datasets (e.g. high-frequency or full-field data), or (3) validate, assess, or critique calibration methods on distinct applications. Contributions from solid mechanics, fluid mechanics, and other physics are welcomed.
Keywords:
model calibration, optimization, Validation
Composite materials have gained significant attention in modern industries due to their outstanding properties and wide-ranging applications. The emergence of advanced fabrication methods, such as additive manufacturing and automated fiber placement, enables the tailoring of structures to meet specific design needs beyond the scope of traditional materials and techniques. Despite these advances, challenges remain in designing and validating composite structures, mainly due to limitations in existing methodologies and analysis tools.
The proposed Mini-symposium, titled âMulti-Scale Advanced Modeling for Analysis and Design of Composite Structures Including Variable-Angle Tow,â aims to provide a comprehensive overview of the current state-of-the-art and future directions in simulating advanced manufacturing techniques and composite structures. This forum will bring together scientists and researchers to exchange ideas and share recent developments in the modeling and design of classical laminates and variable-stiffness composites produced through additive processes.
The symposium will cover topics such as the development of beam, plate, and shell models; multi-scale methodologies bridging micro- and macro-scale behavior; homogenization techniques; and capturing complex interactions across different length scales.
A key focus will be on addressing challenges related to modeling manufacturing defects, uncertainty in material properties and loading, and their implications for design optimization. Additional areas of interest include simulation of additive manufacturing processes, failure and damage modeling, delamination, robust design, and surrogate- or AI-based optimization methods.
This Mini-symposium aims to foster interdisciplinary dialogue and promote knowledge exchange by offering a collaborative platform for innovative research. The event will support the formation of a network of experts committed to advancing composite materials research and driving their real-world application in engineering.
Researchers from both academia and industry are invited to contribute their expertise. This Mini-symposium offers a unique opportunity to build connections and advance novel methodologies and tools for the analysis and design of additively manufactured composite structures.
Keywords:
Additive manufacturing, Advanced theories, Automated fiber placement, Composite structures, Design, Laminates, Manufacturing Process modeling, Multi-scale modeling, Optimization, Tow-steered composites, Uncertainty
Many problems in various fields of science and engineering are posed mathematically as
inverse problems. In contrast to the more standard âforward problemsâ, where the model
describing the physical system is fully described, and the goal is to find the response due to given sources or known material and topology of the structure, the objective of an inverse problem is to find missing information on the model, based on some given information (acquired by measurements from sensors or by a priori design) on the response in space and/or time or to find optimal parameters or functions. Examples include:
(a) identifying damage in a structure;
(b) reconstructing the displacement field of a structure from strain measurements;
(c) locating the epicenter of an earthquake from measurements on the ground;
(d) finding the optimal topology of an object that yields a desired function;
(e) finding the optimal melt cooling strategy (in space and time) in a crystal-growth process;
(f) identifying the main acoustic sources in a city;
(g) finding the optimal model and material parameters of a solid from experiments.
There is a rich literature on computational methods for inverse problems. Known methods
include Linear Sampling Method, Arrival Time Imaging (Kirchhoff Migration), Time
Reversal, Parameter/Topology-sensitivity based analysis, Stochastic (e.g. Bayesian)
approaches, Full Waveform Inversion, the latter often relying on a gradient-based
optimization and on the adjoint method, and more. Research in this area is very active, since inverse problems are notoriously hard. In order to be effective, methods for the solution of inverse problems must be robust, efficient, and perform well even in the presence of noisy or uncertain data. Additional interesting and important challenges arise from the need for computationally-intensive solution methods. Examples include reduced order and surrogate models, and linear/nonlinear solvers.
Keywords:
Detection, Identification, Optimal, Inverse Problem
This mini-symposium aims to bring together researchers working on various aspects of topology optimization applied to fluids, solids and structures. In particular, we are interested in recent advances in topology optimization. Suggested topics include, but are not limited to:
* Novel and efficient topology optimization algorithms
* New methods to handle manufacturing, stress and other constraints
* Exact solutions to topology optimization problems
* New methods to solve multi-objective topology optimization problems
* Recent advances in reliability-based topology optimization (RBTO)
* Efficient solution of industrial large-scale topology optimization problems
* Inclusion of microstructure in topology predictions
* Recent advances in topology optimization applied to multi-physics problems
* Exploiting high-performance computing in topology optimization considering parallelism by CPU and/or GPU
* New methods of adaptive mesh refinement in topology optimization
* Multiscale topology optimization
* Topology optimization applied to fluid problems
Keywords:
large scale optimization, multi-physics topology optimization, multiscale optimization, reliability-based design optimization, topology optimization
Decisions about the analysis and management of engineering systems are frequently made in the presence of uncertainty. Optimization methods have widespread application in engineering decision-making. Recent advances in the fields of uncertainty quantification, data science, and machine learning provide effective optimization frameworks for handling various challenging situations, e.g., high-dimensional applications, rare and high-impact events in optimization, thereby enhancing the decision-making process. This mini-symposium aims to bring together researchers, academics, and practicing engineers interested in the various forms of optimization under uncertainty for engineering systems. We seek contributions discussing novel optimization algorithms and methods, decision analysis frameworks, and metrics, as well as applications in engineering and science. Areas of interest include, but are not limited to, multi-criteria optimization for engineering systems, robust optimization, performance-based optimization, stochastic optimal control, machine learning-based frameworks for sequential decision-making, optimization for engineering risk management, reduced-order modelling, multi-fidelity formulations, and data-driven optimization.
Keywords:
Computational methods, Decision making, Design, Machine learning, Optimization, Uncertainty Quantification
Fibre-reinforced plastic (FRP) compositesâranging from continuously fibre-reinforced structures to components reinforced with short or long fibresâhave become indispensable in a wide array of industries, including automotive, aerospace, energy, and construction. Their increasing adoption stems not only from their outstanding specific mechanical properties but also from the high design freedom they offer in tailoring both geometry and material composition.
A particularly promising frontier lies in the simulation-driven optimization of FRP components with graded fibre distributions. These allow local adaptation of stiffness, strength, and failure behaviour to the specific performance requirements of complex load cases. Advances in computational methods now make it possible to design and virtually validate such graded structures, where fibre orientation, volume content, and reinforcement architecture vary continuously across the part. This enables truly functionally optimized, lightweight designs with unprecedented material efficiency and mechanical performance.
This minisymposium will provide a platform for researchers and practitioners to exchange insights on the latest developments in simulation, modeling, and optimization of fibre-reinforced plastic components, with a focus on:
⢠Multi-scale and multi-physics simulation of FRP materials and processes
⢠Optimization of fibre architecture for load-adaptive structures
⢠Computational design and validation of functionally graded composites
⢠Process-structure-property relationships in FRP manufacturing
⢠Integration of simulation and experimental validation
⢠Application case studies involving structurally and materially optimized components
Keywords:
design optimization, heterogeneous materials, inverse problems, Composites
Additive manufacturing (AM) offers key advantages over conventional manufacturing processes, including immense form and design freedom, as well as localized control of material properties. Moreover, the ability to deposit multiple materials and create architected materials further widens the design space. However, fully leveraging these advantages is a complex task, where intuition-based design approaches fall short. To systematically address this challenge, increasing efforts are made to develop AM-oriented computational design approaches. These include techniques such as topology optimization and generative design, where specific aspects of AM are integrated into the computational design process through (efficient) AM process simulations or specification of the fabrication characteristics. The computational demands of these methods have also prompted exploration of machine learning to rapidly predict the process conditions or speed up design iterations.
This session aims to provide a forum to exchange ideas and the latest developments regarding computational design techniques for AM. The topic includes contributions focusing on new aspects of:
- Topology optimization / generative design in combination with AM design rules.
- Development and integration of computationally efficient, physics-based AM process simulations into the computational design process.
- Design for spatial distribution of material properties under AM process considerations.
- Optimization of support structure layout for AM parts.
- Computational design of lattice structures including AM considerations.
- Deposition sequence or laser trajectory/tool path optimization.
- Recent advances in machine learning and AI for AM-oriented design and process modelling.
- Industrial AM case studies and/or identification of new design and simulation challenges.
Keywords:
Design for additive manufacturing, topology Optimization
The mini-symposium deals with the current state of efficient and effective methods for large-scale shape optimization and form finding tasks. In particular, it aims to promote the synthesis of innovative technical solutions by combining creativity, simulation, and modelling at the highest level. The goal is to present and discuss the status and challenges in the development and implementation of modern computational technologies that can be used in real-world engineering design applications for structures, fluids, and their interactions. The mini-symposium is aimed at ambitious representatives from science and practice.
Keywords:
engineering design, form finding, Shape optimization
Fiber-reinforced composites have become a crucial element of modern lightweight structures due to their excellent stiffness-to-weight ratio. Unlocking the true potential of these materials requires optimizing the structural geometry as well as fiber orientations. However, available geometric design tools, such as topology optimization, primarily address isotropic materials, where stiffness tailoring by fiber orientation control is mainly performed for standard geometries such as panels. Recent developments in the structural design methods and additive manufacturing technologies have triggered a paradigm shift toward combined optimization of geometry and fiber paths. Despite boosting attainable structural performance levels, this approach brings various requirements such as managing a multitude of variables with different physical meanings, preventing numerical instabilities, and ensuring the manufacturability of output designs. To address these challenges, the development of novel design strategies is critical.
This mini-symposium invites researchers from academia and industry to share their recent findings and insights regarding the design of lightweight fiber composites, considering both form and anisotropy of the structures. Studies concerning both classical constant-stiffness laminates and variable-stiffness composites with curved fibers are welcome. Contributions may involve different geometric design techniques such as density-based, discrete-variable, and component-wise topology optimization. Sequential, collaborative, and concurrent geometry and fiber orientation optimization approaches are all within the scope. In addition to prevalent compliance minimization, other potential design objectives of interest include, but are not limited to, maximization of buckling load, eigenfrequency gaps, and damage resistance. Works involving reliability-oriented, metamodel-based, and/or AI-assisted optimization techniques are also encouraged. Studies centering on manufacturability, experimental testing, and industrial applications are of great relevance as well.
The mini-symposium will offer a unique opportunity for researchers working on lightweight design methodologies to connect with peers in the field and stay informed about the latest developments. We aim to facilitate progress in the design of composite components and the adoption of new techniques in real-world problems by providing a platform for interdisciplinary discussions and knowledge exchange.
Keywords:
Fiber Path Planning, Fiber-reinforced Composites, Lightweight Design, Stiffness Tailoring, Topology Optimization
The solution of optimal control problems (OCP) and sensitivity analysis (SA) has recently received important progress due to its multiple applications and underlying symmetries of its mathematical structure. The minimisation of a time dependent cost function subject to ordinary or partial differential equations [1] may be adapted towards the solution of optimal trajectories in aeronautics and automation, path planning in robotics and elastodynamics, optimal design of mechanisms and structures, or the solution of inverse problems in computational mechanics and biomedicine, among others.
This MS focuses on the formulation and numerical solution of optimal control problems and the analysis of sensitivity in mechanical systems with control parameters. Contributions studying theoretical and geometrical aspects, or practical applications and numerical implementations are all welcome [2,3].
We intend to gather engineers and researchers in applied sciences that investigate the development of numerical strategies for OCP and SA, and that can contribute with novel formulations, applications or numerical algorithms for solving the optimality conditions.
REFERENCES
[1] AManzoni , A Quarteroni , S Salsa. Optimal Control of Partial Differential Equations, Springer, 2021.
[2] P Eichmeir, T LauĂ, S Oberpeilsteiner, K Nachbagauer, W Steiner. The adjoint method for time-optimal control problems. Journal of Computational and Nonlinear Dynamics 16 (2), 021003, 2021.
[3] A. Bijalwan, S. Schneider, P. Betsch, J J MuĂąoz. Monolithic and staggered solution strategies for constrained mechanical systems in optimal control problems. Int. J. Num. Meth. Eng., 2024.
Keywords:
Control, Control Problems, Optimization, Sensitivity Analysis
Modern engineering systems increasingly rely on the integration of multiple interacting components â each potentially made from distinct materials, manufactured through different processes, or subject to different loading and performance requirements. From aerospace and automotive assemblies to biomedical implants and consumer products, the design of such multi-component systems presents several challenges not typically encountered in single-part designs. These include the need to account for part-to-part interactions, interface behaviour, joint mechanics, manufacturing constraints, and system-level objectives. Additional complexity arises in product family design, where multiple product variants must be developed using shared components and design principles to balance performance, cost, and manufacturability across a range of configurations.
This Minisymposium aims to bring together researchers working on computational methods for the design and optimization of multi-component systems. We welcome contributions that explore a wide range of approaches, including:
⢠Topology optimization (density-based, level-set, evolutionary, etc.)
⢠Shape and size optimization for individual parts and assemblies
⢠Gradient-based, heuristic, and ML-based optimization methods
⢠Integration of optimization with CAD, meshing, or automated modelling workflows
Both methodological advances and application-driven studies are encouraged, particularly those that provide insight into how optimization tools can be used to design assemblies that are high-performing, lightweight, robust, or easily manufactured. Topics of interest include, but are not limited to:
⢠Topology and shape optimization for assemblies and modular structures
⢠Multi-material and graded-material optimization in multi-part designs
⢠Optimization of joints, interfaces, and connections (mechanical, adhesive, welded, etc.)
⢠Design of systems with detachable or reconfigurable parts
⢠Decomposition-based approaches
⢠Product family design and optimization of shared component platforms
⢠Optimization problems involving contact, interface behaviour, or assembly sequencing
⢠Integration of design for manufacturing and assembly (DfM/A) in optimization
⢠Numerical techniques for handling large-scale or hierarchically structured models
⢠Case studies from industry demonstrating optimization of complex assemblies
⢠Applications involving structural, thermal, fluidic, or multi-physics performance metrics
Keywords:
design optimization, multi-component, multi-material, product family design, multi-joint , Topology optimization
Design processes in engineering are rapidly evolving from empirical, experience-based approaches to computer-centred methodologies grounded in numerical simulation and data-driven strategies. Within this transformation, topology optimization has become a key enabler, advancing structural, multi-physical and multidisciplinary design, including solid mechanics, acoustics, fluid dynamics, and fluidâstructure interaction problems. These methods drive innovation by revealing novel design concepts, enhancing performance, and improving material efficiency.
Recent progress in advanced manufacturing, for example additive manufacturing, has further expanded the potential of topology optimization, enabling the realization of complex, high-performance structures previously beyond reach. While applications in the automotive and aerospace industries are well established, adoption is rapidly growing across a wide range of engineering fields.
This mini-symposium will primarily address recent methodological advances in topology optimization - spanning solids, fluids, and fluidâstructure interaction - while also presenting illustrative applications that demonstrate its expanding impact in engineering practice.
Keywords:
Additive Manufacturing, Multi-physics
For decades, computational simulations have been fundamental for modeling in science and engineering. Researchers and practitioners have relied on simulations for analyzing properties of physical systems, performing trade studies, and use with design and optimization. Forward simulations alone, however, are particularly inefficient for solving inverse and design problems; they require many forward solves because they do not directly provide sensitivity information, such as gradients, that are needed for optimization. Adjoint-based methods provide this gradient information by solving the adjoint equations backwards in time and have been incredibly impactful for engineering and design applications using relatively few queries [1]. While modern autodifferentiation libraries and advances in GPU technology have been key enablers for the modern success of machine learning, these tools have also served as a convenient framework for creating scalable and fully differentiable physics simulators, where gradients are instead computed through backpropagation. Recently these solvers have been used in a variety of domains [2] for challenging applicationsâsuch as directly optimizing geometries with respect to design parameters and developing optimal controllers by differentiating through the control objective. These approaches continue to show extraordinary promise as an enabling technology for science and engineering.
This minisymposium places an emphasis on (1) recent advances and opportunities in developing differentiable simulations, (2) obtaining gradients for sensitivity analysis of physical systems, and (3) exploiting differentiability to solve challenging control, inverse, and design problems.
[1] Giles, Michael B., and Niles A. Pierce. "An introduction to the adjoint approach to design." Flow, turbulence and combustion 65.3 (2000): 393-415.
[2] Newbury, R., Collins, J., He, K., Pan, J., Posner, I., Howard, D., & Cosgun, A. (2024). A review of differentiable simulators. IEEE Access.
Keywords:
Adjoint Methods, Automatic Differentiation, Differentiable Control, Differentiable Physics, Optimization, scientific machine learning, Sensitivity Analysis
The use of fiber-reinforced composites has been consistently increasing across various industries due to their superb stiffness/strength-to-weight proportion. Earlier studies on the design of such composite structures mainly involved simple quasi-isotropic laminate configurations. Over time, stiffness tailoring techniques have been adopted to utilize the directional properties of the material. Recently, variable-stiffness (VS) composites with curved fiber paths have started to gain prominence thanks to the advancements in relevant design techniques and manufacturing technologies. VS composites enable attaining higher structural performance levels due to improved load distribution, and they have found various applications including vehicle panels, building components, sports equipment, and prosthetics. Stiffness properties of fiber composites can be further tuned as a function of time using mechanical actuation techniques. Despite its significant utility, VS concept involves several challenges including (i) increased difficulty of the design problem due to the presence of locally optimal solutions and a greater number of variables, (ii) fabrication-related defects (e.g., fiber breakage) and imperfections (e.g., tow gaps and overlaps), and (iii) limitations in materials and manufacturing technologies. To tackle these hurdles, innovative design and production approaches are essential.
This mini-symposium welcomes studies on the development of variable-stiffness composites. Contributions may involve structures with different forms of spatially/temporally changing elastic properties, which can be attained using curved fibers, laminate blending, domain discontinuities, and/or actuation elements. Design objectives of interest include, but are not limited to, maximization of stiffness, strength, buckling load, damping, and natural frequency gaps. Gradient-based, derivative-free, reliability-oriented, metamodel/AI-assisted, and single/multi-objective optimization techniques are all within the scope. Works concentrating on manufacturability, experimental testing, and industrial applications are also of great relevance.
The mini-symposium will be an exceptional occasion for researchers from academia and industry to share their latest discoveries and insights regarding the design and fabrication of variable-stiffness composites. In addition, the attendees will have the opportunity to network with other professionals and engage in stimulating discussions.
Keywords:
Fiber Path Optimization, Fiber-reinforced Composites, Laminate Blending, Stiffness Tailoring, Tunable Structures, Variable-stiffness Design
Partial differential equations (PDEs) are one of the most important modeling paradigms in science and engineering. For enabling a sophisticated use of these PDE models, the optimal control, design, calibration, and inversion of them is indispensable. The resulting optimization problems with PDE constraints form a challenging class of optimization problems. Tackling them suitably requires a close interplay of theory and numerics for both PDEs and optimization. Often, further challenges, such as nonsmoothness or random inputs enter the scene. Also, machine learning components are increasingly being incorporated into PDE models.
This minisymposium addresses, from different angles, new solution approaches to the many challenges in PDE-constrained optimization.
Keywords:
Optimization, PDEs, Minisymposium
The sustainable transformation of multiple technological domains is one of the key challenges of this century. Achieving this goal requires an enhancement of the structural performance with respect to predefined criteria. The growing emphasis on sustainability in engineering design, combined with the rapid development of computational methods in engineering mechanics, has fueled the adoption of numerical structural optimization approaches. These methods provide a powerful and versatile tool kit for meeting the demands of present-day engineering.
This minisymposium focuses on current methodological advances in numerical structural optimization. Shape and topology optimization stand as prominent examples, accessible through a multitude of different approaches and enabling the creation of highly efficient, material-saving designs across diverse application domains. Current research expands upon classical formulations by incorporating innovative materials, multi-material concepts, optimization across multiple length scales, manufacturing requirements, and multi-physics coupling to address complex, real-world scenarios. These developments are essential for unlocking the full potential of contemporary technological capabilities.
The contribution of computational innovation to the accomplishment of sustainability goals defines the focal point of the session. In doing so, it will not only showcase the state of the art in numerical structural optimization but also highlight its vital role in shaping the next generation of engineering solutions for a sustainable future. The scope of the session includes, but is not limited to, theoretical developments, innovative optimization algorithms, unusual application contexts or problem formulations, and novel parametrizations.
Researchers active in this forward-looking field are cordially invited to present their latest findings and engage in fruitful discussions aimed at shaping the future of sustainable engineering solutions.
Keywords:
computational mechanics, engineering design, numerical methods, shape optimization, topology optimization, structural optimization, Sustainability
The convergence of agentic artificial intelligence (AI), physics-informed machine learning (PIML), and advanced computational mechanics is opening new frontiers in engineering design and manufacturing. Agentic AI systems move beyond passive prediction by autonomously planning, reasoning, and adapting across designâsimulationâfabrication loops, enabling faster and more reliable solutions to complex, high-dimensional optimization problems. In parallel, PIML embeds governing physical laws into AI models, delivering robust generalization even with sparse, biased, or noisy data.
This integration enables next-generation engineering pipelines that are data-efficient, physics-consistent, and human-centered. These methods have already shown impact in areas such as materials discovery, topology and shape optimization, process control, and multi-physics system design. However, significant challenges remain in scaling these approaches to industrial environments, bridging computational innovations with manufacturing realities, and ensuring trustworthiness and interpretability in deployment.
The aim of this minisymposium is to bring together researchers and practitioners working at the intersection of computational mechanics, AI, and manufacturing. We seek to highlight recent methodological advances, share successful application cases, and discuss remaining barriers to adoption. By fostering a cross-disciplinary dialogue, the symposium will explore pathways toward scalable, industry-ready AI-driven design frameworks capable of transforming manufacturing ecosystems. Topics include, but are not limited to:
Agentic AI for autonomous designâmanufacturing workflows
Physics-informed neural networks (PINNs) and neural operators
Multi-agent and LLM-based collaborative optimization
Generative AI for inverse design of materials and processes
AI-augmented topology optimization and multi-physics simulation
Keywords:
Agentic AI, Generative Design
In the context of simulation-based optimisation, time-dependent systems are notoriously time consuming. This MS seeks to gather researchers working towards reducing the computational cost.
First, the forward state must be solved for the time frame of interest, which is usually done using time-stepping schemes. This solution can take a significant time for realistic problems with high spatial mesh resolution and small time steps. Thereafter, if using the adjoint method for sensitivity analysis, the backward-in-time adjoint must be solved in order to compute the gradients. This essentially doubles the computational time directly and in terms of memory usage, the full forward time history must be saved to be used during the backwards problem. Wrapping an optimisation loop around this, further increases the computational time significantly due to hundreds, if not thousands, of design iterations typically being needed.
This MS focuses on efficient and fast algorithms and methods for optimisation of time-dependent systems to try to avoid or alleviate the abovementioned problems. Contributions should focus on reducing the computational cost or memory consumption for topology, shape or parameter optimisation for time-dependent systems through novel means. Not necessarily applied to optimisation but clearly suited for this context. This could be through improvements to time stepping schemes and checkpointing schemes, or through the application of parallel local-in-time adjoint methods, multiple shooting, or parallel-in-time methods. Relevant application areas include, but are not limited to, thermal diffusion, phase-change materials, fluid dynamics, solid dynamics, fluid-structure interaction, acoustics, and so on.
Keywords:
adjoint sensitivity, parallel-in-time, shape optimisation, time-dependent, Topology optimisation
This mini-symposium highlights emerging topology and shape optimization techniques that are advancing the computational design of materials and structures. It brings together researchers and practitioners working on innovative algorithms, machine learning-assisted and data-driven approaches, and optimization under uncertainty. Applications span aerospace, biomedical, automotive, and civil infrastructure, with emphasis on additive manufacturing, multiscale and multifunctional design, and multiphysics integration. Contributions on architected, nonlinear, bioinspired, and smart materials, as well as supporting software tools, are also welcome. Topics of interest include, but are not limited to:
⢠New topology and shape optimization algorithms
⢠Topology optimization for aerospace, biomedical, automotive, and civil infrastructure applications
⢠Topology and shape optimization for additive manufacturing
⢠Machine learning-assisted, data-driven, and surrogate-based topology and shape optimization
⢠Multiscale, multifunctional, multi-objective design of materials and structures
⢠Multiphysics and multidisciplinary optimization
⢠Stress-constrained topology optimization
⢠Reduced-order multiscale modeling for design
⢠Simultaneous material and structure optimization
⢠Optimization under uncertainty
⢠Design of architected materials
⢠Design of nonlinear materials
⢠Bioinspired design of composites
⢠Design of metamaterials
⢠Smart material design
⢠Software
Keywords:
Material Design, multidisciplinary design analysis and optimization (MDAO), shape optimisation, structural optimization, Topology Optimization
With the increasing complexity of engineered systems and the prevalence of uncertainty, design methodologies have progressed beyond traditional deterministic frameworks. This workshop convenes researchers and practitioners to discuss advances of optimization methods aimed at enhancing the robustness of general high-dimensional, nonlinear systems. This includes robust design approaches with given uncertainty or perturbation model, and tolerance optimization approaches that maximize permissible uncertainty or perturbation.
Areas of application include, but are not limited to structural/fluid/solid mechanics, thermal systems, control systems and mechatronics. Methods to address uncertainty within the design optimization may include all sorts of probabilistic approaches, such as surrogate-based Monte Carlo methods or Taylor expansion-based approximations, embedded in reliability-based or robust design optimization, or stochastic gradient approaches. Alternative methods to handle uncertainty, such as Fuzzy set theory, solution spaces, interval analysis or worst-case approaches are also encouraged to be submitted.
Keywords:
optimization under uncertainty, reliability-based design optimization, solution space optimization , tolerance optimization, Robust design optimization
1400
Manufacturing and Materials Processing
Eco-Efficient Smart Manufacturing of Metallic Materials (ECOSMART-MM) aims to bring together researchers and practitioners from academia and industry to explore the integration of advanced computational methods and sustainability principles into modern metallic materials processing. As manufacturing sectors face increasing pressure to reduce energy consumption, waste, and emissions, this Minisymposium will serve as a collaborative platform to discuss transformative innovations that enhance both performance and eco-efficiency. This session will focus on the implementation of Artificial Intelligence (AI), Machine Learning (ML), and digital technologies in traditional and additive manufacturing processesâsuch as casting, forging, forming, and metal 3D printing[1â4]. Topics include predictive defect modeling, real-time monitoring, digital twins, and reinforcement learning-based process control. Emphasis will also be placed on sustainability-driven innovations like near-net-shape forming, closed-loop recycling, smart energy optimization, and the use of environmentally friendly production practices. In addition to technical innovations, the Minisymposium will address academic-industry collaboration models, including the successful transfer of computational tools and data-driven approaches into industrial production. Speakers will present state-of-the-art research, case studies, and visionary insights to inspire the development of next-generation manufacturing systems that are intelligent, adaptive, and sustainable. This session is intended for researchers, engineers, and decision-makers engaged in computational mechanics, material science, digital manufacturing, and sustainable 2
engineering. The goal is to promote interdisciplinary dialogue and identify actionable strategies to align high-performance metal processing with the imperatives of Industry 4.0 and environmental stewardship.
Keywords:
Computational Mechanics, Eco-Efficiency, Industry 4.0, Metallic Materials, Smart Manufacturing
3D Concrete Printing (3DCP) is an additive manufacturing technique that involves the layer-by-layer deposition of cementitious materials to create complex structures, eliminating the need for traditional formwork. In addition to 3DCP, other techniques such as material jetting and shotcrete-based 3D printing have also recently emerged, further expanding the range of applications for cement-based materials. These advanced additive manufacturing approaches enhance automation, reduce construction time, and enable the creation of customized, optimized designs. However, despite their significant potential, the lack of standardized regulations and the continued reliance on trial-and-error practices often lead to resource waste and inefficiencies. Reliable analytical and numerical models are thus needed to understand, control, and optimize the printing process during all its phases. Accurately modelling 3DCP comes with significant challenges, due to the interaction between printing parameters and material properties. Furthermore, achieving accurate and stable designs requires numerical tools capable of capturing phenomena across multiple spatial and temporal scales, enabling the prediction and prevention of buckling failure, plastic collapse, and cold-joint formation.
This minisymposium aims to bring together researchers from different backgrounds to exchange ideas and provide a platform for presenting and discussing innovative modelling and simulation approaches for additive manufacturing of cementitious materials. Particular focus is given to addressing key challenges, includingâbut not limited toâprocess simulation, cementitious material characterization, and constitutive modelling at the early ages, real-time monitoring strategies, data-driven and AI-informed methods, structural analysis, design, and optimization approaches for 3D-printed structures.
Keywords:
cementitious materials, Numerical approaches, Optimisation
High-performance components produced using Additive Manufacturing (AM) technologies are a reality in many industrially-relevant applications. Nowadays, complex structural parts can be produced using AM technologies with almost unlimited design freedom and locally varying material properties. Therefore, in recent years, the so-called metamaterials, e.g., architected cellular or lattice structures, have known an increasing interest due to the possibility of designing structural parts with tailored mechanical properties and performance.
Despite their huge potential, widespread adoption of metamaterial structures has been hindered by concerns about their structural integrity. Considering the complexity of the manufacturing process, many potential process-induced defects can be present in the final component, limiting their reliability and applications.
Numerical models can thus be a crucial tool to shed light on the complex process-structure-property-performance relationships occurring in AM metamaterials. Only a deep understanding of these relationships will allow us to fully control process errors and their effects on the mechanical properties and performances of these kinds of structures.
In the present mini-symposium, the following topics, related to AM metamaterials, will be considered:
ď§ Cell-based modeling for periodic structures
ď§ Instability and structural behavior
ď§ Material modeling, calibration, and validation
ď§ Effects of defects
ď§ Image-based analysis
ď§ Machine learning techniques
ď§ Process simulations
ď§ Fatigue analysis
ď§ Engineering applications
Keywords:
Additive Manufacturing, Lattice structure, Material modelling, Metamaterials
Heterogeneous materials consist of multiple phases randomly distributed throughout the medium. Examples include voids and minerals in rocks, aggregates and cement paste in concrete, inclusions/fibres and polymer matrix in composites, and polycrystalline grain structures in alloys.
The random morphology of these materials significantly influences their processing, structure, properties, and performance. For porous media, the distribution and connectivity of pore spaces critically affect flow transport within the material. In alloys, features such as grain size, orientation, and boundaries determine physical properties like strength and stiffness.
This mini-symposium, without limiting itself to specific materials, focuses on the characterization, representation, simulation, and application of heterogeneous materials, respecting the heterogeneity of their microstructures. Topics include:
⢠Characterization: Physical testing, microscopic imaging, and statistical analysis of various heterogeneous materials.
⢠Representation: Digital reconstruction of microstructures or representative volume elements (RVEs) of heterogeneous materials.
⢠Simulation: Multiscale modelling, multi-physics simulation, data-driven analysis and artificial intelligence of heterogeneous materials.
⢠Applications: Microstructure design, process optimization, performance evaluation, and fabrication methods of heterogeneous materials.
⢠Other related topics.
Keywords:
Alloy, Multi-physics Simulation, Porous Material, Uncertainty Quantification, Multiscale Modelling
Due to their excellent strength-to-weight ratio, the use of composite materials in various industries (e.g., aviation, automotive, renewable energy systems, etc.) is constantly increasing. To keep up with the rising demand for composite components and simultaneously ensure consistent quality, automation is more and more inevitable and therefore in the spotlight of modern production strategies. Especially processes such as automated fiber placement (AFP) or tape laying (ATL), diaphragm and press forming or filament winding and are becoming increasingly established for serial production. This increase in automation, however, introduces new challenges to the development, adaptation and optimization of manufacturing processes. A key factor to efficient development and optimization is process simulation, which enables the reduction of cost- and time-expensive trial-and-error experiments and a better process understanding and process control. However, there is still a lot of research potential in the field of process simulation, for example regarding computational efficiency, detailed outcome predictions for thermoset and thermoplastic fiber reinforced composites, effects of effects as well as inclusion of the process simulations results in the structural analysis of the part . The mini symposium âProcess simulation for composite manufacturingâ addresses innovations in process simulation, especially regarding detailed process modelling and accurate prediction of process outcomes. It should offer the audience the opportunity to see recent advances in the simulation and validation of automated manufacturing processes. We invite contributions with focus on (but not limited to): simulation and experimental validation of manufacturing processes including e.g. draping, material deposition, curing, consolidation; flow simulation; process induced deformations and effects of effects; digital twins as well as process monitoring, machine learning and artificial intelligence in process optimization. Participants should gain insights into innovative techniques in process simulation and identify current gaps and future research directions.
Keywords:
Composite Manufacturing, Finite Element Simulation, Validation, Process Modelling, Process Simulation, Thermoplastic Composites, Thermoset Composites
Numerical modelling of welding and WAAM processes becomes a decision making tool used to speed up the development and qualification of welding and repair techniques. Both computation welding mechanics which concerns the modelling of welding effects in the base metal and in the weld at solid state (temperature field, microstructure, stress and strain distributionâŚ) and multiphysics simulation which focuses on arc/plasma and weld pool modelling can be implemented for this purpose.
This minisymposia is organised by the Scientific and Technical Committee on Numerical Welding Simulation with the help of the French Association of Mechanics (AFM). The goal is to make an update of the progress made in welding and WAAM concerning:
⢠modelling of the process what can we model today and with what accuracy, what are the couplings effects to be taken into account;
⢠behaviour laws (metallurgy, hardening recovery, viscous effects, simplified methods, ...);
⢠real-life size structures (life time, match computation time with industrial needs, ...).
These elementary bricks will be helpful to characterize the overall welding process in order to numerically simulate the behaviour of a structure (distortions, fatigue resistance, damage), while relying on cases validation tests (calculation / test comparison).
Topics of the minisymposia on modelling and simulation of welding and wire arc additive manufacturing processes in the broad sense will include:
⢠Very large structures, thick components, how to simulate the very large number of passes?
⢠Performance and process control: multiphysics advances for the simulation of welding processes (molten bath and arc) allowing high quality welding.
⢠What are the benefits of integrating the manufacturing history to justify the lifetime.
⢠Effect of welding on the service behaviour of welded joints (low cycle fatigue, stress corrosion, fracture ...).
⢠New models of welding simulation to improve the controllability of structures and make NDT diagnostics more reliable.
⢠Simulation of heterogeneous welding.
⢠Special processes (reloading, repair, FSW, resistance, Hybrid, ...).
⢠Residual stresses and distortions, control of the risks of cracking during welding.
⢠State of modelling materials for welding simulation.
⢠Wire Arc Additive Manufacturing process.
⢠Research and experimental computational tools.
Keywords:
Computational Mechanics, cracking risks, distortions, manufacturing, multiphysics, repair, residual stress, WAAM, Wire Arc Additive Manufacturing, Additive manufacturing, Welding process
The accelerating shift toward intelligent, data-driven manufacturing is transforming how products are designed, produced, and optimized. To meet tight process performance targets and ambitious sustainability goals, manufacturers increasingly rely on modeling frameworks that combine predictive accuracy with computational efficiency. High-fidelity simulations and surrogate modeling techniques have emerged as critical enablers of this transition. They support a broad spectrum of applications across advanced manufacturingâfrom microstructural predictions in additive processes to adaptive decision-making and automation in machining, forming, and joining operations.
This MS aims to engage both the classical computational mechanics and the growing scientific machine learning (SciML) communities in manufacturing by providing an inclusive platform for researchers working at the intersection of high-fidelity simulation, model order reduction, applied AI, and surrogate modeling. Contributions advancing computational methodsâsuch as finite element and meshfree formulations, phase-field and multiphase flow models, and particle-based or discrete element methods for granular and powder systemsâare especially welcome.
Equally important are emerging approaches that integrate physics-based models with SciML to develop hybrid frameworks that preserve interpretability while enabling accelerated predictions essential for real-time control. Relevant topics include reduced-order modeling, physics-informed neural networks (PINNs), Bayesian surrogate modeling, uncertainty quantification (UQ), and adaptive sampling tailored to manufacturing.
The symposium welcomes contributions spanning scalesâfrom microstructural evolution and thermo-fluid/mechanical coupling to part-scale distortions and residual stressesâand materials, including metals, polymers, ceramics, and composites. It also encourages dialogue between communities focused on deterministic high-fidelity modeling and those leveraging data-driven paradigms for design space exploration, optimization, and qualification.
By bringing together expertise in computational mechanics, materials science, and SciML, this MS aims to showcase methodological innovations and real-world applications that demonstrate how integrated modeling workflows can enhance process understanding, enable predictive digital twins, and accelerate the transition toward intelligent manufacturing systems.
Keywords:
Manufacturing, Process Simulation, Scientific machine learning , Surrogate Modeling
Additive Manufacturing (AM) offers highest production flexibility and almost unlimited freedom of design. While truly predictive computational modeling approaches are highly desirable to foster digital process chains in part design and qualification, the multi-scale nature inherent to most AM technologies makes process simulation very challenging.
This minisymposium focuses on recently developed part-scale simulation techniques in AM, including experimental model calibration and validation. Computational modeling and simulation approaches for any class of AM processes (e.g., laser power bed fusion, electron beam melting, directed energy deposition, binder jetting, material jetting, fused deposition modeling, stereolithography) and materials (e.g. metals, plastics, ceramics, concrete and their composites as well as biological materials), and also for related processes (e.g., laser or electron beam welding), are welcome.
Topics of interest for this minisymposium include (but are not limited to):
⢠Simulation of the manufacturing process to predict heat transfer, residual stress, thermal distortion, composition, and microstructure on part-scale
⢠Adaptive discretization strategies, non-standard numerical solution schemes and model order reduction approaches allowing for increased computational efficiency
⢠Physics-based, data-driven, and hybrid approaches
⢠Multi-scale and multi-physics approaches
⢠Thermo-mechanical material modeling including functionally graded materials
⢠Effects of microstructure and defects on mechanical properties
⢠Combined simulation and in-situ monitoring for rapid build qualification
⢠Feedback control for minimizing defects and residual stress in as-built structures
⢠AM-oriented topology optimization including lattice and cellular structures
A companion minisymposium âModeling and Simulation of Additive Manufacturing: Meso- and Microscale Approachesâ focuses on simulation approaches on the meso- and microscale.
Keywords:
3D Printing, Additive Manufacturing, modeling and simulation, Part-Scale Modeling
The intend of this minisymposium is to share the latest advances in modelling the additive manufacturing of industrial components, focusing on wire additive manufacturing and the use of finite element codes and methods.
Keywords:
computational mechanics, mobile heat source, numerical simulation
Additive Manufacturing (AM) offers highest production flexibility and almost unlimited freedom of design. While truly predictive computational modeling approaches are highly desirable to foster digital process chains in part design and qualification, the multi-scale nature inherent to most AM technologies makes process simulation very challenging.
This minisymposium focuses on recently developed micro- and mesoscale simulation techniques in AM, including experimental model calibration and validation. Computational modeling and simulation approaches for any class of AM processes (e.g., laser power bed fusion, electron beam melting, directed energy deposition, binder jetting, material jetting, fused deposition modeling, stereolithography) and materials (e.g. metals, plastics, ceramics, concrete and their composites as well as biological materials), and also for related processes (e.g., laser or electron beam welding) are welcome.
Topics of interest for this minisymposium include (but are not limited to):
- Simulation of AM processes at meso- and microscale resolution, resolving individual powder particles, filaments and subcomponents, or crystal structures
- Mesoscale prediction of heat transfer, multiphase flow and powder phenomena (powder spreading, melt pool dynamics, meltâvaporâpowder interaction, liquid binder flow, filament deposition), surface topology and composition, including defect evolution
- Multiphase interface modeling in melt pool / binder simulations using mesh-free (e.g., smoothed particle hydrodynamics), moving grid (e.g., ALE), or fixed grid approaches (level-set, phase-field, immersed boundary, ghost-fluid, XFEM, CutFEM)
- High-fidelity energy absorption models (ray tracing) for laser- or electron beam-based AM
- Microstructure modeling including phase and grain evolution (e.g., phase-field modeling, cellular automaton modeling), and linking microstructural features (grain size, orientation, phases) to mechanical properties
- Adaptive spatial and temporal discretization strategies, non-standard numerical solution schemes and model order reduction approaches allowing for increased computational efficiency
- Multi-physics and multi-scale coupling techniques
- Physics-based, data-driven, and hybrid approaches
- Combined simulation and in-situ monitoring for rapid build qualification
A companion minisymposium âModeling and Simulation of Additive Manufacturing: Part-Scale Approachesâ focuses on simulation approaches on part-scale.
Keywords:
3D Printing, Mesoscale
models, Microscale models, Modeling and Simulation, Additive Manufacturing
Additive manufacturing (AM) and advanced materials testing increasingly demand predictive simulation frameworks that are not only high-fidelity but also agile enough to enable fast and close to real-time decision-making, control, and design optimization. Traditional numerical approaches offer accuracy but are hindered by computational cost, limiting their integration into iterative design and process control loops. This minisymposium will focus on next-generation computational paradigms that unify large and small language models (LLMs/SLMs), Software 3.0 concepts, and AI-enhanced numerical methods to achieve concurrent calibration, training/learning, and solving of governing physics equations relevant to additive manufacturing.
This minisymphosium will focus on Software 3.0, where models, solvers, and AI components co-evolve during runtime, enabling the direct embedding of reduced-order modelling (ROM) into AM simulation software, streamlining the offlineâonline bottleneck. Examples include tensor decomposition methods such as TAPS1 and INN-TD2, interpretable symbolic surrogates from Ex-HiDeNN3, and scalable convolutional hierarchical deep networks (C-HiDeNN-TD) for ultra-large-scale PDEs4. These approaches eliminate or drastically reduce offline data generation, offering speedups of several orders of magnitude while retaining mechanistic interpretability critical for certification in safety-critical industries.
The minisymposium will also explore AI-enhanced multi-level variational multiscale frameworks that couple macro-, meso-, and micro-scale physics with data-driven inference. Topics include Digital Twin5 for defect prediction, physics-informed AM modelling6, and computational models leveraging differentiable solvers7 for AM and gradient-based inverse design. Novel workflows that jointly calibrate models from in-situ monitoring data, update parameters on-the-fly using various deep learning architecture8, and immediately propagate changes through multiscale solvers will be emphasized.
Particular attention will be given to the interaction between LLM/SLM technologies and AM modelling: large models for multi-modal data fusion9, code generation, and simulation orchestration; smaller, domain-specialized models for real-time inference embedded in machine controllers. These capabilities align with Software 3.0âs vision of self-improving, self-documenting computational ecosystems.
Through case studies in various AM techniques and mechanical testing, such as l
Keywords:
Digital Twin, Software 3.0, Additive Manufacturing, Reduced Order Modelling (ROM)
1500
Materials by Design
The convergence of computational tools and materials science is revolutionising the way we design, model, optimise, and manufacture advanced materials. This mini-symposium will serve as a platform for researchers, engineers, and industry professionals to exchange knowledge on state-of-the-art computational approaches that accelerate materials innovationâfrom fundamental scientific discoveries to practical, real-world applications.
We invite contributions on a broad range of topics, including but not limited to:
⢠Computational Materials Innovation: AI-driven and physics-based methodologies for the design of novel materials, such as high-entropy alloys, composites, metamaterials, and multifunctional systems.
⢠Multiscale and Multiphysics Modelling: Integrated frameworks linking atomistic, mesoscopic, and continuum-level simulations to capture material behaviour across length and time scales.
⢠Digital Twins and Virtual Characterisation: In-silico experimentation, real-time data assimilation, predictive modelling, and machine learning for property prediction, validation, and decision-making.
⢠Optimisation and Inverse Design: Computational strategies for microstructural design, performance enhancement, and manufacturing process optimisation.
⢠Process Modelling for Advanced Manufacturing: Simulation and optimisation techniques for additive manufacturing, hybrid processing, and novel fabrication technologies.
⢠AI and Data-Driven Approaches: The role of artificial intelligence, big data, and digital ecosystems in accelerating materials R&D pipelines.
⢠Industrial Applications and Case Studies: Deployment of computational tools and digital twins in aerospace, biomedical, energy, and defence applications.
This mini-symposium aims to foster interdisciplinary collaboration between academia and industry, promoting innovative solutions that push the boundaries of computational materials research and its impact on society.
Keywords:
Computational Materials Science, machine learning, Manufacturability, Multidisciplinary design, Multiscale and Multilevel Methods, Optimisation, Physics-Based Data-Driven Modeling, Topology optimization
The concept of metamaterial is a new frontier of science, encompassing physics, materials science and engineering [1]. In a broad sense, the term âmetamaterialâ indicates an engineered material with effective properties arising from a tailored design of its architecture, either at the nano/microscale or at the macroscale. The architecture is conceived to attain exotic properties that may not be found in nature, opening new and unforeseen opportunities especially for mechanical and acoustic applications. Furthermore, recent studies of special interest in this field are now targeting metamaterials with programmable properties in space and time, with the purpose of developing materials with multifunctional properties, including mechanical, acoustic, thermal, piezoelectric and optical. In this context, the mini-symposium aims to gather the most recent theoretical and computational studies on mechanical and acoustic metamaterials, with the purpose of stimulating a fruitful discussion among experts in this broad field. Topics covered include, but are not limited to:
⢠Locally resonant metamaterials
⢠Metasurfaces
⢠Origami/kirigami-based designs
⢠Bioinspired designs
⢠Hierarchical designs
⢠Multiphase metamaterials
⢠Time-varying metamaterials
⢠Topological metamaterials
⢠Design, size/shape optimization, topology optimization
⢠Fabrication processes
REFERENCES
[1] C. Lu, M. Hsieh, Z. Huang, C. Zhang, Y. Lin, Q. Shen, F. Chen, L. Zhang, âArchitectural design and additive manufacturing of mechanical metamaterials: A reviewâ, Engineering, Vol. 17, pp. 44-63, (2022).
Keywords:
Computational Mechanics, Design, Metamaterials
The rapid advancement of innovative numerical methods and cutting-edge technologies has revolutionised the development of advanced materials and structures, enabling them to be precisely tailored for specific applications. In material design, the core objective is to meticulously select the chemical structure, composition, and processing conditions to achieve challenging design criteria. These criteria often focus on optimising key properties such as toughness, strength, and durability while simultaneously minimising cost, weight, and environmental impact. In structural design, similar principles are applied to ensure that the architecture of materials and components not only meets functional requirements but also enhances performance, safety, and efficiency across a wide range of applications, from aerospace to civil engineering and beyond.
This mini-symposium aims to bring together researchers working on state-of-the-art topology, shape optimisation techniques and material design. Participants will have the opportunity to exchange ideas, present novel developments, and discuss recent advances in the field. Topics of interest include but are not limited to:
⢠Manufacturing constraint modelling,
⢠Multi-scale and multi-physics optimisation for metamaterials,
⢠Optimisation of functional properties (thermal, acoustic, fluid, etc.),
⢠Incorporating material microstructures into topology optimisation,
⢠Structural and multidisciplinary applications,
⢠Multi-material topology optimisation,
⢠Modelling of manufacturing defects,
⢠Machine learning-assisted, data-driven, and surrogate-based topology and shape optimisation
⢠The integration of computational methods with experimental data to enhance material performance and structural efficiency.
⢠Simultaneous material and structure optimisation
⢠Design of nonlinear materials
⢠Smart material design
Keywords:
Optimization
Over the past decade, there have been significant advances in understanding the mechanics of nano- and bio-materials through in silico investigations. Statistical mechanics-based multiscale modeling provides a rigorous mathematical framework that enables the explanation, understanding, and prediction of macroscopic physical properties based on microscopic observations and parameters. Artificial intelligence (AI) has increasingly been utilized to develop new methods and applications across various fields. The integration of AI with multiscale modeling holds great promise for breakthroughs in the discovery and design of novel materials, particularly for advanced engineering applications.
The design of nanostructured and self-assembled materials aimed at enhancing strength and performance for mechanical and energetic uses has attracted growing attention. A deep understanding of the mechanics and fundamental mechanisms of these materials is essential for the development of innovative material systems.
This symposium will highlight computational mechanics and multiscale analysis for a broad range of engineering applications, aiming to showcase cutting-edge research in multiscale science and engineering. Topics include, but are not limited to:
Biological materials
Biomaterials
Bio-inspired structural materials design
Composite materials design
Multiscale modeling
Artificial intelligence in materials science
Keywords:
AI for materials design, Bio-inspired structures, Biomaterials, Composite design, Biological materials
Improving and accelerating materials development is an important goal in science and industry, as tailored and optimized materials are key to innovation. With recent advances in modeling, simulation, and machine learning, tailoring the chemical composition and the microstructure of materials to a targeted property has become feasible. Beyond improving the performance of materials, this allows us to explore how to adapt materials to recycling requirements. It opens up pathways to mitigate negative effects of contamination and to avoid issues with availability and sustainability of critical elements.
This inverse design process requires computational techniques to characterize and reconstruct the materialsâ microstructures and to quantify and model the mechanical response at the microscale. Computational homogenization of these microstructures then allows for an automated prediction and understanding of the interplay between microstructural features and effective properties. To describe the nonlinear, inelastic effective behavior of materials with high precision while explicitly considering design variables, suitable machine learning approaches incorporating knowledge from fundamental underlying physics enable an improved extrapolation capability and the use of sparse training data obtained from computational homogenization. Beyond constitutive modeling, data analysis and machine learning help to exploit knowledge from simulations in terms of surrogate models and are therefore key to the exploration and prediction of process-structure-property linkages as well as for inverse design.
Topics of interest covered within this mini-symposium include but are not limited to:
- techniques for exploration and inversion of process-structure-property linkages,
- inverse design of heterogeneous materials from metals to architected materials,
- approaches that improve recyclability or sustainability through data-driven design,
- design approaches that account for constraints e.g., related to manufacturing processes,
- data-driven multiscale simulations including microstructure characterization and reconstruction, e.g., 2D and 3D image-based methods, definition of descriptors.
Keywords:
Multiscale Modeling, PSP Linkages, Inverse Design, Sustainability
Phononics is an emerging field focused on controlling elastic and acoustic waves across different frequency regimes to trigger exotic responses in a broad scope of applications, including sound attenuation with lightweight structures, vibration control, shock absorption, wave guiding, stabilization of flow perturbations, and thermal transport at the nanometric scale, among others. Metamaterials are artificially engineered to realize such responses, through specific topological design of their inner structures.
The inherent complexity associated with metamaterials usually requires the use and development of new advanced computational techniques to model and simulate their behaviour, and to optimize their design for specific applications. These techniques include structural and fluid dynamic simulations combined with homogenization, optimization and machine learning methods, addressing challenges like eigenvalue problems and model order reduction for complex systems.
This minisymposium aims to discuss topics related with metamaterials and phononic structures for any kind of application, focusing on challenges and innovations in modelling, simulation or design of such materials. Discussions may extend beyond computational aspects to explore metamaterialsâ exotic properties, applications, or manufacturing challenges.
The following topics fall within the scope of the minisymposium:
- Modelling and design of acoustic and elastic metamaterials for sound and vibration mitigation.
- Modelling and design of architected materials for tailored mechanical response and energy dissipation (mechanical metamaterial, lattice-based structures, etc.).
- Modelling and design of phononic subsurfaces and metamaterials for flow manipulation (applications in pipelines, lifting bodies, wind turbines, etc.).
- Modelling and design of metamaterials and phononic structures at micro and nanoscales (applications in waveguiding, thermal transport, etc.).
- Reduced-order and multi-scale modelling, optimization techniques and AI-driven design of metamaterials and phononic structures.
- Band structure analysis, novel applications and addressing manufacturing challenges in the field of phononics and metamaterials.
Keywords:
Architected materials, Computational design, Metamaterials, Phononic structures
The rapid rise of advanced manufacturing methodsâsuch as additive manufacturing, nanotechnology, and bioinspired designâhas enabled the development of next-generation materials with complex architectures and tailored functionalities, including architected lattices, fibre-reinforced composites, shape-memory materials, electroactive polymers, and metamaterials. While these materials offer vast potential across sectors like biomedicine, aerospace, and energy, their complex microstructures and behavioursâmarked by anisotropy, inelasticity, and multiphysical couplingâpose major challenges for traditional modelling and simulation approaches, calling for new, more capable computational strategies.
This mini-symposium aims to bring together researchers working at the forefront of material modelling, characterisation, and computational design. We are particularly interested in contributions that integrate machine learning and AI, physics-informed modelling, multiscale and multiphysics approaches, and advanced numerical methods. We also welcome experimental, data-driven, and hybrid strategies that inform, validate, or augment computational models.
We invite submissions across a wide spectrum of topics, including but not limited to:
⢠Advanced constitutive modelling of complex and functional materials
⢠Multiscale and multiphysics modelling techniques
⢠Data-driven and AI-enhanced material models
⢠Inverse design and topology optimisation of materials and structures
⢠Computational design of mechanical and multifunctional metamaterials
⢠Bioinspired, programmable, and soft materials modelling
⢠Integrated experimentalâcomputational approaches for model calibration and validation
⢠Digital twins for materials and structures
⢠Emerging methods for uncertainty quantification and robust material design
This symposium aims to foster interdisciplinary discussion and cross-pollination of ideas between mechanics, materials science, computational engineering, and data science communities.
Keywords:
Compuational Material Design, digital twins, Machine Learning, Toplogy Optimisation
ABSTRACT
Mechanical metamaterials are engineered materials with unconventional mechanical behavior originating from artificially designed microstructures along with intrinsic material properties. Advanced computational analysis and design play a vital role in identifying the optimal metamaterial architectures, which can subsequently be manufactured and tested. This mini-symposium aims to focus on the recent advances in computational design and identification of metamaterials, including analytical and semi-analytical approaches, multi-scale finite element analysis and molecular dynamics simulations, along with the recent trends in machine learning, artificial intelligence and material informatics. Besides the static and dynamic property modulation of conventional mechanical metamaterials at different length scales, the mini-symposium will also cover the rapidly emerging trends in this field of coupling the mechanics of material behavior and metamaterial architecture with different other multi-physical aspects such as electrical or magnetic fields, and stimuli like pneumatic pressure, temperature, light or chemical reactions to explore the scope of programming on-demand mechanical responses. The interest in this context would include (but not limited to) the evolving trends and challenges concerning the notions of real-time reconfigurability and functionality programming, meta-surfaces, nano-scale metamaterials, artificial intelligence and machine learning in metamaterials, inverse design and topology optimization, multi-physical origami/kirigami, soft and conformal metamaterials, robotic matter, intuitive understanding in metamaterial design, and computational additive manufacturing.
Keywords:
AI and ML in metamaterial design, Mechanical metamaterials, Meta-surfaces, Origami and kirigami, Programmable metamaterials, Soft metamaterial robotics
Metamaterials offer exceptional design potential across mechanical, thermal, and wave propagation properties beyond those achievable with conventional materials. Among them, auxetic structures, characterized by a negative geometrical Poissonâs ratio, have attracted significant attention for their unique deformation mechanisms, enhanced energy absorption, and tuneable mechanical responses. The rapid convergence of advanced computational methods, precision fabrication technologies, and high-fidelity experimental techniques now enables the integrated development of such materials from concept to application. This mini-symposium aims to provide a multidisciplinary forum for researchers, engineers, and practitioners engaged in the design, simulation, experimentation, and optimization of metamaterials and auxetic structures.
Topics of interest include (but are not limited to):
⢠Novel auxetic and non-auxetic metamaterial geometries
⢠Topology optimization and generative design methods
⢠Multiphysics modelling of coupled mechanical, thermal, and acoustic responses
⢠Experimental validation techniques and advanced characterization
⢠Additive manufacturing and scalable fabrication strategies
⢠Data-driven design and machine learning approaches
⢠Application-oriented metamaterials for aerospace, biomedical, protective, and energy systems
By bringing together expertise from mechanics, materials science, manufacturing, and applied mathematics, this mini-symposium will foster dialogue on the integrated pathways from concept to implementation of metamaterials.
Keywords:
Auxetic structures, design optimization, Experimental Validation, Metamaterials, Multiscale Modeling
Metamaterials are engineered structures whose extraordinary properties arise from their geometry rather than composition, enabling novel functionalities in acoustics, elasticity, energy absorption, electromagnetics, and thermal applications. This minisymposium focuses on the synergy between modelling, simulation, and fabrication in advancing metamaterial science and engineering.
We invite contributions on analytical and numerical modelling, multiscale and multiphysics simulation, topology optimization, nonlinear and tunable metamaterials, as well as advanced manufacturing methods such as additive manufacturing and micro-/nano-fabrication. A key emphasis lies on integrated workflows where computational design guides fabrication and experimental feedback refines models.
By uniting researchers from applied sciences, engineering, and materials science, the symposium will foster exchange on state-of-the-art methods, emerging trends, and real-world applications, bridging the gap between theory and practical implementation.
Keywords:
Additive Manufacturing, Multiscale Modelling, Topology Optimization, Metamaterials, Multiphysics Simulation
The design of next-generation materials requires innovative strategies that combine advanced computational modeling, experimental data, and artificial intelligence to achieve unprecedented functionality, adaptability, and performance. Architected, adaptive, and morphing materialsâranging from metamaterials to bio-inspired structuresâoffer unique opportunities for tailoring mechanical, thermal, and multifunctional properties through the careful arrangement of geometry, topology, and composition [1-3].
This minisymposium will focus on data-driven approaches to the design and optimization of architected and adaptive materials across scales, encompassing studies at any length scale as well as fully integrated multiscale methodologies.
We invite work that explores the integration of AI/ML with physics-based modeling to accelerate material discovery, property prediction, and performance optimization. Topics of interest include, but are not limited to:
⢠Physics-informed and hybrid modeling frameworks for architected materials
⢠Integration of experimental data (e.g., materials imaging) into predictive models
⢠Coupling micro-, meso-, and macro-scale analyses for material and structural design
⢠Data-driven multiscale topology optimization and inverse design
⢠Design under uncertainty, including robust and reliability-based approaches
⢠Adaptive and morphing materials with tunable stiffness, shape, or properties in response to external stimuli
⢠Bio-inspired and functionally graded architectures for enhanced performance
A special emphasis will be placed on approaches that explicitly address uncertainty, where the combined use of stochastic modeling, uncertainty quantification, and robust optimization ensures reliable performance in real-world applications.
Contributions that demonstrate cross-disciplinary linksâsuch as between aerospace, biomedical devices, soft robotics, and energy systemsâare particularly encouraged.
By uniting researchers from computational mechanics, materials science, and data science, this minisymposium will provide a platform for sharing cutting-edge research, fostering collaborations, and shaping the future of data-driven design of architected and adaptive materials. The broad yet distinctive focus ensures relevance to participants working at different scales, while highlighting emerging trends in adaptive and intelligent materials.
References
[1] Kadic et al. 2019
[2] Osanov and Guest 2016
[3] Bishara et al. 2023
Keywords:
Adaptive Materials, Architected materials, Data-Driven Design, Materials Design, physics-informed machine learning, Uncertainty Quantification
Material properties, such as elastic modulus, Poissonâs ratio, mass density, electric permittivity, magnetic permeability, thermal conductivity, heat capacity, or others, are so defined that they are in general greater than zero. Thermodynamic bounds must be satisfied in conventional materials. However, when materials are allowed to have internal microstructures and processes, they may exhibit negative characteristics, such as negative stiffness due to postbuckling processes. Several physical considerations, such as violation of conservation laws in the non-Hermitian, non-reciprocal systems, or odd elasticity, are adopted to examine their stability or exceptional points, where eigenvalues change from real to complex numbers. Possible applications of materials with negative characteristics are abundant, e.g. vibration mitigation, noise reduction, cloaking in electromagnetic, or unbounded effective material properties in composites. In this minisymposium, all aspects from numerical viewpoints in understanding materials with negative characteristics are welcome, including, but not limited to, machine learning techniques to generate material microstructures or properties, and novel analytical or computational methods in calculating physical properties for materials in solid, liquid, or other states. Experimental or theoretical studies to correlate numerical results are also welcome.
Keywords:
Auxeticity, Computational Mechanics, Effective Properties, Microstructure, Negative stiffness
This minisymposium is dedicated to the mechanics of entangled networks, including but not limited to woven and knitted fabrics, knotted structures, polycatenated and fibrous networks. These systems are ubiquitous in both natural and engineered environments, from the cytoskeleton of living cells to advanced textiles and soft robotics. Their unique and often programmable mechanical properties emerge from the intricate interplay of their topology, geometric constraints, and the frictional and constitutive behaviour of their filamentary components.
We will explore the principal challenges in characterizing and modelling these complex networks, focusing on the paramount role of topology in dictating emergent behaviours such as nonlinear elasticity, anisotropy, damage tolerance, and energy dissipation. We seek contributions that employ a wide range of theoretical, computational, and experimental methodologies to advance our understanding. Specific topics of interest include multiscale modelling that bridges the gap between microscopic fiber interactions and macroscopic network response, computational analysis of knotting and unknotting dynamics, the mechanics of polycatenated and interlocked molecular materials, and the influence of fluid-structure interactions on fibrous materials in applications like filtration and biomechanics.
This minisymposium aims to foster interdisciplinary collaboration and provide a platform for new research at the intersection of materials science, computational mechanics, additive manufacturing and quantitative biology. We welcome submissions that not only present new findings but also highlight opportunities for future research and the development of new design principles for these fascinating material systems.
Keywords:
Computational Fabrication, Functionality through Nonlinearity, Inverse Design, Multiscale Modelling, Topology, Entangled Networks
Nature provides a powerful source of inspiration for the design of advanced materials and structures [1]. Biological systems, from bone, tendon, and shell to plant stems and protective cocoons, achieve combinations of strength, toughness, adaptability, and multifunctionality unmatched by synthetic systems. These remarkable properties arise from hierarchical organisation, compositional gradients, and nonlinear mechanics across multiple scales. Translating these principles provides a new paradigm for engineering lighter, safer, and more sustainable technologies across diverse industries.
This minisymposium will highlight advances in bioinspired design and optimisation, bridging fundamental science and practical engineering applications. Topics of interest include the discovery of novel architectures, multiscale biomechanics of natural materials, topology optimisation and evolutionary algorithms, complex nonlinear loading, fabrication of complex structures by additive manufacturing, and emerging applications in energy-efficient transport, aerospace safety, biomedical devices, and sustainable manufacturing. By uniting expertise in computational mechanics, materials science, design optimisation, and advanced manufacturing, this minisymposium will provide a forum to accelerate the translation of biological principles into next-generation materials and systems with global impact.
Keywords:
Bioinspired design, multiscale biomechanics, sustainable technology
This mini-symposium is intended as a forum for presentation and discussion of the results and problems related to mathematical modelling, numerical simulation and experimental testing of advanced materials and smart structures. Of great interest are innovative adaptive engineered cementitious composites with constituent phases, e.g. with shape memory materials (alloys, polymers), which provide the composites and structures with special functionalities, e.g. the ability of self-healing or self-centering, or high damping capacity.
Different scales of observation, including electron and optical microscopy and digital image correlation, as well as descriptions via multiscale averaging methods and computational homogenization, can be considered, including microscopic, mesoscopic and macroscopic scales. The main goal is to find the constitutive relations, taking into account the influence of the interfaces between the constituent phases, to assess the effective properties of advanced materials and finally to build the computational model (digital twin) of the material and structures made thereof in the era of digital technology.
Presentation of both deterministic and uncertainty models, also generated using deep learning techniques, and solution techniques for the coupled chemo-hygro-thermo-mechanical processes and deterioration processes in advanced materials and smart structures, are welcome within the framework of this mini-symposium.
Keywords:
Advanced Materials, Advanced Numerical Methods, Brittle and cohesive fracture, Building materials, cementitious materials, Experimental investigation and validation, Mathematical Modelling, mechanics design of materials, Multi-scale methods
1600
Multiscale and Multiphysics Systems
The Minisymposium focuses on non-conventional techniques for solid and fluid mechanics, including experimental, theoretical, and computational aspects. Attention is focused on heterogeneous/multiscale/multiphase/multifunctional materials and fluids and their behaviour, especially in the framework of coupled field problems and with respect to their fracture and failure arising from scale-bridging processes.
Topics:
⢠Non-conventional theoretical techniques for description of heterogeneous/multiscale/multiphase/multifunctional materials and fluids:
o fractional continuum mechanics,
o tolerance and non-asymptotic modeling,
o peridynamics,
o fractal media,
o nonlocal continuum,
o relativistic continuum mechanics,
o multiscale techniques as combinations of conventional techniques and/or new techniques, etc.
⢠Non-conventional techniques for solving coupled field problems for heterogeneous/multiscale/multiphase/multifunctional materials and fluids (computational aspects including implementation and hardware/software point of view).
⢠New set-ups for experimental testing of heterogeneous/multiscale/multiphase/multifunctional materials and fluids (miniaturized equipment, digital imaging, etc.).
Keywords:
fracture and failure, heterogeneous materials, multifunctional materials, multiphase materials, multiscale materials
The rising global demand for engineering infrastructure and the urgency to minimize the carbon footprint pose great scientific challenges regarding the development of sustainable concretes. As far as computational methods are concerned, models are needed to link the chemical/microstructural composition of cementitious materials to the behavior of engineering structures made from plain and reinforced concrete. Such models help to understand and quantify how physico-chemical processes occurring at nanoscopic and microscopic scales interplay with the macrostructural behavior of engineering infrastructure. Interdisciplinary approaches are expected to be particularly well suited to bridge all relevant spatial and temporal scales. The objective of this minisymposium is to discuss recent advances in computational modeling of concrete and concrete infrastructure. Computational models addressing various length and time scales and physical phenomena relevant for the behavior of concrete and concrete infrastructure subjected to different environmental and loading conditions are welcome. Innovative approaches providing insight into complex phenomena, predictive models increasing safety, durability, and sustainability in practical applications, and models leading to new design concepts in the field of structural engineering science are especially encouraged. Contributions linking different fields of research, e.g., physical chemistry, material science, multiscale and probabilistic mechanics, as well as structural engineering science are also particularly welcome.
Keywords:
multiscale mechanics, physical chemistry, probabilistic mechanics, reinforced concrete structures, cementitious materials, engineering science, Verification
This symposium will be dedicated to various aspects of the multiphysics of fibrous materials including nonwovens, textiles, paper and cellulose products, as well as biological network materials including connective tissue, arterial walls and the cytoskeleton. Fiber composites and composites in which a stochastic network is embedded in a matrix are of interest. Studies of the mechanics, transport, self-healing, activity, and their interactions, in fibrous materials are invited.
These materials exhibit complex behaviors controlled by strong nonlinearities associated with large deformations, instabilities, contacts between fibers, the presence of active elements, dynamic re-organization of fibers, etc. The development of computational methods and of constitutive representations that consider the complex mesoscale physics are of interest for this symposium.
While the focus is on advanced computational methods, presentations that discuss experimental and theoretical aspects and establish links between the structure, topology and physics of the networks and their emerging mechanical response are welcomed. Topics include but are not limited to:
- Multiphysics of semi-flexible fiber networks
- Mechanics of paper, nonwovens, yarns and 2D and 3D fabrics
- Synthetic and biological active networks and networks with transient bonds
- Deformation and forming of composite reinforcements
- Multiscale modelling and homogenization of fibrous materials
- Modelling of reinforced soft matter
- Application of machine learning and data-driven methods to modelling of fibrous materials
Keywords:
deformation, fracture, multiscale computation, transport, Network materials
preCICE is an open-source coupling library for partitioned multi-physics and multi-scale simulations. It enables the efficient, robust, and parallel coupling of separate single-physics solvers. This includes, but is not restricted to fluid-structure interaction. preCICE treats these solvers as black-boxes and, thus, only minimally-invasive changes are necessary to prepare a solver for coupling. Ready-to-use adapters for well-known open-source solvers, including OpenFOAM, SU2, CalculiX, FEniCS, and deal.II, are available. The software offers methods for equation coupling, fully parallel communication, data mapping, and time interpolation. This minisymposium brings together users and developers of the software. It enables the exchange of users among themselves, which otherwise would not know much of each other. Furthermore, the developer team can get direct feedback from users, who they sometimes only know from forum conversations. Lastly, the software and its capabilities can be presented to others in a full and broad sense as not only the developers talk about their software, but also users report on experiences. Recent work focuses on extending preCICE towards two-scale macro-micro coupling, volume coupling including large-scale data mapping, dynamics meshes, and other applications than fluid-structure interaction. For more information, please visit https://precice.org.
Keywords:
co-simulation, Coupled problems, Fluid-Structure Interaction, Multi-physics, Multi-Scale
ABSTRACT
Multiscale computational homogenization methods refer to a class of numerical homogenization techniques for determining the effective behavior of complex and highly heterogeneous materials, and for computing the response of structures composed of these materials. The main added value of computational homogenization consists in surpassing limitations of analytical approaches, e.g. incorporating realistic multi-phase morphologies and complex nonlinear material behavior.
This minisymposium focuses on the developments and applications of either multiscale computational homogenization methods, including all pending challenges in this area, or on modeling and simulation methods at the scale of heterogeneous microstructures with an implicit or explicit connection to another scale. Particular emphasis is given on complex models to incorporate specific phenomena at a given scale and related simulation challenges (complex morphologies, large models, lack of deterministic description of constituents, presence of interfacesâŚ) and emergent behavior (effective behavior not described by individual constituents).
The topics covered include (but not limited to):
⢠FE2 methods and alternatives (e.g. FE-FFT);
⢠Machine-learning/artificial intelligence techniques and surrogate modeling for multiscale analysis
⢠Advanced algorithms for reduction of computational costs associated with multiscale algorithms (model reduction, parallel computingâŚ)
⢠Data-driven multi-scale mechanics
⢠Numerical or virtual material testing across the scales;
⢠Emergent behavior through upscaling
⢠Scientific computing and large data in multiscale materials modeling
⢠Coarse-graining of nano- and micromechanics
⢠Numerical modeling of materials based on realistic microstructures, e.g. provided by high resolution 3D imaging techniques;
⢠Computational homogenization of heterogeneous, linear, time-dependent and nonlinear heterogeneous materials, including material dynamics and metamaterials;
⢠Heterogeneous materials with coupled multi-physics behavior (phase change, chemo-mechanics, nonlinear thermo-mechanics...), including extended homogenization schemes;
⢠Multiscale damage modeling, capturing the transition from homogenization to localization;
⢠Computational homogenization including size effects, higher-order gradients or lack of scale separation;
⢠Numerical modelling of the macroscopic behavior of microstructures with complex interfaces, microcracking, instabilities or shear bands;
Keywords:
Computational Homogenization, Materials, Multiscale methods, Solid Mechanics
Accurate, efficient and robust solvers for complex-rheology flows remain a frontier topic in computational mechanics. This minisymposium concentrates on methodological advances and comparative studies, providing a venue for developers and analysts to exchange cutting-edge ideas, benchmarks and implementational details.
Topics of interest include, but are not limited to
⢠Variational, finite-element, finite-volume, spectral and lattice-Boltzmann formulations for viscoelastic, thixotropic, yield-stress and shear-thinning fluids
⢠Stabilisation strategies for the high-Weissenberg-number and creeping-flow limits (log-conformation, DEVSS-G/SUPG and DG schemes), solvers for integral constitutive equations
⢠Coupled microâmacro and multiscale frameworks
⢠Data-driven and machine-learning approaches
⢠Physics-informed neural networks (PINNs) and operator-learning surrogates
⢠Hybrid solvers that embed ML modules within structure-preserving numerical cores
⢠Reduced-order and latent-space models enabling digital-twin applications
⢠Data-driven discovery of material laws using ML approaches
⢠Sensitivity analysis, adjoint-based optimisation and uncertainty quantification in non-Newtonian settings
⢠Stability, bifurcation and nonlinear wave phenomena in complex-fluid flows
⢠Industrial and biomedical case studies: injection/compression molding, polymer extrusion, film coating, additive manufacturing, inkjet printing, blood rheology, soft-robot actuation, micro-fluidic devices
By bringing together experts on algorithmic theory, high-performance implementation and application-driven validation, this minisymposium aims to accelerate next-generation numerical technologies for complex fluids. We invite contributions from academia and industry, and particularly welcome studies that promote reproducibility and synergy across different methodological communities.
Keywords:
Computational Rheology, Structure-Preserving Discretization
In recent years, considerable progress has been made in connecting the micro- and mesoscopic mechanics of materials to the structural engineering level supported by advances in multiscale modeling. To this end, different classes of computational scale-bridging methods have been developed, spanning various disciplines, e.g. engineering, computational mechanics, mathematics, physics, and chemistry. Although these methods are usually designed for specific research problems, from a methodological point of view, similarities and distinctive features can be identified. This session intends to bring together scientists from different disciplines working on scale bridging problems (both spatial and temporal) in materials and structures. The topics addressed in this invited session include, but are not limited to:
⢠Homogenization-based methods: mathematical and computational homogenization etc.
⢠embedded domain and domain decomposition methods, global-local techniques
⢠heterogeneous multiscale method (HMM), equation-free method
⢠(non-equilibrium) thermodynamics-based coarse graining methods
⢠methods for bridging distinct models, e.g. atomistics-to-continuum
⢠methods for phenomena with (partially) non-separating scales, e.g. localization, damage and fracture or transient phenomena
⢠methods for interfaces and contact conditions
⢠methods for bridging temporal scales
⢠multiscale methods for coupled multi-field phenomena
⢠model reduction techniques for multiscale algorithms and complex microstructures
⢠integration of experimental data (e.g. imaging) into multi-scale numerical algorithms
⢠data-driven and machine learning based multi-scale approaches
⢠quantum-assisted solution methods and numerical algorithms
Keywords:
Computational Homogenization, Coupled problems, Homogenization, Multi-scale modeling
In recent years, significant progress has been made in describing coupled thermo-hydro-mechanical and chemical processes in porous media. These include innovative modelling, numerical and experimental approaches at different scales and for various engineering applications.
For this session, we particularly invite contributions on the following topics:
⢠Phase-field porous media fracture under isothermal/non-isothermal conditions.
⢠Recent techniques applied to study porous media coupled behaviors, such as physics-guided neural networks, convolutional neural networks for predictions, AI image processing of micro-CT and SEM images, inverse design of porous metamaterials, microstructure realization, and reduced-order modeling.
⢠Phase-change processes in porous media, including numerical and experimental approaches related to freezing and thawing in porous materials.
⢠Multi-physics electro-chemical-hydro-mechanical processes.
⢠Additional topics related to solid-phase transitions, microstructure evolution, multi-scale modeling of porous media, and others will also be considered.
Keywords:
Machine Learning in Porous Media, Multiphase Processes, Thermo-Hydro-Mechanical-Chemical Coupling, Hydraulic Fracturing, porous media
Fibrous materials frequently appear, at the macroscopic scale, as thin layers (or "plates"); for instance, as polycrystal-like, porous networks of nanofibers or as composite-like, epoxy resin matrices reinforced by fibres. Such materials are typically lightweight, strong, and often also renewable as well as biodegradable; attributes that offer them considerable versatility and therefore utility in engineering applications. Accordingly, innovative products incorporating such materials and/ or related structures (such as "paper-based" transistors) have been emerging and reshaping our technological landscape. Yet, research and development unleashing such products is still very much experiment-, trial-and-error-based; so that, our understanding of the multiscale organization, as well as the mechanical and physical behaviour of their constituting materials and/ or structures remains poor. Consequently, the innovation process is quite chaotic, and the potential to generate yet unforeseen product applications largely untapped. In addition, considerable uncertainty surrounds the long-term performance of such innovative products throughout their entire life cycle. This is true even for century-old products such as "paper", for which there is still huge room for improvement in terms of controlled production and use [1, 2]. This symposium aims at bringing together scientists and engineers involved in cutting-edge theoretical, computational, and experimental research on multiscale organization, mechanics, and physics of fibrous, thin-layer materials and related structures. Submitted contributions should address recent advances in the following areas: 1) Multiscale, experimental determinations on hierarchical organization (at both continuum and atomistic/ sub-atomistic scales); 2) Multiscale, experimental determinations on mechanical, physical, and coupled physical responses; 3) Theoretical developments and their computational implementation (at both continuum and atomistic/ sub-atomistic scales); 4) Comparison of multiscale, theoretical predictions with respective, experimental determinations.
REFERENCES
[1] P.M. Godinho, S. Scheiner and C. Hellmich, Realistic Multi-Step Micromechanics Model for Paper Sheets considering Fiber Collapse Degree, International Journal of Engineering Science (Submission imminent).
[2] P.M. Godinho, Multiscale, Continuum Micromechanics of Paper: Elasticity and Strength, Doctoral thesis, Vienna University of Technology (Submission imminent).
Keywords:
Fibrous, Materials, Mechanics, Multiscale;, Organization, Physics, Thin Layer
This Minisymposium aims to showcase recent advances in multiscale modeling techniques for the design and optimization of complex and engineered materials, with a focus on computational strategies that bridge microscale material features and macroscale structural performance. The session will gather contributions addressing the development, implementation, and application of numerical methods capable of capturing material behavior across multiple length scalesâan essential capability for the accurate design of architected and heterogeneous materials such as metamaterials, functionally graded structures, and hierarchical composites. A central objective of the Minisymposium is to highlight how multiscale modeling can inform and enhance material and structural optimization processes. Contributions are expected to span a broad range of topics, includingâbut not limited toâhomogenization techniques, computational micromechanics, topology optimization, hierarchical and concurrent multiscale frameworks, and the incorporation of fine-scale material descriptors into large-scale design workflows. Particular emphasis will be placed on models that support the design of materials with tailored mechanical, thermal, acoustic, or multifunctional properties, and that can be integrated into computational pipelines for structural optimization. The Minisymposium will also explore emerging paradigms in material design, such as the integration of data-driven approaches, reduced-order modeling, and machine learning techniques to accelerate multiscale simulations and inverse design processes. Contributions highlighting practical applicationsâranging from additive manufacturing and energy systems to biomechanics and lightweight aerospace structuresâare especially encouraged, particularly when they demonstrate the synergy between theoretical developments and real-world engineering challenges. In addition to presenting technical advances, the session will serve as a platform for discussing open challenges related to computational cost, model validation, uncertainty quantification, and the translation of optimized designs into manufacturable solutions. By fostering dialogue among experts in applied mathematics, solid mechanics, materials science, and computational engineering, the Minisymposium will offer a comprehensive overview of current efforts in multiscale modeling for material and structural design, while also helping to outline future research directions.
Keywords:
Metamaterials, multifunctional materials, Multiscale Modeling, multiscale optimization, Optimal material design, Topology optimization
In the quest to advance material design and optimisation, bridging molecular- and continuum-scale modelling is crucial. This mini-symposium on âBridging Molecular and Continuum Modelling for Innovative Material Discovery and Designâ aims to explore cutting-edge multiscale approaches that seamlessly integrate molecular and continuum frameworks.
Molecular models based on Quantum Mechanics (QM) and Molecular Mechanics (MM) provide essential insights into the fundamental properties of materials, while continuum models, such as Finite Element Methods (FEM), offer robust tools for larger-scale simulations. This mini-symposium highlights methodologies that translate molecular parameters into continuum-scale predictions to refine and innovate material applications.
We will cover a range of topics, including corrosion processes, alloy optimisation, and the development of new sustainable materials. Special emphasis will be placed on translating microstructural effects from the molecular to the continuum scale, improving predictions and material design.
This mini-symposium invites experts from academia and industry to engage in collaboration and knowledge sharing. Our goal is to drive innovations in material science modelling that support new material development for a sustainable and resilient future.
Join us in exploring new computational solutions in material science.
Keywords:
Continuum Modelling, Corrosion processes, Density Functional Theory, Finite Element Methods, Material Design, Material Discovery, Material Optimization, Microstructural Effects, Molecular Mechanics, Molecular Modelling, Quantum Mechanics, Sustainable Materials, Multiscale Modelling
We encourage researchers investigating the various aspects of chemo-mechanically coupled phenomena as they drive microstructural evolution in solids. A special focus will be put on strong coupling effects, i.e. to what extent will mechanical fields influence phenomena such as precipitation, phase segregation, species diffusion, phase transformation. Another interesting aspect within the scope of this minisymposium is the issue of kinetic laws describing the various types of microstructural changes. As these phenomena may have a significant impact on industrially relevant processes, e.g. in heat treatment or additive maufacturing, the minisymposium shall cover also applied engineering aspects alongside with studies on the fundamental physics and thermodynamics initiating and perpetuating microstructural changes. The question as to how mechanical quantities such as stresses or plastic deformation inhibit or promote these changes will also be of interest. We invite reports on research activities at all scale levels possibly also exploiting the bridges between them. Aside from analytical considerations typical modeling tools comprise numerical analyses using phase field methods, FFT techniques, cellular automata, full field finite element models, all supplemented by experimental characterization techniques. Moreover, alternative approaches will be more than welcome.
Keywords:
chemo-mechanical coupling, microstructure evolution, multiphysics, phase transformation
The macroscopic, observable behavior of advanced materials is governed by the structure at different scales. Non-exhaustively, these scales include atomistic and mesoscopic levels. Of utmost interest, understanding failure and evaluating strength and fracture toughness requires multiscale approaches. Advanced materials include nanocomposites, polymer blends, inorganic amorphous materials, smart materials, and hierarchical materials. Advanced materials can also be extended to biological materials, which display complex multiscale features.
A possible classification categorizes the required multiscale approaches into sequential and concurrent methods. Sequential methods obtain findings on the fine scale, which are then applied to the coarse scale in a separate simulation. In contrast, concurrent methods simultaneously consider the coarse and fine-scale in hierarchical or partitioned-domain approaches. In hierarchical methods, both scales are evaluated in the entire simulation domain, while the partitioned-domain strategies only resolve the regions of interest, e.g., the vicinity of fillers in nanocomposites at the fine scale.
Commonly, multiscale strategies do not only bridge scales but also methods and disciplines, which is the scope of this mini-symposium.
Materials: bulk polymers (e.g., thermosets, thermoplastics, elastomers, gels), composites, bio-based materials, biological materials, graphene, inorganic glasses, piezoelectric materials, meta-materials, dielectrics, phase-change materials, architected materials, liquid crystals.
Methods:
- Continuum approaches: Finite element method, peridynamics, numerical homogenization, phase-field methods, topology optimization;
- Particle-based methods: Ab initio, molecular mechanics/dynamics, dissipative particle dynamics;
- Multiscale methods: Atomistic-continuum coupling (sequential, concurrent, hierarchical, partitioned-domain methods), heterogeneous multiscale method, quasicontinuum method, QM-MM, FE2, FE-FFT;
- Multi-physics coupling: Piezoelectricity, flexoelectricity, thermoelasticity, photomechanics, magnetorheology, mechanochemistry, phase transition.
Scales: Atomistic, molecular, coarse-grained, mesoscale, macroscale.
Applications: Prediction of mechanical properties, characterization of processing conditions and production methods, understanding of fracture mechanisms, development of new nanocomposites, identification of structure-property.
Keywords:
Atomistic-to-Continuum Coupling, Methods for fracture and failure, Molecular Modelling, Multi-physics Simulation, multifunctional materials, multiscale mechanics, Numerical Homogenization
Multiphysics problems involve the interaction of multiple physical phenomena, such as fluid-structure interaction, conjugate heat transfer, or aeroacoustics and are prevalent in various fields, including aerospace engineering, biomedical applications, and environmental science. These problems often encompass multiple time and spatial scales; for instance, the heat transfer between a cooled surface and a hot turbulent flow exhibits small convective time scales in the fluid, while the heat transfer into a solid is governed by possibly orders of magnitude larger heat conduction time scales. Therefore, numerical simulations of multiphysics problems often become computationally expensive and require high-performance computing capabilities.
Casting the complete multiphysics problem into a single unifying formulation can be challenging and expensive. Since the physical processes are usually governed by different sets of governing equations, individually optimized solution schemes are frequently used for their solution. While the performance of the dedicated solution schemes is important, the coupling between solvers and the required data exchange between numerical methods can create significant bottlenecks especially on parallel high-performance computing systems. On such systems, minimizing the time needed for communication of coupling information and synchronization of solvers is essential to achieve acceptable computational efficiency. This overhead can become even larger when adaptive mesh refinement is utilized in each solver, necessitating dynamic load balancing not only for each solution method, but also for the fully coupled problem.
This minisymposium invites papers, that address numerical methods for solving multiphysics problems with a focus on their applicability to HPC systems. We welcome contributions related to unifying formulations, coupling algorithms and techniques from various areas of computational engineering sciences with an emphasis on their computational performance. By fostering discussions around innovative approaches and strategies in this domain, we aim to advance our understanding and capability in tackling complex multiphysics challenges effectively.
Keywords:
Computational Fluid Dynamics, Fluid-Structure Interaction, High-Performance Computing, Multi-physics Simulation, multiphase flows
It is a well-known fact that the first 70-80% of the lifetime of a material or a component subjected to real operation conditions occurs in a damage or crack initiation phase, where damage or cracks are on microstructural scale and damage progression is influenced by the materials local microstructure. Hence, to correctly assess the lifetime of materials or components it is very important to take the influence of the materials microstructure into account. However, to address the influence of the local microstructure on the macroscopic behaviour of the material and component for example homogenization and multiscale methods are needed, which are quite numerically demanding.
This mini-symposium is focused on the numerical assessment of the lifetime of materials during this early damage or crack initiation phase, where damage progression is dominated by the materials microstructure. In addition, in real applications materials and components operate under multi-physical conditions, i.e. the material is exposed to elevated temperatures and/or a hostile environment causing degradation for example due to:
- Fatigue loads particularly in the low-cycle fatigue regime causing cyclic plastification
- Corrosion, for example due to corrosive combustion products such as sulfur dioxide, carbon monoxide, nitrogen oxide and water vapour. Here, the impact of the increased water vapour content of synthetically and environmentally friendly produced sustainable fuels is of particular interest
- Hydrogen embrittlement and hydrogen assisted cracking of materials in hydrogen environments. Here, the impact of hydrogen on the materials in fuel cells, storage tanks or in hydrogen combustion engines is a particular focus
- Fretting wear due micro-sliding in contact areas
Particularly under low-cycle fatigue loads damage occurs at the surface of the component, which strongly interacts with other degradation mechanisms acting on the surface. Hence, for this mini-symposium presentations and contributions focusing on the interactions between the aforementioned damage mechanisms are of particular interest. While the mini-symposium is focusing on the development of numerical methods to address the problem areas described above, examples, where a validation of the numerical methods using experimental data was demonstrated shall also be included.
Keywords:
Fatigue, Microstructural Effects, Multi-physics, Multi-scale modeling
Understanding complex systems in biology and fluid dynamics often requires the integration of processes across multiple space and time scales. Cellular environments and macroscopic flows are governed by partial differential equations (PDEs); therefore, the development of accurate and efficient multiscale strategies becomes essential. This minisymposium will focus on recent advances in multiscale modeling and simulation, with particular emphasis on applications in biology, chemistry and engineering.
A key theme will be the effective treatment of problems involving multiple interacting scales. In many applications, different levels of details must be combined to capture both local features and global behavior efficiently. This is particularly relevant in systems where geometry, dynamics and material properties vary significantly across scales.
Natural and biological systems often exhibit complex hierarchical structures. For instance, transport networks in living organisms or plants involve branching patterns that span several orders of magnitude, with varying properties at each level. In such cases, adopting a single model across all scales becomes impractical, and specialized techniques are needed to account for the interplay between scales.
Alongside numerical methods, the minisymposium will also address recent theoretical approaches that enhance our understanding of multiscale systems and help ensure that structural and physical properties are respected during modeling and simulation. Topics of interest include Asymptotic-Preserving numerical schemes, mathematical modeling in arbitrary and evolving geometries, and the numerical challenges posed by stiff systems and moving interfaces. The interplay between accuracy, efficiency, and the ability to capture cross-scale behavior will be a central focus.
We invite researchers working in this area to share recent developments and discuss open challenges. Applications of interest include ion transport and sorption across membranes, vascular and plant network formation, flow in fractured or heterogeneous media, and coupled fluid-structure problems in physiological systems.
Keywords:
biological network formation, highly oscillatory problems, partial differential equations, uniformly accurate schemes
Modeling the behavior of building materials is still a challenge from the point of view of computational mechanics. This is due to the nonlinearities arising in the material behavior during softening/hardening and the complexity of the yield criterion that may describe their deformation capacity under generic triaxial stress states.
Further, when analyzing the interaction between phases (like aggregates and cement paste, or aggregate and Interfacial Transition Zone (ITZ) in concrete) or between materials (like, e.g., in reinforced concrete, or in presence of fiber reinforced polymer (FRP)) some numerical strategies are needed to overcome the additional nonlinearities due to contact at the interface.
Coupled problems arise in many aspects of computational mechanics applied to building materials. Specifically, coupling may involve the formulation of sophisticated mechanical constitutive laws (e.g., elasto-plasto-damage, visco-plasto-damage), or a plethora of multiphysics (e.g., thermo-mechanics, thermo-hydro-mechanics, chemo-hydro-mechanics), even non conventional (e.g., involving radiation, species diffusion, or magnetic/electric field) applications.
Optimal numerical strategies may be useful, depending on the specific application field (FEM, DEM, (semi-analytical methods, machine learning techniques, molecular dynamics, stochastic techniques, âŚ) to reach the adequate level of accuracy and predictability of the models.
The aim of this Mini Symposium is to gather researchers/scientists/experts in the field of computational mechanics applied to cementitious materials at the material and structural scale, with novel approaches on the most challenging numerical issues related to these materials in an efficient way.
Authors are encouraged to present their innovative contributions in the field of theoretical and numerical models for predicting the mechanical behavior of these materials under service and ultimate scenarios. The studies are not limited to conventional materials but are open to their use in combinations with polymers, metallic fibers, improved mixtures/components, as well as eco-sustainable materials. The modeling approaches may pertain to the framework of continuous or discrete mechanics, poro-mechanics or homogenized scale. Both probabilistic and deterministic approaches are encouraged. Investigations spanning all scales are also welcome, from the nanoscale, film, or lattice levels up to the homogeneous macroscale.
Keywords:
3D numerical modeling, Building materials, Civil engineering, Composite materials, Constitutive laws, Multiscale
Physical mechanisms responsible for the behavior of polymers and polymer composites typically take place at a wide range of time and length scales. To capture these appropriately is a true challenge for reliable simulations of this kind of materials: On the one hand, the computational techniques employed here should be able to take into account the relevant time and length scale, but, on the other hand, find a reasonable compromise between computational effort and physical accuracy.
This symposium welcomes contributions focusing on the multiscale computational treatment of polymers and their composites, which are based on profound theoretical knowledge and / or experimental evidence. Of specific interest are considerations of structure-property relations and coupled multi-physics problems including, e.g., chemical reactions, biological processes, electromagnetism, or phase transformations. Additionally, uncertainty quantifications related to the aforementioned fields are of specific interest. In terms of materials, possible contributions may discuss materials like thermosets,
⢠thermoplastics,
⢠elastomers,
⢠gels,
⢠liquid crystal elastomers,
⢠bio-inspired materials as well as composites and nanocomposites thereof.
Further issues to be covered are aspects of
⢠structures in 3D (bulk polymers), 2D (membranes), and 1D (fibres including muscle fibres),
⢠physical states (melts, solids, semi-crystalline and amorphous polymers) and their evolution (polymerization, curing, and crystallization during processing),
⢠mechanical properties (viscoelasticity, plasticity, damage, creep, fracture, adhesion, instability),
⢠coupled problems (piezo-elasticity, electro-elasticity, magneto-elasticity, flexo-elasticity, photoelasticity,
⢠magneto-rheology, crystallization, effects of physical aging and chemical degradation on the mechanical behaviour),
⢠interfacial phenomena like surface and confinement effects, interfaces, and interphases.
Simulation techniques fitting into this minisymposium comprise individual methods like
⢠ab initio approaches,
⢠molecular dynamics,
⢠finite elements,
⢠coupling methods including sequential, hierarchical, and domain-decomposition approaches for atomistic-continuum coupling, coarse-graining, and homogenization, as well as phase-field methods,
⢠big-data and data-driven strategies, and optimization techniques focusing, e.g., on topology optimization.
Keywords:
Biocomposites, Biopolymers, Composite materials, Coupled multi-physics, Multiscale Modeling
This mini-symposium focuses on pore-scale modeling and simulation of porous media, welcoming diverse research to advance both fundamental theory and practical applications [1], [2], [3]. Key topics include:
⢠Novel Algorithms & Computational Methods: Development of advanced numerical approaches (LBM, PNM, finite volume/element methods, DNS, machine learning) for complex geometries, multiphase flow dynamics, and integration with experimental/digital rock data.
⢠Model Validation: Rigorous benchmarking of computational models against lab/field data using pore-scale imaging, flow visualization, and experimental measurements.
⢠Emerging Complexities in Porous Media Flow: Theoretical and numerical advances in multiscale flow dynamics, including multiphase flow, reactive transport, and coupled processes (e.g., chemical-pore-structure interactions).
⢠Upscaling Pore-Scale Results: Bridging pore-scale simulations to continuum models via effective medium properties, constitutive relationships, and multiscale frameworks.
⢠Digital Rock Physics: Advanced in digital representation of porous media structures using high-resolution imaging, segmentation, and statistical characterization of pore networks for accurate simulation inputs.
⢠Coupled Processes with multiphase flow, reactive transport: Modeling interactions such as fluid-fluid displacement and reactive transport that alter pore structures.
⢠CCS/CCUS Applications: Simulations of COâ injection, storage, and trapping mechanisms in subsurface formations. It covered the flow and transport of supercritical COâ, mechanisms of capillary trapping, dissolution processes, reactive transport in carbonate formations, mineralizatin in basalts and saline aquifer, and assessing the long-term integrity of COâ storage sites
The symposium aims to foster cross-disciplinary collaboration and highlight cutting-edge advances in porous media research. Submissions integrating experiments or novel computational methods are encouraged.
Keywords:
CCS, CCUS, Digital Rock Physics, Finite Element Methods, Fluid Flow in Heterogeneous Porous Media, Multiâphase flow, porous media, Reactive Transport, transport phenomena, Upscaling
Ferroelectric materials are a class of multifunctional or so-called âsmartâ materials that have gained increasing scientific and technological interest over recent decades. Beyond their established applications as actuators and sensors in high-precision positioning systems (e.g., semiconductor manufacturing and microelectronics), ferroelectrics are also being explored for advanced functionalities, such as energy harvesting and ultrasonic transducing. The technical achievements of recent years are also increasingly pushing their use in cryogenic environments, notably in the field of quantum computing.
A core challenge in modeling ferroelectrics lies in their inherently nonlinear and hysteretic behavior, stemming from irreversible domain switching processes. These phenomena are accompanied by dissipation, self-heating, and other loss mechanisms, which must be accurately captured in predictive models. Approaches range from phenomenological constitutive models to micromechanical, atomistic, and phase-field descriptions, each offering insight at different length and time scales. The verification of the validity of any modeling approach is highly dependent on the possibility of experimental comparisons.
Since domain switching occurs at the sub-grain level within representative volume elements (RVEs), resolving these effects necessitates scale-bridging techniques. The resulting multi-scale character of the problem demands a comprehensive methodological framework to adequately describe the thermo-electro-mechanically coupled response of such materials.
This minisymposium invites contributions addressing recent advances in the modeling, simulation, and experimental characterization of ferroelectric materials. Topics of interest include, but are not limited to:
⢠Thermo-electro-mechanically coupled modeling including phenomenological, mi-cromechanical, phase-field or atomistic approaches
⢠Multi-scale methods and homogenization techniques
⢠Experimental characterization methods
⢠Ferroelectricity, ferroelasticity, piezoelectricity, electrostriction, electro-/elastocaloric effects, pyroelectricity and flexoelectricity
⢠Domain kinetics and switching phenomena
⢠Rate-dependent phenomena: viscoelasticity, creep
⢠Fatigue behavior and degradation mechanisms
⢠Quasi-static, dynamic, and resonant regimes
⢠Lead-based and Lead-free ferroelectric materials
⢠Applications in sensing, actuation, energy harvesting, transduction and cryogenic environments
Keywords:
Electro-mechanics, Experimental characterization, Ferroelectrics , Multi-scale methods, smart materials
One of the most pressing challenges among material scientists and engineers since the last decade of the twentieth century is to create human-like soft bodies mimicking nature. Machines consisting of a set of materials are usually designed to perform some specific tasks such as generating motion or lifting an object. Hence, one of the most active fields of current research is syntheses, experiments, modelling, and designs of responsive materials that can integrate within machines or act as machines. Responsive materials are smart and innovative substances that can be activated under the application of external or internal stimuli including electric field, magnetic field, pH, light, temperature, humidity or combinations of two or more of them. One of the most promising features of these materials is their ability to undergo large deformations upon the (remote or contactless) application of active fields. Their multifunctional properties make them outstanding candidates for innovative technical applications ranging from large-displacement actuators over smart sensing devices to synthetic soft tissues in flexible electronics. Most of the smart materials have unique microstructures which can be tuned/optimized to further enhance their properties. In case of magneto- and electro-active composites, these are usually composed of a soft matrix and embedded inclusions. From a theoretical and computational viewpoint, this calls for the development of homogenization schemes to help at conceptualizing customized composite's effective properties [3]. Moreover, recent advancements in additive manufacturing (3D printing) provide ample opportunities to intricately design these materials from the micro and nanoscale to âprogramâ their macrostructural response. At the same time, the advance of experimental techniques allowing for precise and reliable validation and testing is paramount.
The goal of this minisymposium is to bring together researchers from experiment, modeling and simulation in order to discuss recent advancements and new directions in the field.
Complex heterogeneous materials offer unique combinations of mechanical, thermal, and functional properties resulting from their intricate microstructures. Accurate multiscale modelling is key to understanding their behaviour, predicting their performance, and enabling their optimisation in advanced applications [1]. In recent years, major advances in deterministic and stochastic methods have made multiscale modelling an expanding research area [2]. This Minisymposium will highlight recent progress in advanced analytical and computational tools for studying complex heterogeneous media, with emphasis on rigorous deterministic approaches and developments in stochastic and data-driven modelling. Topics of interest include, but are not limited to:
⢠Advanced analytical and computational methods for deterministic and stochastic modelling of organised/random microstructures and fracture/damage evolution
⢠Random field modelling of heterogeneous media
⢠Design and optimisation of composite materials and structures under uncertainty
⢠Scale-bridging techniques, including analytical and computational homogenisation and multiscale finite element methods
⢠Multiphysics modelling of heterogeneous materials, including coupled mechanical, chemical, thermal, and electrical processes
⢠Machine learning, data-driven techniques, and high-performance computing for efficient multiscale material analysis
[1] T. Sadowski and P. Trovalusci, Multiscale Modeling of Complex Materials, 556, Springer, 2014.
[2] G. Stefanou, D. Savvas and M. Papadrakakis, Stochastic finite element analysis of composite structures based on mesoscale random fields of material properties, Comput. Methods Appl. Mech. Eng. 326 (2017) 319-337.
Keywords:
analytical and computational tools, complex materials, multiscale and multiphysics modelling
In the relentless pursuit of energy efficiency, sustainability, and the electrification of various sectors, energy storage systems, particularly batteries, have emerged as the linchpin of modern technological advancements. The pivotal role of electrochemical systems, particularly batteries, in powering electric vehicles, renewable energy integration, portable electronics, and grid stabilization underscores the imperative for a comprehensive understanding and accurate modeling of their complicated multiphysics behavior. As we stand at the precipice of a transformative era in energy storage, the need for sophisticated, physics-based models and advanced computational techniques at various length scales to predict, optimize, and control battery performance has never been more pressing.
Our objective is to explore and promote the development of models that go beyond empirical approximations, delving into the intricate electrochemical processes within batteries. These models should encapsulate the physical and chemical phenomena governing energy storage and release, accounting for factors such as electrode kinetics, electrolyte behavior, thermal effects, and structural changes. Through this conference, we hope to facilitate knowledge exchange, collaborative research, and the dissemination of novel computational tools (including both physics-based and AI-assisted modeling) that enable accurate battery performance predictions, state-of-health monitoring, remaining functional life prediction, and the design of sustainable, long-lasting energy storage solutions.
This symposium seeks to provide a prominent platform for researchers, engineers, and industry experts to convene and deliberate on the latest developments, challenges, and breakthroughs in battery modeling and computational methodologies. With a keen emphasis on physics-based and data-driven approaches, this symposium aims to bridge the gap between fundamental electrochemical principles and practical battery applications. By fostering an interdisciplinary dialogue among experts from various domains, including materials science, chemistry, physics, electrical engineering, and computer science, we endeavor to accelerate the pace of innovation in battery technology. We cordially invite researchers, practitioners, and enthusiasts to participate actively in this stimulating intellectual exchange. Together, let us harness the power of physics-based and advanced data-driven battery modeling and computation to shape
Keywords:
Energy storage, Multiphysics simulations
Cauchy continuum-based theories typically employed to model conventional solids may not be able to capture the complex or exotic behaviour of certain materials. In particular, materials exhibiting size effects or atypical mechanical behaviour, like architected materials, metamaterials, and materials undergoing rather complex microscopic phenomena, require models that include additional information concerning their microstructure, such as generalised continua theories and multi-scale approaches. This mini-symposium provides a place for discussion and exchange of ideas regarding the modelling, design and analysis of materials, taking into account their microstructure and their (possibly) non-classical behaviour at different scales. On the one hand, recent advances on the numerical description of this class of materials are foreseen, with focus on the multi-scale modelling through homogenisation schemes, techniques for optimal design of macro or microstructure and constitutive modelling based on generalised continua (Cosserat, micromorphic, strain gradient, ...). On the other hand, there is also place to share the applications and potential industrial transference of this sort of techniques to specific classes of materials, like metamaterials, fibre reinforced composites, polycrystalline materials, biological structures, and architected materials, not to be exhaustive.
Contributions addressing but not limited to the topics listed in what follows are welcomed:
- Multi-scale models based on homogenisation of standard or generalised continua;
- Modelling of size effects, fracture and damage across the scales;
- Reduced-order and surrogate models for multi-scale and/or generalised continua;
- Constitutive modelling and parameters calibration in generalised continua;
- Multi-scale design and optimisation of high-performance materials and metamaterials;
- Numerical methods to solve generalised continua and multi-scale problems;
- Industrial multiscale analysis of complex and high performance engineering materials.
Keywords:
Composites, Metamaterials, Reduced-order models , Size-effects, Homogenisation
In silico engineering increasingly relies on predictive, scaleâbridging simulations that couple mechanics with transport, electrochemistry, heat, and biological growth. Synthetic architected lattices, geomaterials, metamaterials, and living tissues, though chemically diverse, all display hierarchical heterogeneity, evolving microstructure, and strongly nonlinear interactions. Capturing such behaviour demands computational strategies that are simultaneously multiscale and multiphysics. This minisymposium will provide a dedicated forum to present and discuss advances that push these frontiers across synthetic and natural materials.
We invite contributions on hierarchical and concurrent coupling schemes: molecular-to-continuum upscaling, FE², peridynamics, phase-field, DEMâFEM, and other hybrid particle-continuum approaches, etc. Similarly, we invite rigorous experimental validation pathways, data-driven surrogates, physics-informed neural operators, adaptive reduced-order models, and exascale/GPU implementations. Particular emphasis will be placed on tightly coupled phenomena: metamaterials, mechano-chemo degradation in (cementitious) composites; thermo-fluid-structure interaction in additive manufacturing and porous media; electro-mechanical actuation in soft robotics; diffusive-mechanical phenomena in fibrous and granular materials, and growth-remodelling processes in biological systems.
The minisymposium has three objectives: (i) transfer algorithms and best practices across material domains to accelerate innovation and avoid redundant reinvention; (ii) benchmark emerging techniques for accuracy, efficiency, robustness, and uncertainty quantification, with encouragement for openâsource dissemination; and (iii) identify grand challenges from constitutive upscaling to realâtime highâfidelity digital twins and inverse design that warrant coordinated community action. By catalysing dialogue between mechanicians, materials scientists, bioengineers, and data scientists, the session aims to stimulate collaborations that translate advanced modelling into safer infrastructure, more durable energy devices, and nextâgeneration biomedical interventions and monitoring. Diversity in geography, gender, career stage, and industrial engagement will be actively promoted when selecting speakers to ensure an inclusive, impactful exchange.
Keywords:
Biological materials, Computational Fluid Dynamics, Multiphysics, Multiscale and Multilevel Methods
Many problems in computational mechanics involve one or more boundaries that are interfaces between different materials, including interfacial transport, fluid-fluid, solid-solid, or fluid-solid interactions. Problems in this class present many difficulties for numerical solution techniques since they introduce moving boundaries and consequently evolving geometries whose location and topology are unknown a priori. In this Minisymposium, we will provide a forum for researchers to meet and share ideas and experiences in this challenging area of computational mechanics. We seek submissions on all aspects of this problem: theory, formulation, analysis, and applications. Presentations including numerical verification and experimental validation are encouraged. We also encourage submissions on manufacturing flows and free surface flows with non-Newtonian fluids.
Methods include, but are not limited to, level set and volume-of-fluid (VOF), arbitrary-Lagrangian-Eulerian (ALE) methods, immersed boundaries, sharp and diffuse modeling of interfacial zones, deformed geometry remeshing, fictitious domain methods, particle methods, embedded boundary conditions, shifted boundary methods, interface enriched finite element methods, cut mesh/finite element methods. We encourage papers incorporating scientific machine learning for free surface applications etc.
Applications include, but are not limited to, manufacturing flows including additive and other advanced manufacturing techniques, fluid-solid interactions, multiphase flow, dynamics reaction fronts, low capillary flows, surface tension formulations, interfacial mass transfer, melt/solidification front modeling, bubble and suspension dynamics, mold filling, dynamic wetting lines, suspension and emulsion rheology, drying and pattern formation, crystallization and precipitation, and polymer extrusion and mixing.
Keywords:
advanced manufacturing, fluid mechanics, heat and mass transport, multiphysics, solid mechanics, Moving boundaries
Cementitious materials, such as concrete, cement, and mortar, play a pivotal role in modern society as they form the basis of infrastructure development. Given the accelerating rate of urbanization and the ongoing ageing of existing infrastructure worldwide, there is an urgent need to improve the sustainability, durability, and performance of these materials. Virtual laboratories can use modern computational methods to gain deeper insights into the complex behavior of the materials during production, processing, and on site application. This helps to design durable and high-performance materials while reducing carbon emissions.
However, the development of an accurate and comprehensive virtual laboratory addressing these aspects is still in progress due to the immense complexity involved. Cementitious materials are multiphase materials whose properties in the fresh and hardened state are governed by chemical and (multi-)physical properties and processes that range over multiple spatial (Âľm â cm) and temporal (seconds - years) scales.
This mini-symposium will focus on recent advances, challenges, and perspectives in the computational modelling and simulation of cementitious building materials. The focus will be on suitable mapping of time-dependent effects occurring with these materials, starting from the first few minutes after mixing up to years-long processes of damage and degradation. Given the complexity of the materials, the symposium will also address reduced order strategies and materials informatics. Among others, the following topics will be covered by the mini-symposium:
⢠Multiscale and multilevel models (continuum micromechanics, numerical multiscale models)
⢠Reduced-order modelling strategies
⢠Data-driven methods, materials informatics, and machine learning tools for building materials
⢠Methods for simulating damage, fracture, transport and physico-chemo-mechanical processes (e.g. creep, shrinkage, chemical dissolution, chemically expansive processes)
⢠Thermodynamic modelling
⢠Analytical and numerical modelling of cement hydration
Keywords:
Cement Hydration, Concrete, Damage, Durability, Micromechanics, Thermodynamic-Modeling, Voxel-Finite-Element-Method
Multiphysics problems with moving boundaries and interfaces, where multiple physical phenomena interact across distinct and evolving surfaces, represent some of the most challenging and impactful areas in modern computational science and engineering. Such problems occur in diverse applications, including fluidâstructure interaction, phase change processes, multiphase flows, and coupled thermal, electrical, and mechanical systems.
Accurate and efficient simulation of these systems demands advanced numerical methods capable of addressing their inherent difficulties, such as strong inter-physics coupling, discontinuities at interfaces, and the need for fine resolution across widely varying spatial and temporal scales. Recent advances have also included data-driven or physics-informed surrogate models, constructed via machine learning from high-fidelity simulations or experimental data. These methods can greatly accelerate simulations while preserving accuracy.
This mini-symposium will present recent advances in numerical methods for moving boundary problems in multiphysics, bringing together leading researchers who are developing cutting-edge algorithms and their applications. Topics will include novel discretization strategies, high-fidelity simulation techniques, and robust algorithms for tackling nonlinearities, complex geometries, and large deformations. Special emphasis will be placed on approaches that improve stability and accuracy, such as adaptive mesh refinement, innovative interface tracking and interface capturing methods, multiscale modeling, and machine learning-enhanced simulations.
Beyond methodological developments, we encourage submissions which feature real-world applications demonstrating how to improve predictive capabilities and optimized designs. Some areas of interest are, but not limited to, aerospace, energy, materials science, and biomedical engineering. By fostering discussion and collaboration, this session aims to accelerate progress in the simulation of interfacial multiphysics problems, pushing forward the state of the art in this critical area of research.
Keywords:
Fluid-structure interaction, Moving boundaries and interfaces, Numerical methods, Machine learning-enhanced modeling, Multiphysics simulation
Interfaces play a critical role in the behaviour of multi-phase and composite materials, governing mechanical response, transport phenomena, ageing, and failure. Advances in computational mechanics enable the detailed representation and analysis of interfaces across multiple scales in space and time, and under coupled physical processes [1,2].
This minisymposium will provide a platform for researchers to discuss novel methodologies and present their latest findings on the modeling, analysis, and simulation of interfaces, with emphasis on multi-scale and multi-physics applications. Contributions may address modeling concepts, discretization techniques, or both, applied to a broad range of interface-related problems. Topics of interest include, but are not limited to:
⢠Computational homogenization of microstructured and multi-phase materials with interfaces
⢠Dimensional reduction techniques for thin interphases
⢠Coupled transport processes (diffusive, convective) across or along interfaces
⢠Chemical reactions at interfaces
⢠Interface contact under coupled mechanical, thermal, chemical, or electrical effects
⢠Damage initiation, propagation, and failure at interfaces
⢠Aging and degradation of interphases in multi-physics environments
Applications of interest include, for example:
⢠Battery electrode materials including novel solid state battery designs
⢠Fuel cell membranes and electrode interfaces
⢠Fiber-reinforced polymerâmatrix composites
⢠Grain boundary modeling
⢠Geomechanical and geological interfaces in multi-phase porous media
⢠Thin-film coatings and protective layers in harsh environments
[1] D.R. Rollin, F. Larsson, K. Runesson and R. Jänicke, Variationally consistent homogenization of diffusion in particle composites with material interfaces using dual macroscale chemical potentials, Computational Mechanics (2025), doi.org/10.1007/s00466-025-02605-5.
[2] M.G.D. Geers, V.G. Kouznetsova, K. Matous and J. Yvonnet, Homogenization methods and multiscale modelling: Nonlinear problems, in: Encyclopaedia of Computational Mechanics, second edition, Wiley, 2017.
Keywords:
Computational Homogenization, Material Interfaces, Multi-physics
Materials with a complex microstructure have several special features manifesting at different length scales and causing various behaviour patterns. This motivates the development of numerous numerical methods for their investigation. On the one hand, the multiscale approaches summarize the information on the material behaviour at lower length scales to predict effective structural behaviour, following different assumptions such as volume averaging, statistical averaging, or energy balance expressed through the Hill-Mandel upscaling condition. On the other hand, the complexity of the material behaviour under different loading cases pushes the development of ML-based approaches. Here, classical methods characterized by the evolution of internal state variables are replaced by Neural Networks (NN) that are able to memorize. Typically, the focus is placed on RNN, LSTM, and GRU, but the alternative NNs, such as Physics-Informed NNs architectures, are also in use. The latter are especially advantageous since they address the question of a limited amount of informative data, the question of physical admissibility, the need to explicitly deal with uncertainties, and the need to provide explainable and interpretable inferences. The minisymposium covers novel numerical aspects in both kinds of approaches, multiscale and ML-based ones, as well as their contribution to a better understanding and improvement of the material behaviour of structural materials, especially 3D printed materials.
Keywords:
3D Printing, Machine Learning, Material modeling, Multiscale Modeling, neural networks
Engineering materials are inherently heterogeneous at micro- and meso-scales, often exhibiting variability due to manufacturing imperfections or natural processes. Capturing the mechanical response of such materials at the macroscopic scale requires a modeling framework that accounts for both aleatory (random) and epistemic (knowledge-based) uncertainties.
This minisymposium explores modern approaches to multiscale material modeling that integrate stochastic methods and machine learning to represent and propagate uncertainty across scales. A central focus lies in learning effective macroscopic constitutive laws from microstructural data and understanding how uncertainties at lower scales affect system-level behaviorâparticularly in nonlinear regimes.
We invite contributions in the following areas:
⢠Stochastic modeling of heterogeneous and anisotropic materials
⢠Stochastic and classical multiscale methods
⢠Scientific machine learning for uncertainty-aware multiscale modeling
⢠Uncertainty quantification in nonlinear material behavior
⢠Model order reduction techniques for multiscale problems
Keywords:
Computational Mechanics, Machine Learning, Stochastic Multiscale, Uncertainty Quantification
This minisymposium aims to foster communication and discussion on recent developments and
applications of novel multiscale computational and data-driven approaches for advanced
materials and structures, covering a wide range of spatial and temporal scales.
Recent progress in multiscale computational approaches has focused on constitutive modeling
and the more comprehensive description of nonlinear deformation, physical failure, damage
evolution, and environmental aging of advanced materials and structures (including nanoscale
aggregates, mesoscale structures and segregations, and macroscale laminates). In particular, the
reactive nature of materialsâ behavior is investigated at extreme scales to establish structure
property relationships.
On the other hand, data-driven computational approaches have introduced an innovative
paradigm shift in computational engineering, supported by the rapid growth of methodologies
such as proper orthogonal decomposition, deep learning, and machine learning. Recent
achievements in this area have demonstrated the potential to enhance the performance of
multiscale computational modeling and simulations.
Keywords:
Data-driven analysis, Multiscale analysis, Multiscale fracture mechanics
Kinetic equations bridge microscopic particle dynamics and macroscopic hydrodynamic theories. They arise in a wide range of natural phenomena and engineering processes, such as rarefied gases, plasmas, particulate flows, and collective behaviors of active/living matter. Despite the solid physical foundation, the high dimensionality of kinetic equations makes them computationally prohibitive for most practical applications.
Moment methods provide an effective model reduction framework, offering physical interpretability and computational tractability while preserving key physical properties such as conservation laws. Recent years have witnessed rapid advances in both theory and computation of moment systems. Novel closure strategies have emerged, ranging from classical Grad-type and quadrature-based approaches to entropy-based formulations, macro-micro decompositions, and data-driven or hybrid closures. Parallel developments in mathematical analysis of realizability, hyperbolicity, dissipativeness and well-posedness have provided a stronger theoretical basis for these methods. Furthermore, advanced numerical techniques are expanding the applicability of moment systems in practice.
This Minisymposium aims to bring together researchers working on applied mathematics, computational science, and engineering to present the latest progress, exchange ideas, and enlighten future research directions of moment methods and numerics for kinetic equations. Contributions are particularly encouraged in:
(1) Moment closure strategies, either mechanistic or data-driven, with solid mathematical foundations and computational efficiency.
(2) Advanced numerical methods, including (but not limited to) high-order, asymptotic- and/or bound-preserving schemes, and methods for non-conservative moment systems.
(3) Applications of moment methods to problems in applied sciences and engineering.
Keywords:
Moment methods, Numerical methods, Structure-preserving models, Kinetic equations
1700
Numerical Methods and Algorithms in Science and Engineering
Partial differential equations (PDEs) provide fundamental models for simulating complex phenomena in science and engineering, with applications ranging from fluid dynamics to materials science. Solving (parametric) PDEs efficiently remains a critical challenge due to their high-dimensional and computationally intensive nature. Recent advances in computational methods have highlighted neural networks (NNs) as promising discretization tools for addressing these challenges, leveraging their ability to serve as universal function approximators.
Neural networks have demonstrated success in diverse fields, offering innovative approaches to regression, classification, and the solution of initial and boundary-value problems. Unlike traditional numerical methods, NNs represent functional manifolds with nonlinear approximation properties, facilitating the exploration of high-dimensional and complex solution spaces.
Neural networks are increasingly used in applications such as solving high-dimensional linear and nonlinear PDEs, uncertainty quantification (UQ), optimization problems, and inverse problems. Additionally, Physics-Informed Neural Networks (PINNs) have been introduced to enforce PDE constraints by collocating the strong residual on training points, proving effective in forward and inverse problem simulations. However, challenges persist in terms of runtime efficiency, error control, the choice of loss formulations, control of integration errors, design of proper optimization tools, and understanding the mathematical principles underpinning NN-based methods.
This mini-symposium highlights recent advances in scientific machine learning (SciML) and neural network methodologies for solving parametric PDEs, with emphasis on mathematical analyses and computational improvements. Topics of interest include convergence and stability properties, a posteriori error estimation, adaptive strategies for network architectures, optimization techniques, proper integration strategies for neural networks, and neural network technologies combined with other traditional numerical methods like finite elements.
Keywords:
machine learning, numerical techniques
The objective of this mini-symposium is to bring together experts from academia, industry and national labs to discuss the most recent advances and emerging research directions in the field of stabilized and variational multiscale (VMS) methods, which play a crucial role in the accurate and efficient numerical modeling of complex fluid mechanics problems. The symposium will cover theoretical developments, efficient numerical schemes, multiphysics coupling, adaptive strategies, novel mesh-moving techniques, embedded formulations, advanced discretization approaches, error estimation, uncertainty quantification, reduced-order modeling, and the integration of stabilized and VMS methods with data-driven approaches and machine learning. Discussions will also focus on applications of stabilized and VMS methods in biomedical engineering, aerospace (from low subsonic to hypersonic regimes), renewable energy, additive manufacturing, marine and automotive industries, among others. By fostering interdisciplinary collaboration, this mini-symposium will provide a platform for researchers to exchange ideas and discuss cutting-edge advancements in stabilized and VMS methods, high-performance computing (HPC), and software development for complex fluid dynamics applications.
Keywords:
CFD, Fluid Mechanics, Stabilized methods, VMS
This mini-symposium is dedicated to the discussion of recent developments and applications in the field of numerical simulation of multiphase flow in porous media, encompassing petroleum reservoirs, aquifers, nuclear disposal, carbon storage, hydrogen storage, geothermal energy, transport of contaminants, poroelasticity and related disciplines, including new gridding, mesh adaptation, advanced numerical formulations, artificial intelligence methods, multiscale and multilevel methods. The goal is to bring together researchers, students, and professionals in the field of petroleum reservoir simulation and all areas involving porous media flows. The scope of the mini-symposium ranges from mathematical and computational methods to the modeling and simulation of challenging applications in multiphase flow in porous media.
Keywords:
Advanced Numerical Formulations, Flexible and Adaptive Meshes, Multiscale and Multilevel Methods, Fluid Flow in Heterogeneous Porous Media
Slender structures are employed as key elements in various engineering applications. Their utilisation is further driven by the introduction of new materials that allow the design of highly optimised shapes. Geometrical nonlinearities drive the mechanical response of these structures, while their material may behave either elastically or inelastically. Furthermore, imperfections can significantly influence their response and safety. As such, the development of robust, efficient, and accurate numerical methods for analysing slender structures remains a fundamental research area in computational mechanics. In light of these considerations, this mini-symposium aims to bring together researchers from around the world who are advancing methodologies for the geometrically nonlinear analysis of structures in civil, mechanical, marine, aerospace, and biomedical engineering.
Contributions may involve:
- Discretisation methods as weak and strong formulations (i.e., Finite Element Method, Virtual Element Method, Boundary Element Method, Galerkin formulations, Isogeometric Analysis).
-Advanced methods and algorithms to recover the nonlinear behaviour of lightweight structures.
-Reduced-order, surrogate, machine learning, and data-driven models.
-Geometrically nonlinear phenomena in coupled problems.
-Multi-level and multi-scale analysis of geometrically nonlinear structures.
-Numerical methods for the imperfection sensitivity analysis and reliable safety assessment of slender structures.
-Optimisation of lightweight structures in nonlinear range.
-Enhanced structural 3D beam and shell models undergoing large deformations.
Keywords:
buckling and post-buckling, discretization techniques, nonlinear analysis, slender structures
Lightweight composite structures are widely employed across various engineering fields, due to their advantageous properties such as high specific strength/stiffness and low weight. Despite these benefits, many lightweight structures are prone to unstable post-buckling behaviour under compressive loads, often amplified by significant imperfection sensitivity, which can lead to potentially dangerous reductions in load-carrying capacity. However, this geometrically nonlinear response is not inherently undesirable. In specific contexts, particularly those requiring adaptive or morphing capabilities, nonlinear deformations and instabilities can be strategically harnessed to enable complex shape reconfigurations and facilitate the rapid deployment of structural systems. It is therefore of pivotal importance to develop and employ methods capable of accurately and efficiently predicting the geometrically nonlinear behaviour of lightweight composite structures. This mini-symposium seeks to bring together researchers engaged in the development of advanced methodologies for analysing, tracing, and optimising the nonlinear structural response of lightweight composites. Contributions may address, but are not limited to, the following topics:
- Advanced computational methods and discretisation techniques for evaluating the stability of lightweight composite structures, while also accounting for multiphysics interactions.
- The use of AI/ML techniques to construct surrogate models for accelerating the design and optimisation of lightweight nonlinear structures.
- Optimisation algorithms to solve convex and non-convex problems with single or multiple objectives for composite structures dominated by geometrically nonlinear behaviour.
- Uncertainty quantification techniques to assess the influence of parameter variability on structural performance in the nonlinear regime.
Keywords:
Composite Structures, Machine Learning, Optimisation, Nonlinear analysis, Uncertainty Quantification
Computational Fluid Dynamics (CFD) simulation can be considered undoubtedly a valid means for investigating the fluid dynamic properties of complex configurations for different engineering applications. The study of the aerodynamic degradation due to ice-accretion represents a typical example of research effort to prevent in-flight disasters. In particular, aero-icing tools have been recognized in the last decade by the aeronautical regulatory agencies as future and desirable key instruments for building-up proper certification roadmaps [1]. Different European projects have been launched in the last years [2,3] linking aeronautical industries with academic and research entities. The development of accurate and robust numerical methods is crucial for mitigating the high costs linked with wind-tunnel experiments and flight tests on icing conditions [4], while increasing safety.
A number of numerical issues arise due to the multi-physics nature of the problem which involves multi-phase flow, thermodynamics as well as geometry handling. Moreover, the mesh generation process itself represents one of the major bottlenecks during icing analyses especially for body-conforming methods. Indeed, the study of flows involving complex geometries requires extensive manpower for generating body conforming meshes.
An alternative is the use of immersed boundary (IB) techniques based on Cartesian meshes. Their non-body conforming nature allows the coding of automatic and very-fast algorithms for grid generation as well as wall-surface tagging. Adaptive mesh refinement (AMR) procedures help clustering cells in proximity of the wall and in zones of high flow gradients. In the past, questions were raised about near-wall accuracy due to the use of wall-models especially at high Reynolds numbers. Besides, the unstructured nature of the AMR meshes makes the Cartesian methods less efficient than classic body-fitted structured codes in terms of data management and computational effort. Moreover, the application of IB methods to the icing topic is today uncommon.
All the above points, and others such as High-Performance Computing, geometrical pre/post-processing and Level-Set boundary interface, represent open research areas that deserves to be addressed to allow the IB methods raising themselves as assessed tools of analysis and design towards a new generation of icing-free surfaces and/or ice-protection systems (IPS).
Keywords:
Aero-Icing, Aircraft Aerodynamic Design & Analysis, High-Performance Computing, Immersed Boundary Methods, Computational Fluid Dynamics, geometrical pre/post-processing
Meshfree (or meshless) methods provide significant flexibility in spatial approximation by eliminating the need for element connectivity, making them particularly advantageous for problems involving complex geometries, large deformations, and evolving discontinuities. In recent years, these methods have seen broad development and application across diverse disciplines, including solid/fluid dynamics, extreme event simulation, solid-fluid interactions, isogeometric analysis, nonlinear mechanics, inverse problems, peridynamics, geomechanics, biomaterial modeling and emerging machine learning techniques. A wide range of meshfree formulations have been proposed, such as smoothed particle hydrodynamics, element-free Galerkin methods, reproducing kernel particle methods, material point methods, generalized finite difference schemes, strong-form collocation approaches, and physics-informed neural networks, and many others. This minisymposium aims to provide a platform for researchers from engineering, mathematics, and industry to present recent advancements, innovations, and applications of meshfree methods. We warmly invite contributions related to the theoretical development, computational implementation, and practical use of meshfree approaches in computational mechanics and applied mathematics, as well as their integration into real-world industrial applications.
Keywords:
Large/Extreme Deformation, Theoretical and Applications, Meshfree Methods, Particle Methods
Multi-physics systems often involve the solution of coupled partial differential equations (PDEs) defined across domains with different topological dimensions. This scenario is common in various fields, among them geology, biomedicine, fracture mechanics, material modeling, or fluid-structure interaction to only name a few. Mixed-dimensional modeling addresses these challenges by simultaneously solving PDEs of varying dimensionality. With its many applications and recent progress in modeling and the analysis of the underlying PDEs, mixed-dimensional modeling has recently become a lively field of research with its own identity.
The differences in dimensionality across the coupled PDEs introduce unique challenges throughout the simulation process: During the modeling phase, suitable coupling conditions must be established to bridge the dimensionality gap. In the discretization phase, careful consideration is required to ensure accuracy and stability, particularly in the imposition of coupling conditions. Furthermore, the resulting systems of equations need to be solved both accurately and efficiently to pave the way for accurate predictions in a variety of applications.
This minisymposium is dedicated to all aspects of mixed-dimensional modeling of multi-physics systems, with particular attention to accurate and robust methods for their discretization, for the efficient preconditioning and solution of the corresponding matrix equations, and to showcase challenging applications of mixed-dimensional models in science, engineering and biomedicine. The minisymposium is aiming to attract scientists form a broad range of scientific communities who are currently using mixed-dimensional PDE models for their research to foster discussions and share insights and ideas.
Keywords:
discretization techniques, Mixed-dimensional modeling, Multi-physics, Solver
Anisotropic phenomena are frequently encountered in advanced engineering and scientific applications. They arise, for instance, in materials science, fluid dynamics, and coupled multiphysics problems. Accurately approximating such problems requires strategies for mesh generation and refinement that capture directional dependencies in material properties and solutions.
Developing computational techniques for anisotropy presents algorithmic and numerical challenges. Resolving thin layers with steep gradients, typical in boundary layers and convection-dominated problems, requires meshes adapted to dominant error directions. Such anisotropic adaptation improves efficiency over isotropic refinement but is more complex due to difficulties in controlling mesh quality and avoiding ill-conditioned linear systems.
Similar issues arise in anisotropic materials like reinforced composites and layered structures, where the mesh must resolve directional dependencies of coefficients and solution fields. Discretization often leads to ill-conditioned systems, necessitating specialized solvers and preconditioners for efficiency in practical applications.
Moreover, anisotropic mesh refinement poses challenges from a mathematical perspective. Error estimates that ensure convergence must be derived not with respect to a general mesh size, but to mesh sizes in different directions. Developing such estimates requires advanced analytical techniques. While this has been studied in classical finite element methods, significant work remains for advanced discretizations. This is especially relevant in the field of Isogeometric Analysis, which employs separate refinement in parametric directions by default. Additionally, complex isogeometric parametrizations can introduce anisotropic mesh features that must be accounted for in the mathematical analysis.
This minisymposium brings together researchers working on the mathematical theory and advanced computational methods tailored for anisotropic material properties and anisotropic solutions. We invite contributions on anisotropic mesh generation, adaptive refinement, and solver development, focusing on computational techniques that resolve anisotropic material properties and solution fields with high accuracy and efficiency. Emphasis is placed on space-time finite element or isogeometric frameworks for evolving anisotropic phenomena, advanced anisotropic error estimation, and anisotropy-aware linear solvers and preconditioners.
Keywords:
a posteriori error estimation, anisotropy-aware solvers, convection-dominated flows, isogeometric analysis, spaceâtime finite elements, anisotropic mesh adaptation
This minisymposium explores the evolving landscape of neural network-based methods for solving partial differential equations (PDEs), focusing on the interplay between data-driven techniques and classical numerical analysis. While approaches such as physics-informed neural networks (PINNs) [1] and neural operators [2] have gained attention for combining machine learning with physics-based modeling, practical deployments have revealed limitationsâincluding slow convergence, sensitivity to problem settings, and limited adaptability across PDE classes.
To address these challenges, recent research has focused on integrating rigorous mathematical insights into neural network solvers. Contributions in this session will showcase a range of methods, includingâbut not limited toâapproaches that incorporate adaptive sampling, structure-preserving architectures, and training strategies to enhance stability and scalability. Attention will also be given to algorithmic innovations that reduce computational cost, enable real-time inference, and improve performance in low-data regimesâfor example, in tasks such as sensor placement and inverse problem solving.
Combining theoretical developments and application-driven case studies, this session brings together researchers from applied mathematics and computational science and engineering to explore how blending numerical PDE methods with neural architectures can push the frontiers of scientific computing.
Keywords:
neural operators, physics-informed machine learning, scientific machine learning, Neural PDE solvers
Isogeometric Analysis (IGA) is a modern advancement in computational analysis that aims to bridge the gap between design and analysis by using the same mathematical representation for both CAD and CAE models. This seamless integration makes IGA an ideal framework for design optimization, facilitating smooth incorporation with manufacturing and fabrication processes. Consequently, IGA has seen widespread application in addressing diverse design challenges involving novel materials, such as metamaterials, smart materials, and functionally graded materials, as well as advanced structures including beams, plates, and shells composed of these materials (FGMs). These optimization problems often involve high nonlinearity, multi-physics interactions, intricate cross-scale coupling and scale dependencies. Hence, this mini-symposium (MS) aims to bring together researchers working on IGA for the optimal material and structural design to exchange ideas, present novel developments, and discuss cutting-edge advances that can address the aforementioned challenges. The MS welcomes, but is not limited to, contributions in the following focused areas:
⢠Development of novel isogeometric optimization frameworks for additive manufacturing.
⢠Data-driven and machine-learning frameworks for isogeometric optimal design.
⢠Optimization of smart/functionally graded/architected materials and structures.
⢠Efficient numerical solvers for concurrent multi-scale simulation and optimization.
⢠IGA-based computational homogenization algorithms.
⢠Isogeometric shape, topology, and material optimization of plates and shells.
⢠Post-buckling optimization strategies for thin-walled structures.
⢠Optimization of micro- and nano-structures accounting for size-dependent effects.
⢠Nonlinear, multi-physics, and multi-material optimization problems addressed through IGA.
⢠Optimized design of materials and structures for emerging applications, such as healthcare, energy harvesting, and energy absorption.
Keywords:
Optimal material design, Structural optimization.
, Isogeometric Analysis
This mini-symposium is intended as a forum for exchange and debate on advanced numerical methods for tackling problems in engineering and applied sciences that involve multiscale and multiphysics couplings.
The aim is to bring together scientists (engineers, physicists, mathematicians) focusing on coupling techniques in numerical modelling in fields ranging from computational mechanics to acoustics, electromagnetism and computational chemistry. Papers may cover a wide range of aspects from modelling to numerical strategies. The main focus will be on computational issues and associated advanced methods, while highlighting the underlying conceptual and theoretical foundations.
Application topics will include (but not be limited to):
⢠Heterogeneous and multiscale media;
⢠Complex materials;
⢠Computational homogenization;
⢠Localized phenomena and singularities;
⢠Multilevel or multi-model strategies;
⢠Hierarchical methods;
⢠Multiphysics coupling algorithms;
⢠Interface and contact problems;
⢠âŚ
Keywords:
Computational Mechanics, Coupled problems, Multiphysics;, Multiscale;, Numerical methods;
During the past decades of the last century researchers concentrated on optimizing mesh-based methods for the solution of partial differential equations in applied mathematics and engineering, e.g. error controlled adaptive meshing, sophisticated constitutive modeling. Beginning of the recent century people recognized, that despite all progress in computer resources and in the algorithmic treatment of discretized systems, for example so called multi-query problems could never be treated with classical mesh-based methods. The success-story of model-order reduction (MOR) has been initiated. Examples for multi-query problems are for example optimization and inverse problems, parametric and uncertainty propagation as-well-as multi-scale problems in space and time.
Nowadays a bunch of methods are available, like proper-orthogonal decomposition (POD), proper-generalized decomposition (PGD) or machine-learning (ML) techniques. In general, they can be distinguished in intrusive and non-intrusive methods, depending on the specific coding effort. Nonetheless, a challenge remains on the efficient treatment of non-linear and high-dimensional problems. Thus, the focus of this mini-symposium will be on latest developments and sophisticated application of MOR for non-linear and high-dimensional problems in solid and fluid mechanics, including methods augmented with experimental data (data augmented simulations (DAS)).
Keywords:
Model Order Reduction
In recent decades, the frequency and severity of extreme, multi-hazard natural events have significantly increased. This escalation has become especially pronounced in recent years and is projected to intensify further due to climate change. Many of these hazards are closely linked to hydrological processes, including floods, mudslides, landslides, avalanches, and tsunamis. Advances in numerical methods, together with a growth in computational power, have driven the widespread adoption of simulation tools to better understand and predict these complex events. Techniques such as shallow water models, advanced finite element and finite volume schemes, and in particular particle-based methods, including Smoothed-Particle Hydrodynamics (SPH), Discrete Element Method (DEM), Material Point Method (MPM), and Particle Finite Element Method (PFEM), enable coupled simulations of natural hazards. Their efficient parallelization on CPUs and GPUs further supports large-scale, three-dimensional modelling efforts.
This thematic session aims to showcase recent progress in the numerical simulation of natural hazard initiation and dynamics, fostering collaboration and dialogue among researchers and practitioners, exploring and discussing how fully three-dimensional methods (or advanced numerical methods) can both complement and ultimately transition beyond classical one-phase depth-averaged approaches in hazard assessment. While emphasizing hydrological hazards, the session welcomes contributions on numerical modelling of a broader range of natural phenomenaâgeological and meteorological alikeâwith particular interest in multi-hazard interactions (e.g., landslides triggered by earthquakes, tsunami waves induced by landslides). Additionally, studies addressing fluid-structure and fluid-soil-structure interactions, reflecting the interplay between natural hazards and civil infrastructure, are highly encouraged.
Keywords:
3D modeling, fluid-structure interaction, multi-hazard, Natural Hazard
The Earth system presents some of the most demanding challenges for simulation and prediction, spanning a vast range of scales in space and time, involving tightly coupled multiphysics processes, and requiring integration of massive observational datasets. As climate, seismic, hydrological, and subsurface models evolve toward higher fidelity, the need for extreme-scale computing and advanced numerical methods becomes more pressing. This minisymposium shall connect researchers at the intersection of geosciences, computational mathematics, and high-performance computing to present recent advances and emerging trends.
We aim to highlight progress in areas including (but not limited to) discretization techniques, robust solvers for multiphysics and multiscale systems, multigrid and domain decomposition methods, uncertainty quantification, and machine learning integration in forward and inverse problems. Of particular interest are innovations enabling simulations on state-of-the-art exascale architectures, e.g., via performance-portable implementations, improved algorithms or simulation software design.
Applications span the solid Earth, ocean and atmosphere dynamics, ice sheet evolution, groundwater flow, and coupled Earth system modeling. We welcome contributions focused on algorithmic development, software frameworks, mathematical analysis, and interdisciplinary case studies that push the boundaries of what is computationally and scientifically feasible in the geosciences.
Keywords:
Computational Earth Science, Computational Geoscience, High Performance
Computing, Simulation Software
This minisymposium brings together researchers from mathematics and engineering to share recent advances in numerical methods for neutral and charged particle transport, radiative transfer, and the Boltzmann equation. These problems, arising in diverse areas in science and engineering such as nuclear engineering, astrophysics, plasma physics, and medical imaging, are often tackled using a broad spectrum of methods tailored to specific community needs. Approaches include space-angle Discontinuous Galerkin methods, discrete ordinates methods, spherical harmonics methods and related moment closure methods, and integral equation methods, among others.
By fostering interdisciplinary exchange, this minisymposium aims to highlight both theoretical and computational developments that address common challenges across these fields. Particular focus will be placed on:
⢠Efficient and robust solvers, including advanced preconditioning strategies, low-rank and hierarchical methods, and error-controlled computation;
⢠Innovative discretization techniques, such as polygonal meshes and high-order schemes;
⢠Eigenvalue problems related to stability and criticality in transport models.
Contributions showcasing cross-cutting insights or comparative studies of numerical techni-ques across different applications are especially encouraged. The minisymposium also welco-mes work at the intersection of scientific machine learning and transport phenomena, as well as survey talks reviewing major recent advances. By bridging communities and methods, this event aims to foster collaboration and stimulate new directions in computational transport.
Keywords:
Boltzmann equation, Eigenvalue problem, low rank method, Radiative transfer, Neutron transport
Digital Twins (DTs) are an emerging technology that promises to reduce CPU time and enable multi-query analyses for a variety of multi-scale and multi-physics problems. As such, coupling solution techniques often serve a critical role in both the development and online deployment of DTs.
One source of improvement in computational efficiency is the invocation of data-driven methods. However, they are still largely used as stand-alone simulation tools and their coupling to conventional methods and other data-driven methods for multi-scale, multi-physics simulations remain underdeveloped both mathematically and algorithmically. The goal of this session is to bring together researchers working on mathematical and software challenges involved in the rigorous and agile coupling of arbitrary combinations of data-driven and conventional methods.
In scope and of interest to this minisymposium are traditional coupling techniques between full-order models as well as coupling techniques between surrogate subdomain models, both intrusive and non-intrusive. Additionally, coupling techniques between either full- or reduced- order models that treat the interface coupling operator itself as a surrogate are additionally of interest, as well as software that supports the efficient implementation of these various methods.
Keywords:
data-driven models, interface-coupled multiphysics, Lagrange multipliers, optimization-based coupling, partitioned schemes, reduced order models, surrogate modeling
This mini-symposium will emphasize a range of Kernel and machine learning-based approaches, which can be applied to the solution of partial differential equations (PDEs) in science and engineering. Contributions dealing with practical applications are encouraged, such as in mechanics, civil engineering, aeronautics, bio-medicine, transport and sensing of pollutants, materials design and processing, remote sensing, non-destructive evaluation, meta-models for high-dimensional problems, etc. Papers on other subjects related to the themes of this symposium are also welcome.
Keywords:
AI
Although physical simulation has become indispensable in scientific and engineering design and analysis, many real-time, many-query decision-making, and uncertainty quantification tasks remain computationally prohibitive with classical high-fidelity methods. Recent advances in artificial intelligence and machine learning have opened new frontiers in model order reduction (MOR), providing scalable, data-driven approaches that dramatically reduce computational costs while preserving accuracy and generalizability. These innovations are particularly powerful for parametrized systems governed by complex nonlinear partial differential equations, where they enable efficient surrogate modeling, rapid multi-query evaluations, and real-time decision-making.
This minisymposium will highlight recent progress in MOR, with a special emphasis on the synergistic integration of physics-based techniques and AI/ML methodsâincluding physics-informed learning, neural operators, manifold learning, and hybrid approaches that embed domain knowledge. Topics of interest include, but are not limited to: nonlinear approximation techniques; model reduction in high-dimensional parametric spaces; hyper-reduction for nonlinear problems; adaptive and error-controlled MOR strategies; structure-preserving MOR approaches; and machine-learning-enhanced surrogate modeling. Applications span optimization, feedback control, uncertainty quantification, and inverse problems in computational physics and engineering.
By bringing together researchers working at the intersection of MOR, physics-based simulation, and machine learning, the minisymposium aims to foster cross-disciplinary dialogue, share emerging methodologies, and critically assess the role of AI in advancing the scope, scalability, and reliability of reduced-order modeling.
Keywords:
AI for Science, Computational Mechanics, Computational Methods, data-driven methods, Large-Scale Problems, Machine Learning, Model Order Reduction, partial differential equations, physics-informed machine learning, reduced order models
Architected materials and structures have gained increasing attention for their unique properties and broad industrial applications. Additive manufacturing now enables the precise fabrication of well-controlled architectures, such as periodic or quasiperiodic lattice structures, allowing, among others, for lightweight, stiff, stretchable, or multifunctional components. However, their complexity poses challenges for numerical simulation. While homogenization-based methods are common, they may face limitations, particularly due to insufficient scale separation in practice. As a result, dedicated efficient strategies based on fine-scale modeling and simulation, i.e., involving the full discretization of the architectural geometry, are emerging. This mini-symposium aims to gather contributions focused on the modeling, simulation, and optimization of architected materials and structures using advanced fine-scale-based methods. Topics of interest include, but are not limited to: (1) leveraging advanced computational techniquesâsuch as isogeometric analysis, beam and shell models, immersed methods, or (machine learning-based) surrogate modeling approachesâfor accurate and efficient simulations; (2) analyzing nonlinear or inelastic responses and stability; (3) developing digital twins that account for process-induced properties and defects; and (4) integrating simulations and experiments to validate models and better understand the underlying mechanical behavior.
Keywords:
beam and shell modeling, Cellular materials, fictitious domain methods, lattice structures
The rapid growth of hardware parallelism, particularly with the rise of GPUs, has opened new avenues for efficiently solving large-scale linear systems. However, the optimal use of
performance capabilities of modern architectures remains challenging due to communication bottlenecks that limit the scalability of classical iterative solvers at extreme scales.
Among the most effective strategies to address these challenges are domain decomposition and multigrid methods, each with a long history of development. Domain decomposition techniques partition global problems into localized subproblems that can be solved separately, enabling high parallelism and reducing global communication. Multigrid methods â including geometric, algebraic, and polynomial variants â accelerate convergence by addressing errors across multiple scales through hierarchies of discretizations, often achieving optimal or near-optimal computational complexity. However, each concept faces potentially severe limitations: domain decomposition may suffer from slow convergence due to frequent interface data exchange and load imbalance, while multigrid methods can struggle with complex geometries, irregular meshes, or highly anisotropic coefficients.
In recent years, systematic efforts to combine these two classes of methods â such as using multigrid as a preconditioner within domain decomposition frameworks, applying domain decomposition smoothers in multigrid hierarchies, or developing hybrid solvers tailored to large-scale, heterogeneous systems â have gained increasing attention. Advances in both domain decomposition and multigrid techniques, whether developed independently or in combination, offer powerful and complementary tools for building robust and scalable solvers for complex physical problems encountered in science and engineering.
This minisymposium aims to bring together various strains of developments for scalable
solvers, inviting novel research on both domain decomposition and multigrid methods as well as their combination. Topics include, but are not limited to, heterogeneous coupling strategies for coupled, multiphysics, and multiscale problems; communication avoiding techniques; and approaches for optimizing performance on modern computing architectures.
Keywords:
Coupled problems, Domain Decomposiiton, Heterogeneous Computing, Iterative solvers, Large-Scale Computing, Multigrid, Scalability
Nonlinear dynamical systems are ubiquitous in both engineering and the physical sciences. They are characterised by complex behaviours such as multi-stability, bifurcations, and chaos. Continuation methods have emerged as powerful tools for exploring the various solutions of dynamical systems as parameters are varied. These techniques enable the computation and tracking of equilibria, periodic and quasi-periodic responses, as well as bifurcation points, allowing for the identification of emerging solution branches and critical thresholds where the system exhibits both qualitative and quantitative changes.
This minisymposium aims to bring together researchers and practitioners working on recent developments in continuation methods and their applications to nonlinear dynamics. We welcome contributions that advance the theoretical foundations, numerical algorithms, and computational tools for continuation analysis, as well as applications in solid and fluid mechanics, multiphysics systems such as fluid-structure interactions or electromechanical systems, biological models, and related areas.
Keywords:
Bifurcation Analysis, Continuation methods, Large-scale modelling, : Dynamical Systems, reduced order models
This Minisymposium aims at exploring the theoretical frameworks, relationships and integration strategies between classical local continuum methods and emerging nonlocal computational models, such as peridynamics and phase-field. While nonlocal models have found extensive application in fracture and damage mechanics, understanding their theoretical foundations, performance characteristics, and their possible combination with local models opens new possibilities in computational mechanics. This Minisymposium invites contributions that evaluate, apply, and compare these methods across a wide variety of application areas, including corrosion and material degradation, heat and mass transport in heterogeneous media, biomedical systems (e.g., bioresorbable implants), and evolving microstructures. Particular attention will be placed onto innovative computational and algorithmic approaches that allow for robust connections, integrations, and unification frameworks for local and nonlocal modeling approaches. Contributions involving computational efficiency, scalability in high-performance computing (HPC) environments, and the application of artificial intelligence and machine learning to facilitate or enhance these connections are especially welcome. This Minisymposium intends to stimulate interdisciplinary discussion among researchers from computational mechanics, applied mathematics, materials science, and biomedical engineering, highlighting shared computational challenges and collaborative opportunities.
Keywords:
Computational Methods, High-Performance Computing, Nonlocal mechanics, Peridynamics, Phase-Field Modeling
Mathematical Modelling and Simulation for Social, Environmental, and Disaster Prevention Issues
TRACK NUMBER
SHINOBU YOSHIMURA 1, DAIGORO ISOBE 2, HIDEKI FUJII 1 AND EISUKE KITA 3
1 University of Tokyo, Hongo, Tokyo 113-8456, Japan
2 Tsukuba University, Tsukuba 305-8577, Japan
3 Nagoya University, Nagoya 464-8601, Japan
* Email: kita@i.nagoya-u.ac.jp
Keywords: Mathematical Modelling, Computer Simulation, Structural Engineering, Fluid Engineering, Machine Learning, Traffic Engineering, Environmental problem, Disaster prevention, Agriculture, Economic problem
ABSTRACT
Large -scale disasters have a great influence on social infrastructure such as buildings and roads, and also change the industrial structure of agriculture. The issues related to society, environment, and disaster prevention are large -scale and complex phenomena. To analyse these phenomena, it is necessary to simultaneously develop the mathematical modelling and simulation technique at the same time. In particular, the progress of machine learning methods is remarkable. These techniques also have a new solution to the problem handled in this session.
This session is planned to exchange opinions on modelling methods and simulation methods by related researchers. We will expect a large number of researchers who are interested in the related fields.
Keywords:
Agriculture, Computer Simulation, Disaster prevention, Environmental problem, Fluid Engineering, Machine Learning, Structural Engineering, Traffic Engineering, Mathematical Modelling
The advent of machine learning (ML) is actively revolutionizing the realm of computational solid mechanics, creating a platform for the development of novel methods with unprecedented efficiency. Nonetheless, several limitations still hinder the direct incorporation of purely ML-based methods in industry-level engineering problems, including the need to generate vast datasets, their limited generalizability beyond the training regime, their numerical sensitivity to the choice of hyperparameters, and more. To this end, there is a growing momentum towards hybrid modeling paradigms which integrate ML-based techniques with conventional numerical methods. The aim of these hybrid frameworks is to capitalize on the strengths of both worlds: the robustness and rigor of conventional methods (such as FEM, BEM, VEM, etc.) and the flexibility and speed of machine learning, ultimately targeting large-scale, multi-physics, real-world solid mechanics problems. In this minisymposium, we invite new contributions focused on hybrid solvers combining ML tools with existing numerical methods, topics of interest include but not limited to:
⢠Novel combinations of ML-tools (networks, operators, etc.) with the Finite Element Method, Boundary Element Method, Virtual Element Method, meshless approaches, and more.
⢠Fusion of ML-methods into multi-scale (FE2) modeling, to accelerate solid mechanics simulations across different length scales.
⢠Efficient and robust pathways to extrapolate the ML predictive capability beyond the bounds of the training regime (unseen geometries, loading conditions, timeframes, etc.).
⢠Challenges and opportunities in the utilization of ML-methods for the solution of coupled multi-physics problems, including fracture, corrosion, thermo-hydro-mechanic-chemical coupling, and more.
⢠Complementing numerical methods of failure analysis with physics-augmented machine learning, with applications on phase-field fracture, continuum damage, peridynamics, cohesive zone models, etc.
⢠Stochastic analysis and uncertainty quantification of such hybrid methods.
Keywords:
Machine Learning, Multi-physics, Numerical Modeling, Coupled problems, Hybrid Methods, Methods for fracture and failure, Solid Mechanics
The objective of this Mini Symposium is to open space for researchers in Boundary Elements to report current advances in these topics as well as the coupling with other numerical methods, particularly the Finite Element Method. The Boundary Element Method is an efficient numerical tool to model the dynamic of unbounded domains and when coupled with the FEM allows to model the dynamics of complex systems, including vibrations of eolic systems, constituted of blades, rotor, structure, foundation and the supporting soil. The modelling of periodic structures, which produce bandgaps, embedded in various kinds of soil profiles and interacting with engineering structures represents another example of current BEM application. The BEM allows the modelling of vibration and waves propagating from the complex systems into the environment, allowing to help assessing the influence of the engineering systems on the surrounding ecosystem. The MS will consider theoretical developments, new numerical strategies as well as engineering applications based on these methodologies. The coupling of Boundary Elements with other numerical methods are welcome. Applications related to acoustics and fluid dynamics as well as geomechanics problems in unbounded domains are encouraged.
Keywords:
BEM-FEM Coupling, Boundary Element Method, Dynamics of Complex Systems
Dealing with Uncertainty Quantification, monitoring, and data assimilation in computational mechanics requires the fusion of several strategies for the sake of computational efficiency. Physics-Enhanced Machine Learning [1] (PEML, also referred to as scientific machine learning) nowadays provide a wealth of tools to enhance the integration of information that is typically extracted from real-world data, physics-based models and domain and expert knowledge, thus representing an overarching paradigm when dealing with complex systems in computational mechanics.
Recent advances in machine learning, among many algorithmic methods of PEML, have allowed to overcome several bottlenecks often hindered by high dimensionality and significant complexity, opening new horizons for data-driven predictive modelling in computational mechanics, and impacting on efficient numerical strategies often used in UQ, such as surrogate models and reduced order models. However, several issues are still open, such as the integration of learning algorithms within UQ techniques, the use of learning paradigms naturally accounting for uncertainties such as Bayesian and kernel methods, the construction of reliable and robust strategies to handle large-scale problems such as multi-fidelity methods for data fusion, the setting of data assimilation procedures relying on latent dynamics modelling capable to adapt in rapidly evolving phenomena [2].
This minisymposium will gather a broad spectrum of contributions in this very vibrant research area, covering the theoretical analysis, computational techniques, and practical use of machine/deep learning, data assimilation, filtering, monitoring, and UQ techniques, as well as their interaction, all towards efficient and accurate predictions in computational mechanics.
REFERENCES
[1] A. Cicirello, Physics-Enhanced Machine Learning: a position paper for dynamical system investigations, Journal of Physics: Conference Series, 2909 (2024), 012034.
[2] L. Rosafalco, P. Conti, A. Manzoni, S. Mariani and A. Frangi, Online learning in bifurcating dynamic systems via SINDy and Kalman filtering, Nonlinear Dynamics, 113 (2025), 14201-14221.
Keywords:
Computational Mechanics, Monitoring, Physics-enhanced Machine Learning, Uncertainty Quantification
Multidisciplinary design analysis and optimization plays a vital role in the transformation of design processes in aeronautics, ground/water transport or the energy sector. Large design spaces and/or high-resolution numerical models call for gradient-based optimization and adjoint methods. Uncertainty handling is crucial in the context of robust design. Algorithmic and parallel/HPC efficiency, interoperability/extensibility and usability of the software solutions are key scalability aspects for large coupled high/mixed-fidelity problems â e.g. CFD-CSM simulations. Multi-fidelity methods, efficient surrogate models or machine-learning/data-driven techniques can accelerate the design evaluation.
This minisymposium aims to review the state of the art in integrating the disciplinary advances into large scale, complex, and/or coupled industrial scenarios.
Particularly, contributions on the following topics are invited:
⢠emerging frameworks, packages and libraries for large-scale MDAO
⢠scalable and robust solution and optimization algorithms
⢠gradient/adjoint-enabled solutions
⢠geometry modelling for MDAO
⢠data-driven techniques
⢠MDAO for transient/dynamic and periodic systems
⢠industrial MDAO applications, benchmark and verification studies
Keywords:
coupled adjoints, data-driven methods, MDAO frameworks, sensitivity analysis, uncertainty quantification (UQ), multidisciplinary design analysis and optimization (MDAO)
Simulation-based engineering and science, including topology design, parameter optimization, and uncertainty quantification, are usually characterized by high-dimensional systems that require immense computational costs. Despite significant advancements in computer hardware and the development of GPU parallel computing over the past few decades, real-time simulations of large-scale, high-dimensional systems remain intractable with conventional methods (e.g., finite element analysis). Regarding that, model order reduction/reduced order modeling methods offer efficient and powerful tools for addressing large-scale, high-dimensional problems by reducing the size of the computational models while maintaining accurate solutions in a fast and efficient manner.
This mini-symposium provide a platform for researchers to exchange insights on the development and applications of model order reduction methods. Topics of interest include but not limited to: model order reduction techniques for high-dimensional problems; low-rank representations, tensor decomposition and proper generalized decomposition; space-time-parameter reduced-order methods; nonlinear model reductions; error estimates and parameter adaptivity; implementation on large-scale engineering problems, topology optimization, uncertainty quantification, operator learning and inverse problems. Additionally, potential topics may include integrated machine learning methods and data-driven simulations.
Keywords:
High-Dimensional Problems, Model Order Reduction/Reduced Order Modeling, Space-Time-Parameter
This symposium focuses on the latest developments regarding space-time methods for numerical simulation and modelling in engineering. These methods have gained interest in recent years with works exploring different aspects and new concepts. These recent works explore the mathematical aspects (from the numerical analysis point of view), the formulation of space-time problems (with Hamiltonian-like approaches), discretisation technics (continuous vs discontinuous Galerkin, IGA, VEMâŚ), and numerical aspects (mesh refinement, domain decomposition, space-time integration, matrix-free approachesâŚ). Such topics are the most representative examples of the recent literature.
Some of the advantages of Space-Time approaches over more classical ones are well known: they open the door to space-time parallelization or space-time adaptivity, they enable to deal with global optimisation or inverse problems in a more direct manner, h- and p- convergence in space and time can be shown to be optimal at least for simple problems, etc. In addition, they can also open up new possibilities for multiphysics and non-linear problems for instance dealing with different time scales in a space-time multi-grid approach just by playing with the different grid sizes or with the approximation functions.
The topics covered by this MS include (but are not limited to):
⢠Mathematical aspect of space-time methods;
⢠Space-time discretization technics;
⢠Space-time parallelization and numerical performance;
⢠Space-time for multiphysics and/or non-linear problems;
⢠Space-time for structural problems;
⢠Numerical and algorithmic aspects.
Keywords:
Coupled problems, Dynamics, Space-Time formulations, ST-FE, ST-IGA, Time-decomposition, Time-parallelization
The brain is a complex and dynamic organ composed of soft tissue, neuronal and glial networks, blood vessels, cerebrospinal fluid, and interstitial space. The coupling of electrical, chemical, and mechanical processes across a wide range of spatial and temporal scales plays a central role in both healthy brain function and the progression of neurological diseases. Capturing this interplay in computational models poses significant scientific and technical challenges but also holds great promise for advancing our understanding of brain physiology through in-silico experimentation. This minisymposium brings together researchers in applied mathematics, computational science, biomedical engineering, and neuroscience to discuss recent advances in brain multiphysics modeling. Topics include but are not limited to: mechanics of brain tissue and cerebrospinal/interstitial fluids, molecular transport and waste clearance (including glymphatic and perivascular pathways), electromechanical coupling, and complex network dynamics. Contributions proposing innovative computational strategies are particularly welcome, such as efficient solvers for coupled systems, physics- and data-based model reduction, and data assimilation methods for imaging or clinical data. Applications to neurological, neurodegenerative, and neurovascular disorders are strongly encouraged.
Keywords:
Brain Electrophysiology, Computational Fluid Dynamics, Neurodegenerative diseases, Numerical Methods for PDEs, Poromechanics, Scientific Machine Learning
Computational electromagnetism plays important roles both to design the electric facilities and to assess the influence of electromagnetic fields; for example, transformer, motor, burn wound, and hyperthermic potentiation. However, as the ability of computers progresses and the demand of more precise approximation, the number of Degrees Of Freedom (DOF) of computational models derived from conventional numerical schemes becomes larger even in case of adaptive mesh refinements. In this mini-symposium, we discuss on the accuracy and efficiency of novel computational methods for electromagnetic field problems from mathematical and engineering points of view. First, we discuss efficient numerical schemes to compute directly such large-scale computational models within the required computational costs, for example, based on Domain Decomposition Methods (DDM) with parallel computations; see [1]. Second, we discuss new approaches based on machine learning and artificial intelligence. Third, we welcome to discuss novel schemes not mentioned above. Moreover, we also consider other related problems as well as electromagnetic field problems because we encounter a kind of interaction problems with electromagnetic field ones in practical situations; see [2].
Keywords:
computational electromagnetism, novel numerical schem, numerical analysis
Reduced Order Methods (ROMs) aim to build surrogate models to simplify complex parametric coupled systems, generating efficient and reliable approximations, especially in real-time simulations and many-query contexts. Intrusive approaches, such as Galerkin projection-based methods, involve the knowledge of the system equations to solve the reduced model. These methods are highly accurate, but they usually lack efficiency in terms of computational time when dealing with complex dynamics, such as non-linear problems and complex coupled systems. Non-intrusive methods, commonly based on data-driven and machine learning techniques, only exploit information from measurements or simulations extracting unseen patterns while efficiently handling a wide range of applications. Despite this, they may require a large amount of data, long training times to achieve acceptable accuracy, and they usually lack error certification, physics consistency, and structure-preserving properties. Bridging data-driven techniques and physics-based approaches enhances the modeling capabilities and the interpretability of the models, enabling efficient and accurate predictions even for complex systems.
The goal of this mini-symposium is to foster idea exchange through a comparative analysis of the applicability of intrusive and non-intrusive ROMs, illustrating their practical benefits and limitations across a wide range of academic, industrial, and engineering applications.
Keywords:
Parametrized PDEs
Computational models play a central role in the simulation, design, and control of complex engineering systems. As these systems grow in complexity, there is a growing demand for efficient and accurate numerical methods; furthermore, real-time prediction, multi-query scenarios, and large-scale optimization and control remain challenging for conventional high-fidelity solvers due to their prohibitive computational cost. Reduced-order models (ROMs) offer a promising path forward by providing significant computational savings while preserving essential features of the underlying high-dimensional systems.
This minisymposium aims to explore recent advances in high-order discretization techniques and reduced-order modelling approaches, with a focus on challenging applications in computational fluid dynamics, structural mechanics, and multiphysics problems. Of particular interest are high-order methods that offer superior accuracy per degree of freedom, and model order reduction strategies that remain robust over wide parameter ranges or preserve critical structures such as conservation laws, symmetries, positivity or entropy.
Topics of interest include, but are not limited to: projection- and collocation-based model reduction techniques for parametric systems; hyper-reduction strategies; structure-preserving high-fidelity numerical schemes and ROMs for nonlinear PDEs; high-fidelity discretization tailored for reduced-order modelling; data-driven and hybrid machine learning-ROM and deep-learning approaches; adaptive and error-controlled model reduction; applications to control, uncertainty quantification, inverse problems, and digital twins.
Bringing together researchers working at the interface of numerical analysis, scientific computing, and engineering applications, this minisymposium will provide a forum for the presentation and discussion of recent methodological developments, benchmarking efforts, and cross-disciplinary applications. A key objective is to foster a shared understanding of the respective roles and limitations of high-fidelity and reduced-order models, promoting their synergistic integration in simulation pipelines. This perspective is essential to enable the next generation of predictive, efficient, and adaptive computational methods.
Keywords:
Modeling and Simulation of Complex Engineering Systems, Multi-Physics Simulations, Reduced Order Modelling (ROM), Structure-Preserving Algorithms, High-Order Numerical Methods
Soft materials are materials of choice in diverse modern technologies with applications including tissue adhesives, bioelectronics, and soft robots, as well as in traditional technologies such as tires and vibration isolators. In all these technologies, the mechanical and physical properties of soft materials play an important role, which are governed by phenomena spanning from the nanometer single-chain level, the mesoscale network level, to the macro-scale bulk material level. Specifically, large deformations coupled with various multi-physical phenomena and instabilities at different length scales open an immensely rich research arena for computational mechanics approaches. Additive and multi-material manufacturing adds another layer of complexity for problems in soft materials. Moreover, soft materials represent essential components in biological tissues, a topic of extreme interest for biomedical applications. These problems require the development of a unique suite of computational approaches that have to deviate from their traditional counterparts in hard materials due to the inherent nonlinearities in the response of these materials. This mini-symposium aims to bring together scientists and engineers working at the forefront of soft matter modeling to exchange and share their experiences and recent research results. Topics of interest include, but are not limited to:
⢠First-principles modeling and multiscale methods: Ab-initio simulations; molecular or micro to macro homogenization/upscaling/coarse-graining methods;
⢠Data-driven and machine learning methods: supervised and unsupervised learning of constitutive models from experimental data; novel neural network architectures for solving equilibrium equations; surrogate modeling;
⢠Multiphysics modeling: computational methods for coupled electro-mechanical-chemical problems, bio-inspired materials, active soft materials; methods for complex applications such as implants, wearable devices, 3D printing; methods for growth and remodeling in tissues; inverse methods for design of multiphysics systems;
⢠Adhesion, friction, contact problems in soft matter: instabilities and failure in thin films and interfaces; surface tension and capillary effects; novel contact algorithms for highly deformable materials;
⢠Fracture in soft matter: novel computational method development; homogenization of fracture; phase-field methods
⢠Integration of computational methods with experimental methods in soft matter.
Keywords:
Biological materials, Computational Mechanics, Soft materials
Batteries are essential in numerous applications, such as mobile devices or electric vehicles, and are a key enabler for the transition towards renewable energy. Recent developments of novel cell technologies, such as solid-state batteries or sodium-ion batteries, promise significant improvements in energy density, safety, or sustainability. These are key criteria for e.g. large-scale stationary storage or even new technology fields such as electric aircraft. The development and design of new energy materials, electrodes, and cell designs significantly benefit from predictions and guidelines provided by computational approaches from atomistic to continuum scale. However, significant challenges remain for the predictive simulation of electrochemical energy storage. Particularly, schemes passing information from atomistic simulations on the material level to continuum simulations for cell-level performance predictions, and backward feeding information from continuum level to meso- and atomic-scale for relevant boundary and real-life cell cycling conditions. Moreover, novel cell technologies use materials with large volume expansion or solid electrolytes, rendering mechanical aspects critical for the performance and degradation of those systems. Therefore, thermo-electro-chemo-mechanical models and efficient numerical solvers are required for battery design and optimization. Finally, integrating batteries into the application and their monitoring and control requires computationally efficient yet accurate tools for state-of-charge and state-of-health estimation. Machine learning techniques promise significant advances in this field, yet superior performance and reliable control in critical applications are still to be demonstrated.
This symposium addresses the development of computational tools from atomistic to continuum scale and their application for analyzing, designing, and monitoring current and next-generation batteries. Special emphasis is placed on the development of coupled mechanical-electrochemical models and implementation of multiscale modeling frameworks that can systematically predict properties of electrochemical devices during operation. Contributions may cover but are not limited to the design of improved materials, hierarchically structured materials, imaging, characterization, and modeling of 3D structures on multiple scales, process-structure-property relationships, advanced physics-based modeling approaches
Keywords:
Computational materials science, Machine-Learning, Multiscale modeling, Thermo-electro-chemo-mechanics, AI/ML, Battery modeling
Quantum computing is considered a promising alternative to classical computational methods and a potential building block for expanding CSE capabilities to answer important future questions using simulation technology. Although industrial applications are still primarily based on classical simulation methods, some research teams have begun to build conceptual algorithmic bridges between the capabilities of new hardware and established simulation methods.
This mini-symposium brings together experts in flow simulation technology with developers of quantum algorithms for solving nonlinear fluid dynamic PDE problems. The goal is to exchange ideas and discuss the latest findings on efficient PDE solution methods based on (a) quantum and (b) quantum-inspired strategies, e.g. tensor network methods, as well as their hardware requirements.
Keywords:
Computational Fluid Dynamics, PDE-solvers, Quantum Algorithms, Quantum Computing, Tensor Network Methods
MS265 â Numerical Advances in the Solution of Saddle-Point System Arising from Fluid Flow Problems
Saddle-point systems lie at the heart of many numerical formulations of incompressible fluid flow, where they naturally arise from pressure-velocity coupling in the NavierâStokes equations and from constaint variational formulations. Efficient and robust solution strategies for such systems are crucial in a wide range of applications, from direct numerical simulations of turbulent flows to fluidâstructure interaction and multiphysics coupling. Saddle-point systems pose significant numerical challenges: they are typically large, sparse, indefinite, and sensitive to discretization choices, boundary conditions, and physical parameters.
This minisymposium highlights recent advances in the numerical solution of such systems, with emphasis on techniques motivated by or applied to fluid flow problems. Contributions span a spectrum of perspectives â from the development of robust and efficient preconditioners and iterative solvers to the design of coupling strategies within fluidâstructure interaction and immersed boundary frameworks. We aim to foster fruitful exchange between numerical analysts, applied mathematicians, and computational scientists working on fluid flow problems. This minisymposium will serve as a platform for sharing insights, identifying common challenges, and exploring promising directions for future research in the numerical solution of saddle-point systems.
Keywords:
fluid dynamics, numerical PDE, preconditioning, saddle-point systems
The modeling and simulation of solid materials across multiple scales remains a central challenge in computational science and engineering. From quantum-scale phenomena to macroscopic mechanical behavior, understanding the interplay between different physical scales requires not only sophisticated numerical methods but also rigorous mathematical frameworks.
This minisymposium focuses on the integration of multiscale modeling techniques and machine learning approaches for solid-state systems. We aim to explore how recent advances in machine-learned interatomic potentials, surrogate models, and physics-informed learning frameworks are accelerating the development of multiscale methods, enabling accurate and efficient modeling of complex materials with defects, interfaces, or evolving microstructures.
Particular emphasis is placed on bridging scalesâfrom atomistic to continuumâthrough approaches such as coarse-graining, atomistic-to-continuum coupling, and hybrid physicsâmachine-learning schemes. Contributions that highlight theoretical developments, algorithmic strategies, or application-driven insights are equally welcome.
By bringing together applied mathematicians, computational mechanicians, and materials scientists, this symposium fosters interdisciplinary dialogue aimed at advancing both the mathematical foundations and practical capabilities of multiscale modeling. We especially encourage submissions that demonstrate how machine learning can enhance traditional numerical models, improve data efficiency, or quantify uncertainty in multiscale simulations.
Keywords:
Computational Solid Mechanics, Data-driven Models, Machine Learning, multiscale methods
The accurate and robust simulation of compressible flows remains a cornerstone of computational fluid dynamics (CFD), with applications ranging from aerospace engineering to astrophysics, weather prediction, and beyond. Traditional numerical schemes have achieved significant success, but they often face limitations when applied to complex, multi-scale phenomena characterized by shocks, turbulence, and wave interactions.
â¨This mini-symposium highlights recent advances in high-order structure-preserving numerical methods for compressible flows, going beyond classical schemes. The focus is placed on modern approaches such as Continuous/Discontinuous Galerkin (CG/DG), Flux Reconstruction (FR), Residual Distribution (RD), and advanced Finite Volume (FV) and Finite Difference (FD) methods that embed essential structures like local conservation, entropy stability, positivity and kinetic energy preservation. Topics include innovative high-order schemes for unstructured meshes, well-balanced and asymptotic-preserving methods, adaptive and multi-resolution strategies. In particular, the mini-symposium welcomes contributions that extend classical schemes, including those based on point and cell-average representations, the use of generalized function spaces beyond polynomial approximations (e.g., radial basis functions, function-space summation-by-parts operators), and structure-preserving approaches to uncertainty quantification and stochastic modeling pushing limitations in turbulent, hypersonic, and reactive flow applications.
These topics align closely with the DFG Priority Programme SPP-2410 âHyperbolic Balance Laws in Fluid Mechanics: Complexity, Scales, Randomness (CoScaRa),â which advances modern solution concepts and numerical methods for multi-scale and stochastic models while preserving hyperbolic structure. The mini-symposium will showcase selected recent results from the SPP-2410 alongside contributions from the wider community, fostering exchange between academia and industry on high-order structure-preserving methods for compressible flows, with emphasis on accuracy, robustness, and practical performance.
Keywords:
Finite element, High-order methods, Structure-preserving, Computational fluid dynamics, finite volume
In many processes in industrial applications and natural sciences, the evolution of interfaces is of paramount importance. Examples occur in a wide range of research areas including multi-phase flows, crack propagation, fluid-structure interaction, solidification, crystal growth and biomembranes. The phase-field methodology is a powerful mathematical modeling approach for systems with moving interfaces like these. In the phase-field method, moving boundary problems are reformulated as PDEs on fixed domains in which the interface evolution is governed by a PDE of a scalar order parameter (the phase field). Phase-field models are diffuse-interface models meaning that the interface is a smooth region described by the smooth phase field.
The phase field method has favorable properties, such as a rigorous thermodynamical
structure and a physical interface description, but introduces new challenges for computations. Important challenges include the discretization of higher order spatial derivatives that typically occur in phase- field models, the design of thermodynamically stable numerical methods (both in space and time) and the treatment of a relatively sharp interface. This minisymposia is dedicated to modeling and computation with the phase-field method. We welcome talks on novel phase-field modeling approaches and numerical algorithms as well as applications in fluids, fracture, solids and biomechanics.
Keywords:
Fluid Mechanics, Fracture Mechanics, Multi-physics, numerical simulation, Phase Field
Meshfree and Other Advanced Numerical Methods for Applied Mechanics and Engineering Problems
TRACK NUMBER 1700
LIHUA WANG *, CHUANZENG ZHANG â
AND ZHENG ZHONG §
* School of Aerospace Engineering and Applied Mechanics, Tongji University
1239 Siping Road, 200092
lhwang@tongji.edu.cn
â School of Science and Technology, University of Siegen, 57076 Siegen, Germany
Center for Mechanics Plus under Extreme Environments, Ningbo University, 315211 Ningbo, China
c.zhang@uni-siegen.de
§School of Science, Harbin Institute of Technology, Shenzhen 518055, PR China
zhongzheng@hit.edu.cn
Keywords: Meshfree methods, particle methods, strong-form collocation, peridynamics, machine learning methods, advanced numerical simulations, applied mechanics and engineering problems.
ABSTRACT
Contributions in all subjects related to meshfree and other advanced numerical methods as well as their mechanics and engineering applications are welcome, which include but are not limited to the following topics:
ď Recent advances in meshfree methods, smoothed particle hydrodynamics, material point methods and other advanced numerical methods.
ď Recent advances in peridynamics and phase-field methods.
ď Recent advances in maching learning and other AI-based methods.
ď Applications of meshfree methods/machine learning and other numerical methods for the numerical simulation of advanced materials and structures, soft materials, inverse problems, fluid dynamics and fluid-structure interactions, geomechanics, large deformation and non-linear problems, multi-phase interactions, multi-field coupled problems, contact and impact, damage and fracture, static and dynamic structural responses, manufacturing processes, nano-mechanics, etc.
Keywords:
advanced numerical simulations, machine learning methods, particle methods, peridynamics, strong-form collocation, Meshfree methods
The increasing urgency of global environmental challenges, such as climate change, biodiversity loss, and resource depletion, require integrated and innovative approaches that combine engineering and mathematical sciences. Modern advances in control theory, optimization, modeling, numerical methods, and data-driven techniques offer powerful tools to understand, predict, and manage complex environmental systems. This minisymposium aims to gather interdisciplinary experts from academia and industry to showcase cutting-edge research and foster collaboration towards sustainable solutions.
We invite contributions addressing a wide range of topics, where engineering and mathematics can effectively contribute to environmental care and sustainability, including but not limited to:
- Mathematical and computational modeling of environmental processes;
- Optimization and control strategies for resource management;
- Data-driven and machine learning approaches for environmental monitoring and forecasting;
- Numerical methods for simulating ecological and physical systems;
- Innovations in agritech and sustainable agriculture to enhance productivity and reduce environmental impact;
- Conservation strategies supported by quantitative analysis;
- Circular economy models promoting sustainable resource reuse and waste reduction;
- Land protection and ecosystem restoration through advanced simulation and control techniques.
This minisymposium provides an interactive forum for exchanging novel ideas, presenting breakthrough research, and discussing emerging challenges. We particularly encourage submissions that push the boundaries of current knowledge and propose innovative methodologies to address real-world environmental problems.
By bringing together diverse perspectives, this event will foster vibrant discussions, encourage knowledge sharing, and promote partnerships aimed at accelerating the translation of scientific and technological advances into impactful environmental care practices.
Keywords:
Data-driven approaches, Mathematical modeling, Optimization and control, Environmental sustainability
Structural responses under extreme loading and environmental conditions, such as impact, penetration, explosion, high-speed machining, and manufacture and surface treatment under high temperature and high pressure, have been paid wide attention in the recent years because of the interesting and important phenomena involved and the great challenges to computer modelling and simulation. As localization, damage, fracture, fragmentation and phase transformation occur, the multi-scale and multi-physics phenomena should be fully considered, and new theories and numerical methods are needed to model and simulate the structural responses under extreme loading conditions in accurate and effective ways. This minisymposium aims at providing an opportunity for academic researchers and industrial engineers in the related fields to discuss the recent progress and to promote collaboration. Those who have been working on in the related fields are cordially invited to exchange their ideas and research results in this minisymposium. Presentations are solicited in all subjects related to the model-based simulations of structural responses under extreme conditions, which include but are not limited to the followings:
1) Development of advanced numerical methods, such as meshfree and particle methods, generalized finite element and isogeometric methods, and peridynamics for modelling and simulation of structural responses under extreme conditions
2) Machine learning, data-driven and related approaches for the problems with extreme conditions
3) Simulation-based disaster prediction and mitigation
4) Numerical methods and coupling algorithms for multi-scale and multi-physics processes
5) Large-scale and highly efficient computation for the problems with extreme conditions
6) Coupled Lagrangian-Eulerian and immersed methods for the problems with moving boundaries
7) Inverse solutions and optimization for the problems with extreme conditions
8) Numerical algorithm implementation and simulation software development
9) Verification and validation for computer modelling of extreme events
10) Other related subjects
Keywords:
Extreme Conditions, Modelling and Simulation, Structural Response
Prediction and control of noise and vibrations are one of the largest environmental challenges for railway exploration in urban areas. These phenomena affect the comfort and life quality of the inhabitants in the railway surroundings. Buildings in residential areas can also be affected by vibrations produced by nearby construction work (some examples: pile driving, compaction work, blasting, etc). This mini-symposium aims to collect numerical studies dealing with the generation, propagation and mitigation of noise and vibrations and its interference with nearby structures.
Keywords:
Construction activities, Noise and vibrations, Railway dynamics, Soil-structure interaction
The numerical solution of systems of PDEs, which describe the behavior of multi-physical phenomena and processes, typically requires the application of efficient linear and non-linear solvers. This is particularly relevant for problems that include a large number of fields, e.g., phase field modeling of grain growth and solidification of or reaction-diffusion in multi-component alloys. These problems require a high spatial resolution increasing the number of unknowns. Therefore, the application of direct (linear) solvers becomes too computationally expensive, and an iterative method has to be employed. However, a suitable preconditioner is essential for iterative solvers to converge within a reasonable number of iterations. Domain decomposition based preconditioners or multi-grid methods are suitable methods of choice, however, their performance, scalability and efficiency crucially depends, among other things, on the construction of the preconditioner. Furthermore, load balancing and updates of the domain decomposition are essential for scenarios, in which adaptive spatial discretization techniques are employed, while reutilization of preconditioners or nonlinear preconditioning techniques are of interest in coupled, nonlinear problems.
This minisymposium aims to explore the latest developments in iterative solvers and their associated preconditioners applied to multi-physics problems using monolithic and staggered approaches. Discussions will highlight large-scale simulations on state of the art high performance compute clusters using established software libraries like Trilinos, PETSc and deal.II.
Keywords:
High Performance Computing, Preconditioner, Coupled problems, Iterative Solver, Multi-physics
Interface problems play a fundamental role in a wide range of scientific and engineering fields, including materials science, fluid dynamics, and solid mechanics. They govern critical phenomena such as phase separation, wetting and dewetting of thin films, morphological evolution, defect dynamics, and the mechanics of biological and soft matter systems. In recent years, remarkable advances have been achieved in multiscale modelling, numerical methods, and theoretical analysis for these problems, enabling deeper understanding and predictive capability.
This minisymposium aims to bring together experts from diverse fields of mathematics, physics, engineering, and artificial intelligence to collaboratively explore innovative solutions for multiphysics and multiscale interface problems, promoting deep integration and synergistic development across interdisciplinary domains. Focusing on core challenges such as the high computational costs of traditional methods and difficulties in cross-scale modeling, the symposium is dedicated to facilitating in-depth discussion on innovative methodologies on modeling and numerics, breaking through the technical bottlenecks of conventional approaches, and providing solid theoretical foundations and technical support for applications in physics, engineering, biology, and related fields.
The symposium centers on cutting-edge challenges in multiscale modeling, numerical simulation and analysis for complex interface problems. The topics of discussion include, but are not limited to:
1. Computable modeling for multiphysics and multiscale interface problems;
2. Advanced paradigms for numerical methods and algorithms;
3. Interdisciplinary applications and practices.
Keywords:
Crystal Defects, Multi-phase Flows, Multiscale Modeling, Numerical Analysis, Biological and Soft Matters, Interface Problem
Isogeometric Analysis (IGA) has been originally introduced and developed by T.J.R. Hughes, J.A. Cottrell, and Y. Bazilevs, in 2005, to generalize and improve finite element analysis in the area of geometry modeling and representation. However, in the course of IGA development, it was found that isogeometric methods not only improve the geometry modeling within analysis, but also appear to be preferable to standard finite elements in many applications on the basis of per-degree-of-freedom accuracy. Non-Uniform Rational B-Splines (NURBS) were used as a first basis function technology within IGA. Nowadays, a well-established mathematical theory and successful applications to solid, fluid, and multiphysics problems render NURBS functions a genuine analysis technology, paving the way for the application of IGA to solve a number of problems of academic and industrial interest. Further fundamental topics of research within IGA include the analysis of trimmed NURBS, as well as the development, analysis, and testing of flexible local refinement technologies based, e.g., on T-Splines, hierarchical B-Splines, or locally refined splines, in the framework of unstructured multipatch parameterizations. We also welcome IGA methods that is combined with data-driven methods (AI and machine learning).
Keywords:
Advanced Numerical Formulations, geometrical pre/post-processing, Isogeometric Analysis
In scientific and engineering simulation frameworks, the need of enhanced accuracy, computational efficiency, and numerical robustness remains mandatory. This mini-symposium aims to investigate recent developments in numerical methods where model order reduction meets scientific machine leanring, with a focus on Computational Fluid Dynamics (CFD) problems across several application fields.
The symposium emphasizes several crucial aspects, covering advanced discretization techniques, reduced order models, and the integration of machine learning for data-driven simulations. These advancements provide a reliable contribution to improve significantly the efficiency and accuracy of the methodologies across different sectors (aerospace, automotive, energy, naval eng.), as well as biomedical and environmental engineering applications. More specifically, this mini-symposium will allow researchers to exchange insights and new ideas on novel approaches going beyond traditional methods. In this framework, novel contributions in finite element, finite volume, finite difference, spectral, meshless, and non-matching methods demonstrating their efficacy in capturing complex flow phenomena are welcome. Moreover, the combination of reduced order models (ROMs) with the aforementioned techniques constitutes another strategic key aspect of this symposium. The discussions will encompass intrusive/non-intrusive and linear/non-linear approaches, as well as data-driven techniques powered by scientific machine learning tools and algorithms used by the scientific community for addressing complex fluid mechanics problems coming from real-world applications.
Keywords:
Computational Fluid Dynamics, Machine Learning, Model Order Reduction/Reduced Order Modeling, surrogate models
The Boundary Element Method (BEM) offers significant advantages for solving partial differential equations, especially for infinite domains and complex geometries. This mini-symposium invites researchers to present the latest theoretical and practical advancements of BEM.
We seek contributions on:
- Formulations: New integral equations and regularization techniques.
- Efficiency: Fast BEM algorithms like FMM and H-matrix.
- Multiphysics: BEM's integration with FEM/FVM for coupled problems.
- Applications: Use of BEM in biomedical engineering and renewable energy.
- Optimization: BEM for topology and shape optimization.
- Synergy with AI/QC: Combining BEM with AI, machine learning, and quantum computing.
This symposium will highlight BEMâs role in solving diverse scientific and industrial challenges.
Keywords:
Boundary Element Method, Coupled problems, engineering applications, high performance computing, machine learning, Mathematical Modelling, Optimization, Quantum Computing
Reduced order models (ROMs) address the significant computational costs associated with solving differential equations using full-order models (FOMs), and provide efficient and reliable approximations essential for simulations in science and engineering, particularly in real-time simulations and multi-query applications.
Combining ROMs with advanced computational techniques such as domain decomposition and machine learning further enhances their ability to tackle complex problems. This mini-symposium aims to delve into these integrated methodologies, exploring their potential and advancements. Key topics include domain decomposition for surrogate modeling, dynamic mode decomposition, and machine learning-based ROMs. These topics will be discussed in the context of various linear and nonlinear problems, such as multi-component systems, multi-physics problems, dynamical systems with random inputs, and data assimilation.
By bringing together researchers and practitioners from diverse fields, this symposium will facilitate the exchange of innovative methodologies and the identification of future research directions, thereby advancing the fields of reduced order modeling.
Keywords:
Data Assimilation, Domain Decomposition, Machine Learning, Model Order Reduction/Reduced Order Modeling
This minisymposium brings together researchers advancing high-order discretizations for large-scale simulations of hyperbolic and multi-physics PDEs. The focus is on structure-preserving numerics (conservation, positivity, entropy stability, kinetic-energy preservation), robust shock capturing, fast time integration, and scalable preconditioning on heterogeneous hardware (CPUs, GPUs, and emerging accelerators). We emphasize rigorous verification and validation through manufactured solutions, benchmark suites, uncertainty quantification, and practical algorithm-hardware co-design that reduces memory traffic and enables mixed-precision performance.
A central theme is the careful use of scientific machine learning (SciML) to augment (not entirely replace) classical solvers. Topics include learned multigrid and preconditioner components, surrogate Jacobian actions for implicit solves, conservative reduced-order models and operator-inference surrogates with certified error control, ML-guided sensors/limiters and hp-adaptivity/AMR that respect discrete invariants, and data-assisted closure or parameter inference integrated with adjoints and ensemble methods. We explicitly invite talks that clarify when ML effectively encodes physics and when it fails or requires remedies to ensure reliability and generalization.
Target applications span compressible aerodynamics and hypersonics, magnetohydrodynamics and space physics, acoustics and electromagnetics, multiphase and reacting flows, and coupled multi-physics in complex geometries. The minisymposium aims to chart practical routes for deploying high-order, structure-preserving and SciML-augmented methods in routine simulation workflows.
Keywords:
Computational Fluid Dynamics, High-order methods, Numerical stability, scientific machine learning
Keywords:
Metal Hydrides, MultiscaleâMultiphysics Modelling, Phase Change Materials, termal energy
1800
Scientific Computing
This mini-symposium is dedicated to methodological and theoretical aspects of hybrid approaches combining Machine Learning techniques with traditional numerical methods or modeling for the resolution of partial differential equations (PDEs). Recent years have seen the emergence of models that intertwine data-driven componentsâneural networks, kernel methods, or learned operatorsâwith well-established discretization schemes (finite differences, finite elements, spectral methods, etc.), which offer new ways to accelerate simulations, increase accuracy and adaptivity of the solvers in complex regimes, and automate parameter selection. They also open up promising avenues for tackling high-dimensional problems, multiscale dynamics, model uncertainties or inverse problems. However, their integration raises important theoretical questions regarding stability, convergence, expressivity, and generalization.
We invite contributions that shed light on:
- the proposal of novel, efficient, and principled hybridization strategies combining machine learning and traditional PDE solvers;
- the mathematical foundations and analysis of hybrid MLâPDE methods;
- theoretical properties of learned components when embedded in numerical schemes;
- the impact of training dynamics and data sampling on the numerical solvers
Our objective is to foster rigorous dialogue around the design principles and theoretical guarantees that underpin successful hybrid application of machine learning to PDEs.
Keywords:
machine learning, Multiphysics problems, Numerical approaches, scientific machine learning
With accelerating climate change, scientists â especially in high-performance computing (HPC) â must address the environmental impact of large-scale simulations, which often produce substantial amounts of carbon dioxide. Although supercomputing infrastructure has become significantly more energy-efficient in recent years, leveraging these advances typically requires major adaptations to algorithms and code bases. With JEDI, the first module of Europeâs exascale computer, ranked 1st in the Green500 list, and similar systems, the use of graphics processing unit (GPU) accelerators is becoming essential to avoid wasting computational resources.
Many researchers have improved energy efficiency by exploiting single instruction, multiple data (SIMD) parallelism, such as via Advanced Vector Extensions (AVX), or by porting applications to GPUs. Fair comparisons with central processing unit (CPU) implementations remain essential to quantify energy savings. Domain-specific redesigns of models and algorithms also play a key role. For example, combining fine-grained agent-based models with coarse-grained metapopulation models has been shown to achieve up to 98% energy savings while still meeting the target outcomes. In general, combining algorithmic and hardware-aware strategies can preserve accuracy while reducing cost.
The idea of algorithmâhardware co-design is particularly advanced in machine learning (ML), which is now one of the main drivers of global computing energy use. As ML models grow in scale and complexity, so does the demand for energy-efficient training and inference. This has led to the development of specialized hardware, such as neuromorphic computing chips and tensor processing units (TPUs), designed in close alignment with the algorithms they support. Other promising directions include field-programmable gate arrays (FPGAs) and mixed-precision computing on GPUs and other accelerators.
This mini-symposium welcomes contributions on applications that enhance energy efficiency through novel hardware usage, algorithmâhardware co-design, or methodological advances. We also invite submissions on measuring and comparing energy consumption (e.g., CPU vs. GPU) and on techniques that reduce energy usage while maintaining results within acceptable precision bounds.
Keywords:
Acceleration, Carbon footprint, Energy, GPUs, High-Performance Computing, Neuromorphic computing, Novel model design, SIMD, TPUs
The numerical simulation of strongly non-linear systems at large scale is one of the most persistent challenges in scientific computing. Recent developments in non-linear preconditioning, domain decomposition, and multi-scale methods, as well as in scientific machine learning, allow for the design of new and scalable numerical methods. This minisymposium provides a forum for the presentation and discussion of these new solution methods and their application to complex large scale problems from engineering, medicine, natural sciences, data sciences, or imaging.
Our discussion will cover established approaches based on structured matrices, mathematical modeling, and numerical simulation, including variational methods and emerging approaches from the growing field of machine learning, as well as their combination.
Keywords:
large scale problems, non-linear systems
This minisymposium focuses on recent advances in space-time finite element methods and high-order implicit time integration for the numerical solution of time-dependent partial differential equations. High-order space-time and implicit methods pose significant challenges. Solving the arising large coupled systems demands new algorithmic approaches and computationally efficient implementations. This minisymposium brings together contributions addressing these challenges through advanced discretization techniques, efficient solvers, and high-performance implementations.
A central focus is the interplay between high-order discretizations and the scalable solvers that make them practical. Talks will cover the design and analysis of multigrid and multilevel solvers tailored to space-time finite element frameworks. These include geometric and algebraic multigrid approaches, domain decomposition methods, and parallel-in-time algorithms that overcome the limitations of sequential time stepping and enable strong scalability on modern hardware. Another major focus is on robust block preconditioners for monolithic space-time systems. These preconditioners are essential for achieving optimal solver performance independent of the mesh resolution and polynomial degree. Recent advances in block-structured approaches have demonstrated significant gains in robustness and efficiency and established them as key building blocks for the development of scalable and reliable solvers for space-time systems.
An emphasis is on space-time tensor-product techniques, variational time integrators, and structure-preserving discretizations. Contributions include novel Galerkin methods and formulations tailored to stiff or multiscale problems. The translation of theoretical advances into practical performance requires hardware-aware implementations that minimize data movement and memory usage. Contributions will address matrix-free operator application, hardware-aware algorithm design, and cache-optimized and performance-portable implementations.
The minisymposium aims to foster exchange across numerical mathematics and scientific computing. The overarching goal is to advance the theoretical understanding, algorithmic design, and computational realization of space-time finite element methods and implicit high-order time integration as scalable and efficient tools for the simulation of complex dynamical systems.
Keywords:
high-order finite elements, implicit time integration, matrix-free, space-time finite elements, space-time multigrid, tensor-product
Computational fluid dynamics (CFD) traditionally has been one of the most compute-intensive disciplines in computational sciences. Typically more than half of compute resources of the Tier 0 computing centers around world is devoted to fundamental-research problems using CFD. CFD has become a standard tool in daily industrial work, disaster prediction and prevention, entertainment industries.
In order to enable more powerful predictive tools, better efficiency, and higher accuracy, CFD always has inspired research in applied mathematics and in computer science. CFD has faced and mastered disruptive changes in computer hardware architecture and incorporated dramatically improved algorithms. While doing so important new concepts for algorithm development such as co-design have been developed, which make CFD more resilient to future disruptive changes.
The mini symposium aims at gathering emergent concepts for next-generation CFD from a wide range of directions, and providing a forum for mutual assessment and discussion. While the inclusion of data-driven surrogates and of full substitutes for classical predictive simulators is already a fast progressing subject, systematic replacement of numerical approximation of ground truth mathematical models by algorithmic embodiments is still at its infancy. The same applies to hardware hybridization, where acceleration by GPU and TPU are well within the current scope, with quantum-acceleration still being a speculative entity.
Specifically we are interested in contributions on:
⢠Performance-portability and exascale computing in CFD
⢠Data-driven surrogates for generative or predictive tasks in CFD
⢠Quantum-inspired algorithms for accelerating CFD
⢠Automatic differentiation and inverse problems in CFD
⢠Advanced hybrid numerical methods and turbulence models for CFD
Keywords:
Hybrid Algorithms, Machine Learning, Computational Fluid Dynamics, next generation HPC, Quantum Computing
The aim of this mini-symposium is to discuss and to share recent ideas from numerics (w.r.t. discretization and solver techniques), scientific computing (w.r.t. computational, algorithmic, and implementational aspects) and data science (w.r.t. machine learning and neural network approaches) for the highly efficient treatment of partial differential equations (PDEs) that arise in the simulation of problems from computational fluid dynamics (CFD) and computational solid mechanics (CSM). The presented approaches shall particularly address new ideas regarding future high-performance computing (HPC) environments which will be in the exascale range and which will include massively parallel, heterogeneous architectures together with specific accelerator hardware (GPUs, FPGAs) including reduced arithmetic precision. The mini-symposium will concentrate on methods and their foundations and will highlight the interplay of these aspects with computational and algorithmic tools and particularly their realization in simulation software. We shall discuss, for instance, aspects regarding hardware-oriented numerics, energy-efficient and extremely scalable numerical approaches, scientific machine learning techniques together with artificial neural networks, numerical cloud computing, and massively parallel solvers exploiting parallelism in time. Other aspects to be included are nonlinear domain decomposition methods and extremely scalable numerical homogenization methods.
Keywords:
CFD, CSM, HPC, Mathematical software, Scientific Computing
The potential of quantum computing to solve scientific and engineering problems has been recognized over the past decade. The power of quantum computers stems from the efficiency in computational time and space for difficult problems by taking advantage of quantum superposition and entanglement. Quantum algorithms have been developed to solve engineering problems such as linear systems, eigenvalue, simulation, optimization, and machine learning. As a continuation of the Quantum Scientific Computing minisymposium held at WCCM in 2024 and USNCCM in 2025, this minisymposium will provide a platform for researchers to exchange the latest ideas of quantum computing to solve engineering and materials problems. The topics of interest include but are not limited to:
⢠Quantum algorithms and methods for computational solid mechanics and fluid dynamics
⢠Quantum algorithms and methods for multiscale simulation (e.g., SchrÜdingerization-based simulation)
⢠Quantum optimization algorithms
⢠Quantum algorithms for materials discovery and materials design
⢠Quantum machine learning and artificial intelligence
⢠Uncertainty quantification in quantum computing
⢠Error correction and mitigation in quantum computing
⢠Simulators of quantum computers on classical computing platforms
⢠Benchmark studies of quantum algorithms and methods
Keywords:
Computational Fluid Dynamics, Computational Solid Mechanics, Quantum Machine Learning, Quantum Optimization, Quantum Simulation, Quantum Computing
This minisymposium will explore the rapidly evolving intersection of quantum computing and computational mechanics, focusing on algorithmic innovations, hybrid quantumâclassical frameworks, and application-driven solutions. As quantum computing progresses from theoretical foundations to practical implementation, it opens up new avenues for addressing long-standing computational challenges in mechanics. These include solving large-scale systems of partial differential equations (PDEs) in elasticity, thermo-mechanics, and fluidâstructure interaction; modeling materials across scales, from atomistic and quantum domains to continuum representations; optimizing complex structural and multiphysics systems; and accelerating inverse problems in system identification, damage detection, and material characterization. Quantum-enhanced solvers also offer the potential to improve the efficiency of reduced-order modeling, nonconvex topology and shape optimization, and high-dimensional parametric studies needed for robust uncertainty quantification.
The primary goal of this minisymposium is to establish a dynamic and interdisciplinary platform for researchers from quantum information science, computational physics, applied mathematics, and engineering mechanics to exchange ideas, identify shared challenges, and forge new collaborations. These communities are increasingly converging around the need to tackle complex, coupled problems in simulation and design. By facilitating dialogue among experts such as quantum algorithm developers, numerical analysts, and mechanics modelers, the session aims to accelerate the integration of quantum computing into classical computational mechanics workflows. Through invited and contributed talks, as well as interactive discussions, the minisymposium will serve as a forum for evaluating the theoretical foundations, hardware considerations, and practical impact of quantum-enhanced methods. It will also contribute to shaping a research roadmap for this emerging field, identifying priority application areas, algorithmic benchmarks, and software needs. Ultimately, this session will help position the computational mechanics community to engage meaningfully with quantum technologies and benefit from their transformative capabilities in the coming decades.
Keywords:
Computational Mechanics, Hybrid Algorithms, Quantum Computing
This minisymposium aims to explore the emerging potential of continuous-variable quantum computing (CVQC) on photonic computers as a natural and scalable framework for advancing computational mechanics. Computational mechanics of practical fluids, solids, structures, and materials typically involves approximating solutions to nonlinear stiff partial differential equations (PDEs) in high dimensions. For practical applications on classical high performance computers today, reduced-order models are often employed for approximation, since high fidelity simulations (e.g., direct numerical simulation) are improbable. With emerging quantum computing technologies, efficient quantum algorithms for achieving high fidelity solutions to these PDEs are desirable. In particular, CVQC is a computational paradigm that utilizes systems with continuous spectra to encode and manipulate information. The infinite-dimensional Hilbert spaces of CVQC align with the functional spaces of PDEs, enabling direct representation of physical fields without spatial discretization. Unlike discrete qubit-based approaches, CVQC encodes information in continuous degrees of freedomâposition and momentum of quantum modesâmaking it inherently suited for modeling the continuous fields that define fluid mechanics, solid mechanics, and multiphysics systems. These qumodes inherently encode continuous variables, such as field quadratures, enabling a more direct representation of continuous physical quantities such as fields. As a result, CVQC offers a promising route toward simulating nonlinear dynamics while greatly reducing the overall degree of discretization, thereby providing a computational framework that is much more closely aligned with the continuous nature of the underlying physics. The main objective of this minisymposium is to bring together researchers from quantum computing, applied mathematics, and computational physics to explore the emerging potential of CVQC as a potential powerful framework to advance simulations of high-dimensional systems in computational mechanics.
Keywords:
algorithms, CFD, Quantum Computing
The solution of large and sparse linear systems represents a major bottleneck in many areas of computational science and engineering, such as computational fluid dynamics, magnetohydrodynamics, and structural mechanics. In this sense, real-world applications demanding a high resolution usually lead to seriously ill-conditioned, extreme-scale problems, triggering a complex interplay between numerical reliability and computational efficiency.
While sparse direct solvers offer strong numerical robustness, their high memory requirements and operational complexity, and reduced parallelism limit their applicability [1]. In contrast, iterative Krylov subspace methods offer an easier implementation and parallelisation, but their convergence is very sensitive to the coefficient matrix's spectrum and always require powerful preconditioners whose choice, design, and implementation remain among the most active research fields in numerical linear algebra [2].
This minisymposium welcomes contributions on sparse linear solvers for large-scale simulations of real-world applications. Topics of interest include, but are not limited to:
- New developments in general-purpose and physics-based preconditioners
- New algorithms tailored for modern HPC systems, such as matrix-free, communication- avoiding, mixed-precision, and GPU-friendly algorithmic redesign
- Data-driven strategies for preconditioning and solver parameter tuning
- Benchmarking, performance modelling, and reproducibility in solver development
- Integration of solvers and preconditioners into simulation frameworks
The goal of this minisymposium is to promote exchange between linear solver developers and application experts, fostering discussion across disciplines and applications.
REFERENCES:
[1] T. A. Davis, S. Rajamanickam, and W. M. Sid-Lakhdar (2016). A survey of direct methods for sparse linear systems. Acta Numerica, 25, 383â566.
[2] M. Benzi (2002). Preconditioning Techniques for Large Linear Systems: A Survey. Journal of Computational Physics, 182(2), 418â477.
Keywords:
GPU computing, HPC, Linear solvers, Preconditioning, Real-world applications
Quantum computing is rapidly emerging as a transformative tool for solving complex engineering problems that have traditionally relied on high-performance classical computing. This session will explore quantum algorithms in computational mechanics. Presentations will address both theoretical foundations and practical implementations, highlighting novel approaches in solving large-scale, high-dimensional problems more efficiently than classical methods alone. Topics will include quantum-accelerated solvers for stiffness matrix inversion, hybrid quantumâclassical algorithms, and quantum-based approaches to uncertainty quantification. Case studies can range from structural mechanics, fluidâstructure interaction, multiscale materials modeling, and other engineering systems. By bringing together experts in computational mechanics, quantum information science, and applied engineering, this session aims to define both near- and long-term roadmaps for quantum-enhanced simulations in engineering practice. This session encourages the participation from students, early-career researchers, and practitioners, aiming to foster a diverse and collaborative session to exchange ideas.
Keywords:
Finite Element Methods, Solvers
This minisymposium presents recent developments on large-scale biomechanical simulations of the human body. We invite contributions from a broad range of topics, including the development of novel discretization schemes for solving partial differential equations, the development of advanced mathematical algorithms for solving linear systems, involving both iterative methods with geometric, polynomial and algebraic multigrid schemes, physics-based block preconditioners as well as sparse direct solvers with novel low-rank factorizations. Furthermore, algorithm implementations on modern high-performance platforms, especially for GPU systems, are at the core of this minisymposium. This includes method development to best utilize these systems, such as matrix-free methods, mixed-precision algorithms, or the embeddeding into modern software packages.
The minisymposium is primarily driven by the European Centre of Excellence (CoE) dealii-X [1], an initiative hosted in four European countries and spanning mathematical development, biomechanical applications, and hardware-driven algorithm design. It covers mathematical abstractions using the deal.II finite element library and various modern third-part packages, such as PSCToolkit and MUMPS, all the way to applications. Contributions in related fields are explicitly invited, and it is planned to present a mix of contributions of the dealii-X CoE as well as other research. We also invite contributions from biomedical applications or progress on biomedical code development, to participate in the minisymposium.
Keywords:
Biomedical Codes, Exascale Computing, Finite Element Method
Advances in high-performance computing (HPC) have opened unprecedented possibilities to improve academic research and the industrial design process. With the rise of heterogeneous hardware architectures and the emergence of exa-scale class computing systems, scale-resolving simulation now supports previously unattainable fidelity enabling the modelling of complex geometries and highly accurate predictions of challenging turbulent flows. These technologies demand new numerical algorithms, scalable solvers and data-driven analysis methods such that these tools effectively advance the transport and energy sectors.
State-of-the-art codes for scale-resolving simulations have transitioned to utilise heterogeneous hardware [1,2,3,4,5,6,7], however, the journey required major changes to
established algorithms. Previously optimised, CPU-based algorithms require a different
approach when integrating GPU-based accelerators. Scale-resolving simulations have specific challenges: extreme memory bandwidth requirements, communication bottlenecks with new CPU-GPU systems and efficient utilisation of accelerator hardware. These challenges set the stage for exchanging experience and discussing novel strategies for efficient usage of new and upcoming HPC systems.
This minisymposium provides a forum for developers of numerical algorithms and
application-oriented scientists to discuss HPC for scale-resolving simulations including DNS, wall-modelled and wall-resolved LES as well as hybrid approaches for complex geometries. Numerical algorithms tailored to achieve high throughput on multi-GPU, CPU-GPU coupled systems and novel hardware architectures. Scalability of solvers and preconditioners for Navier-Stokes type equations. Algorithmic design for load-balancing, communication reduction and mixed-precision strategies. In-situ data processing for handling parallel I/O at large-scale with on-the-fly analysis, data compression and feature extraction. Mesh generation for geometries with highly curved surfaces and complex shapes at large-scale and their adaptive optimisation. Further, multi-physics applications discussing coupling of cross-disciplinary solvers and integration of different code frameworks.
Keywords:
Heterogeneous
Hardware, Scale-Resolving Simulations, Computational Fluid Dynamics, High Performance Computing
Most efficient numerical algorithms in scientific computing and machine learning require the user to provide a routine that can calculate the derivative of a quantity with respect to some underlying parameters. With the increasing complexity of the models used in mechanics and the rise of scientific machine learning (SciML), users are increasingly turning to automatic differentiation (AD) techniques [1, 2] that can automatically derive this routine. This minisymposium aims to bring together researchers, software developers and users of automatic differentiation techniques to discuss the state-of-the-art in the field of automatic differentiation and its application to various challenging problems in mechanics. Possible topics include:
- the solution of challenging mechanics problems leveraging AD,
- novel combinations of AD approaches, e.g. coupling PDE-level AD tools with source-level AD tools,
- end-to-end differentiable numerical solvers and operator learning approaches including those using non-traditional basis functions, e.g. neural networks,
- material constitutive modelling using AD,
- checkpointing strategies for reverse-mode AD,
- and AD software tools and frameworks.
[1] Farrell, P., D. Ham, S. Funke, and M. Rognes. 2013. âAutomated Derivation of the Adjoint of High-Level Transient Finite Element Programsâ. SIAM Journal on Scientific Computing 35 (4): C369â93. https://doi.org/10.1137/120873558.
[2] Naumann, U. 2011. The Art of Differentiating Computer Programs. Software, Environments and Tools. Society for Industrial and Applied Mathematics. https://doi.org/10.1137/1.9781611972078.
Keywords:
Automatic Differentiation, Constitutive Modeling, Mathematical software, scientific machine learning
For an accelerated materials design, digital microstructures can be used, which exhibit the same characteristics as experimentally measured microstructures, but are obtained through microstructure modeling and simulation. This requires the application and development of advanced materials modeling and simulation techniques. In particular, numerical methods based on the phase-field method have become indispensable and extremely versatile tools in materials science, microstructure mechanics, and physics. The method typically operates at the mesoscopic length scale and provides important information about morphological changes in materials by mapping interfacial motions of physically separated regions. It provides a numerically highly efficient treatment of the moving interfaces, as no explicit tracking of the interfaces is necessary. Thus, the phase-field method is widely established for modeling microstructural evolution processes, such as solidification, solid-solid phase transition, growth and coarsening of precipitations, grain growth and martensitic phase transformation. An outstanding feature of the phase-field method is the ability to consider different physical driving forces for interfacial motion due to diffusive, electrochemical, thermomechanical, etc. processes. In addition, large-scale numerical simulations can be performed by numerically solving the coupled multiphysics differential equations on high performance clusters. Due to this versatility, phase-field methods, used in a wide range of fields in materials science and physics, are constantly under development. The main objectives of this WCCM symposium are to establish cross-community standards for phase field modeling by
- Highlighting current issues, emerging applications, and outstanding perspectives in phase field modeling
- Identifying methodological commonalities among different phase field communities
- Discussing analytical challenges of phase field modeling in a transparent manner
- Establishing benchmarks for the verification of models and implementations
Keywords:
Bechmarks, chemo-mechanical coupling, Phase Field, phase field model, phase-field, Phase-Field Modeling, Thermo-Hydro-Mechanical-Chemical Coupling
Over the past decades, computing architectures have evolved such that Mooreâs Law and MPI-type scaling alone cannot solve todayâs challenge problems or maintain market relevance with CPU-only simulation performance. General compute GPU accelerators have become commonplace in both workstation and HPC settings. Bespoke architectures are emerging to further accelerate AI/ML workloads, which can have implications for computational mechanics codes, as well. Porting code, maintaining correctness, and achieving performance across these increasingly diverse architectures is a costly and enduring challenge.
This mini-symposium hosts presentations describing
⢠Methods for developing and modernizing computational mechanics software to run on emerging architectures including, but not limited to, GPU, FPGA, RISC-V, or ASICs.
⢠Use of threading and vector intrinsics on different architectures.
⢠Effective use of performance portability abstractions (e.g., Kokkos, Raja, OpenMP-offload, SYCL).
⢠Testing and verification of computational mechanics software on multiple diverse and emerging architectures.
⢠Portability and performance experiences and studies in any of the previously mentioned topic areas.
Keywords:
ASICs, FPGAs, GPUs, RISC-V, Parallel computing, portability
1900
Structural Mechanics, Dynamics and Engineering
The concept of vehicle-bridge interaction (VBI) originated from the investigation of high-speed trains in 1990s. Based on this idea, further extension to extract bridge frequencies from a passing vehicle has gradually become an important method in the indirect measurement since 2004, especially in the field of structural health monitoring of bridge structures. Since then, various related topics have been widely studied, including the extraction of bridge modal parameters, investigation of frequency variation for a VBI system, full-scale experiment and scale test, three-dimensional modelling, etc. In this regard, this minisymposium aims at providing a forum for worldwide researchers to present new research developments centered on vehicle-bridge interaction and its applications to bridges, but not limited to the above.
Keywords:
Bridge, Modal Parameters, Signal Processing, Vehicle-Bridge Interaction
Based on the previous studies on vehicle-bridge interactions, Yang et al. (2004) proposed the idea of extracting bridge frequencies from the dynamic response of a moving test vehicle and had the idea verified by a field test. This technique was quickly extended to construction of mode shapes and damage identification of bridges. It was referred to as the indirect method for bridge measurement, in that no vibration sensors need to be mounted on the bridge, but only one or few vibration sensors are required on the test vehicle. Compared with the conventional direct method that relies fully on the response of the bridge fitted with vibration sensors, the advantage of the indirect method is obvious: mobility, economy, and efficiency. Over the past years, a rapidly growing number of research has been conducted along the lines of the indirect method for bridge measurement, with significant advances made on various aspects of application. In this minisymposium, we shall invite all the experts to present their state-of-the-art research along the lines of the subject, and to exchange the ideas for enhancing the technique of indirect approach for modal identification and damage detection of bridges using the moving vehicles.
Keywords:
Bridge, damage, modal identification, test vehicle, Vehicle-Bridge Interaction
Theoretical models and their computational formulations for beam and shell structures undergoing large deformations are among the most intriguing and prolific research topics since the advent of computational mechanics. Today, the variety of application fields where the potential of such models can be deployed is dramatically expanding. While traditional fields such as civil, mechanical, aerospace, and offshore/naval engineering continue to pose significant challenges, new and fascinating problems are arising in biomechanics, mechanical metamaterials, soft robotics, 4D printing, programable objects, and beyond. Addressing both longstanding and novel challenges requires the development of advanced computational methods that can efficiently capture the increasing complexity of real-world phenomena. On this basis, the proposed MS aims at creating a platform for sharing and discussing the latest advancements in computational methods, with a special emphasis on new challenges associated with innovative applications. Relevant topics include, but are not limited to, geometrically exact formulations, finite rotations (their parametrization and discretization), finite strains, constitutive models and material nonlinearities, discretization techniques, contact and beams-shells-solids coupling, instabilities and bifurcations, dynamics and time integrators, and innovative applications (biomechanics, metamaterials, programmable objects, 4D printing, etc.).
Keywords:
Finite rotations, Large deformations, Shells, Beams, Material nonlinearities
This minisymposium focuses on both theoretical and practical aspects concerning the transient solution of problems in structural dynamics in science and engineering. Particularly, novel numerical methods and solution strategies as well as discretization schemes in space and time for wave propagation, structural vibration, structural health monitoring, coupled problems (e.g., fluid-structure-interaction) and impact problems are of interest. This includes, but is not limited to the development or the application of
⢠isogeometric and high-order finite element methods (e.g., IGA, SEM, p-FEM, VEM, among others),
⢠fictitious/embedded domain and immersed boundary methods,
⢠meshfree methods,
⢠mass lumping and mass scaling techniques, or
⢠advanced time integration schemes (e.g., novel implicit, explicit, implicit-explicit or asynchronous time integration schemes, sub-cycling, parallel implementation, etc.).
Furthermore, we particularly welcome contributions focused on large-scale challenges and industry-relevant applications.
Keywords:
Advanced Discretization Schemes, Advanced Time Integration Schemes, Computational Structural Dynamics, Mass Lumping, Mass Scaling
This Minisyposium is devoted to recent developments on the various aspects of contact mechanics:
⢠Interface behaviour: unilateral contact, friction, adhesion, CZM, viscosity, fretting, wear, peeling, rolling contact, contact in biomechanics, fluid flow in contact interface.
⢠Computational models: multilevel approaches (molecular and nano-micro-macro models), multi-physics (thermo-piezo- âŚ), coupled multi-field formulations, fractal surface characterization, homogenization, bi-potential.
⢠Computational methods: fast solvers, multigrid, isogeometric analysis, NURBS, virtual elements.
⢠Dynamics of structures and of rigid bodies, instabilities.
⢠Discretization methods for overlapping immersed and embedded meshes.
⢠Mathematical progresses.
⢠Industrial applications involving interface and contact conditions.
Keywords:
Adhesion, Computation, Computations, Contact, Friction, Interface, Modelling
Railway is one of the most sustainable mass transport systems, and it plays a key role in assisting the socio-economic development of citizens and cities. The commercial speed of rail vehicles is being continuously increased to improve the competitiveness of this transport system, introducing additional challenges in the design of safe and efficient vehicles, which is supported by the development of advanced components or subsystems of railway vehicles. Nevertheless, the study of new solutions is often limited to the virtual environment, due to the economic and bureaucratic requirements limiting the organisation of field tests. Therefore, experimental tests are often limited to the characterization of the single components.
In this context, the use of advanced testing techniques is emerging in railway engineering. In particular, the Hardware-In-the-Loop (HIL) methodology is used to support the validation of new components for railway vehicles by quantifying their influence on the dynamic performance of the train. This approach merges the potentialities of computational mechanics with the experimental testing of real components. In the railway engineering field, HIL testing was proposed to support the development of suspension components to improve ride comfort and stability. Moreover, HIL was used to study the pantograph-catenary interaction through dedicated test rigs, and further applications were proposed to study the longitudinal dynamics of railway vehicles, even considering roller rig testing facilities.
This mini-symposium invites scientific contributions focusing on the application of HIL strategies and advanced testing procedures to railway components or subsystems. Topics are expected to be (but not limited to):
- Innovative components of railway vehicles supported by advanced or HIL testing methodologies: validation, component optimization, failure and robustness analysis.
- Development of refined methodologies to foster the application of advanced or HIL testing in the context of railway dynamics, such as real-time modelling of wheel-rail contact, or control systems applied to mechatronic solutions.
Keywords:
railway dynamics, real-time hybrid simulation, experimental testing, Hardware-In-the-Loop
Driven by the rise of innovative materials such as non-metallic reinforced components, there is a general trend to rethink the design of structural concrete buildings and allows conceiving the construction of those. Inspiration from biology is leading to new ways to build lightweight, resource-efficient components with optimized strength. Compared to their conventional steel counterparts, these reinforcements offer superior flexibility in geometric shape, while at the same time offering higher corrosion resistance.
The construction necessitates the utilization of advanced simulation techniques for structural analysis. On the one hand, micromechanical mechanisms have to be investigated which take into account the material behavior of the components and their interaction. On the other hand, simulation methods for the analysis of curved, lightweight construction elements need to be developed. At all levels of analysis, a high degree of accuracy and efficiency is required to exploit the material properties to the optimum.
Topics of interest include (but not limited to)
⢠Material modelling of concrete and the interface between concrete and reinforcement,
⢠Methods for the analysis of debonding, fracture and failure, e.g. phase field or X-FEM approaches,
⢠Simulation methods for thin, curved structures,
⢠Homogenization, multi-scale modeling and mixed-dimensional substructure modeling,
⢠Model order reduction techniques in quasi-static, inelastic analysis,
⢠Topology optimization of slender structures,
⢠Experimental investigation and validation of numerical methods.
This mini-symposium aims to convene researchers specializing in the field of modelling structural concrete and to provide a platform for the exchange of interdisciplinary knowledge. The primary focus is on advanced simulation techniques tailored for slender and brittle concrete structures.
Keywords:
Experimental investigation and validation, Homogenization, Methods for fracture and failure, Simulation methods for thin structures, ; Concrete, Material modeling
The conservation and assessment of masonry and historic structures represent a critical and multifaceted challenge in structural engineering, often requiring the support of advanced and reliable numerical modeling strategies [1].
This mini symposium aims to bring together researchers working at the forefront of computational modeling in the field of masonry and heritage structures. Contributions focused on recent advances in both continuum and block-based modeling approaches [2], as well as on innovative formulations for simulating the dynamic response and rocking behavior of masonry structures [3,4] are invited. Advances in practice-oriented modeling techniques, such as macro-element and equivalent frame models, are also welcome.
In addition, the mini symposium will emphasize modeling strategies that tackle key challenges in historic structures including the effects of material degradation and environmental exposure, and the integration of numerical models with structural health monitoring systems. These modeling strategies may be applied to a wide range of built heritage, including buildings, infrastructure, and monuments. Computational approaches that support decision-making in historic structures conservation are strongly encouraged, also considering digital-twin and artificial intelligence (AI)-based tools.â
REFERENCES
[1] DâAltri, A.M., Sarhosis, V., Milani, G. et al. (2020) Modeling Strategies for the Computational Analysis of Unreinforced Masonry Structures: Review and Classification. Arch Computat Methods Eng 27, 1153â1185
[2] Gatta, C., Nale, M., Addessi, D., Benvenuti, E., & Sacco, E. (2025). Large displacement analysis of masonry structures coupling enhanced virtual elements and damage-friction interfaces. Computers & Structures, 313, 107749.
[3] Ghezelbash, A., Sharma, S., D'Altri, A.M., Lourenço, P.B., Rots, J. G., & Messali, F. (2025). Challenges in HighâFidelity Implicit BlockâBased Numerical Simulation of Dynamic OutâofâPlane TwoâWay Bending in Unreinforced Brick Masonry Walls. Earthquake Engineering & Structural Dynamics.
[4] Vecchio, D., Vlachakis, G., Menon, A., & Lourenço, P.B. (2025). Seismic vulnerability assessment against rocking and sliding failure using nonlinear dynamic analysis: Application to the temples of Bagan, Myanmar. Structures,74, 108584.
Keywords:
Collapse Simulation, Computational Mechanics, Masonry Mechanics, Structural Health Monitoring
The optimal design of lightweight structures enables significant weight and energy savings, with applications spanning aerospace, automotive, and packaging industries. Weight reduction is typically achieved through thickness minimization and the use of architected or metamaterials, while preserving sufficient stiffness to maintain structural integrity under various static and dynamic load conditions. The complex responses of this type of structure reveal multiple solutions due to the buckling phenomenon and instabilities which are difficult to model with classical algorithms. This is also the case for other applications such as those involving soft materials that we can encounter in biomechanics or in the chemical industry. These soft components are capable of sustaining large deformations under diverse stimuli and often lead to complex responses involving multiple bifurcations.
In this mini-symposium, we are interested in advanced experimental and numerical techniques developed for modelling the complex responses of soft materials and slender structures considering the interaction between different loading involved in mechanical systems.
We hope to bring together experts working on these different aspects to review the emerging challenges and share the latest advancements in this research field. Topics of particular interest include but are not limited to :
⢠geometric and material instabilities in soft materials
⢠wrinkling, creasing, folding and ridging in extreme materials under various stimuli
⢠shape buckling of flexural structures such as plates, shells and membranes
⢠growth-induced deformations in biological tissues, biomaterials, bio-inspired structures
⢠micro-structural and macroscopic modelling in composites across length scales
⢠novel mathematical modelling method and constitutive theory
Keywords:
Bifurcation, Buckling, Lightweight Structures, Wrinkling
The aim of this mini symposium is to cover advanced mathematical models, exploring their applications in dynamics, vibrations, control and fractional calculus. The study focuses on how these models can be used to describe and predict complex behavior in physical and engineering systems. The analysis of vibrations and dynamics is essential for the development of efficient and innovative technologies, while fractional calculus offers a differentiated approach to problems involving memory and non-local effects. This combination of mathematical techniques provides powerful tools for optimizing the performance and stability of diverse systems, from turbines and mechanical structures to electronic and biomedical devices.
Keywords:
Fractional Calculus, Vibrations, Dynamics
Dynamic phenomena in coupled problems are of major importance in many multi-physics applications in industry and academia. Their modeling and simulation in the transient or frequency domain are notoriously challenging, even more so in the context of optimization. One must balance the accuracy and complexity of the models with the computational cost of the solvers, enabling compatibility with classical optimization techniques. The aim of this minisymposium is to gather researchers from both industry and academia to review recent advanced developments around two key points:
1. The modeling of the involved linear or non-linear dynamic multi-physics phenomena, including the efficient numerical treatment ranging from suitable finite element technologies, coupling strategies to reduced-order models of the coupled system.
2. The development of multi-objective and/or constrained optimizations in this physical context such as parametric, shape and/or topological optimizations applied to deterministic and/or stochastic settings. Multi-fidelity approaches and surrogate-based techniques are welcome in this minisymposium, as well as methods of gradients computation based on adjoint problems.
With respect to applications, coupled/multi-physics systems are, for example, expected in the following non-exhaustive list of fields:
⢠Fluid-structure interaction (vibroacoustics, sloshing, aeroelasticity, etc.)
⢠Coupled structures incorporating viscoelastics, porous materials, etc.
⢠Coupled structures with smart electric devices such as piezoelectric patches.
⢠Design of multi-physics dynamic systems such as electrical machines.
Keywords:
coupled systems, Dynamics, Multi-physics, Optimization
Physical interactions between solid-solid and solid-fluid domains and coupled interface problems are omnipresent in biomechanics. E.g., they appear in the form of fluid-structure interaction (FSI) between body fluids and external agents [1], or coupling constraints between slender elastica, or contact interactions between deformable elastica of varying dimensions, such as beam-beam, beam-surface, and surface-surface [2]. Inherently, these interface problems are nonlinear and require advanced computational modelling frameworks for their thorough understanding. While state-of-the-art numerical techniques have been proposed to study the FSI phenomenon in biomechanical applications using either monolithic or partitioned schemes [3], mortar-type coupling approaches involving Lagrange multipliers have been adopted to account for contact interactions [4]. For instance, the numerical simulation of two-way FSI coupling between pulsatile blood flow and complex elastic vascular networks spanning multiple spatio-temporal scales is a challenging problem in itself. Another notable example is the high-fidelity numerical simulation of endovascular devices for aneurysm treatment that involves contact interactions. In this mini-symposium, we aim to address the forefront of recent progress, current research, and emerging trends in the computational modelling of nonlinear interface problems in biomechanics.
[1] U. KĂźttler et al., Coupling strategies for biomedical fluidâstructure interaction problems, International Journal for Numerical Methods in Biomedical Engineering 26 (2010) 305-321.
[2] C. Meier et al., A finite element approach for the line-to-line contact interaction of thin beams with arbitrary orientation, Computer Methods in Applied Mechanics and Engineering 308 (2016) 377-413.
[3] K. Surana et al., Mathematical models for fluid-solid interaction and their numerical solutions, Journal of Fluids and Structures 50 (2014) 184-216.
[4] I. Steinbrecher et al., A consistent mixed-dimensional coupling approach for 1D Cosserat beams and 2D surfaces in 3D space, Computational Mechanics (2025).
Keywords:
Contact Interactions , Coupling Constraints, FSI, Interface, Biomechanics
Adaptive and compliant structures offer innovative solutions by leveraging their inherent flexibility to redistribute loads under changing conditions, thereby enhancing load-carrying efficiency and enabling complex morphing capabilities. Simultaneously, the field is advancing towards the design of structures capable of undergoing large, controlled deformations to perform complex maneuvers, opening up new possibilities in architecture, engineering and robotics. Soft robotics, in particular, benefits from computational frameworks that allow for the design of intelligent morphing behaviors, safe interactions with humans and delicate environments, and integration with real-time control systems. These capabilities position soft robotic systems as a frontier domain within adaptive and compliant mechanics.
In developing adaptive and compliant structures, the understanding of its structural behavior using computational mechanics plays a vital role. It enables accurate modeling and simulation of complex and deformable structures, which is crucial for designing, optimizing, and controlling these systems. This minisymposium will bring together researchers working on the computational design, analysis, and control of adaptive and compliant structures, including soft robotics and systems with embodied intelligence. Topics of interest include, but are not limited to:
- Modeling and simulation of compliant structures, including load-responsive and soft robotic systems
- Optimization techniques, including form finding and actuator design for performance and efficiency
- Design methodologies for flexible structures capable of controlled large deformations
- Motion planning and control for adaptive and flexible structures
- Sensor and actuator placement strategies and model-based control algorithms
- Autonomous materials and physical control strategies
- Active and passive control strategies for energy efficiency and complex deformation maneuvers
- Reliability and fail-safe design methodologies for adaptive systems
- Design and computational analysis of deployable and reconfigurable structures
- Integration of machine learning and embodied intelligence in soft systems
This minisymposium aims to foster a multidisciplinary exchange among researchers in computational mechanics, structural engineering, robotics, and applied physics to collectively shape the future of adaptive and compliant design.
Keywords:
actuation and
control strategies, actuator placement, compliant structures, deployable
structures, flexibility, mechanical metamaterials, motion design, programmable structures, soft robotics, structural optimization, Adaptive structures, embodied intelligence
Induced soil vibrations may reduce the serviceability of a building and can cause discomfort or health problems for residents. Ground-borne vibrations in highly populated areas are related, for example, to traffic, vibrating machinery, seismicity, or energy generation processes (e.g. geothermal power plants). Less urbanized areas, e.g. coastal regions, can also be affected by ground-borne vibrations. An example presents the construction and operation of offshore wind turbines and offshore drilling platforms for oil and gas.
Theoretical, numerical, and experimental approaches have been developed to efficiently and precisely describe complex structures and sophisticated materials in an unbounded or bounded soil domain. These cover, for example, accomplishments in developing and describing (non-) linear soil-structure-interaction models or seismic metamaterials. Another field presents the description of complex soil domains, which are characterized by layering, multiphase influences, or plastic behaviour. These approaches often have in common that they mimic an âinfiniteâ soil domain and approximate the Sommerfeld radiation condition.
The purpose of this minisymposium is to bring together academic staff, researchers, post-graduate students, and professional engineers who deal with advanced topics in the field of ground-borne vibrations to present and share novel approaches. Research areas may include (but are not limited to) the following issues:
(a) Modelling complex soil materials, e.g. layering, water saturation, seismic metamaterials
(b) Impact of uncertainties in data sets
(c) Non-linear material behaviour in a soil medium
(d) Wave scattering and diffraction due to incident wave fronts
(e) Soil-structure interaction
Keywords:
complex materials, ground-borne vibrations, soil-structure-interaction, wave scattering and diffraction , infinite soil domain
Architectured metamaterials have attracted increasing interest because of their significant applications in many scientific fields. These materials are endowed with peculiar internal structures to determine unconventional and enhanced mechanical, acoustic, electromagnetic, and thermal properties that cannot be observed in nature. The rational design of such architectured systems represents a rapidly evolving research frontier, with the potential to enable breakthrough technologies and transformative engineering solutions.
This Mini-Symposium will focus on theoretical, numerical, and experimental approaches to model and analyse architectured metamaterials. A special attention will be given to homogenization schemes, multiscale and multi-physics techniques, multi-field coupling phenomena, and computational techniques to predict the overall behavior of these materials. Participants will have the opportunity to discuss the impact of these materials on fields ranging from mechanical, civil, naval, aerospace, and biomedical engineering to robotics.
The topics of the Mini-Symposium include, but not limited to: 1) manufacturing techniques for architectured metamaterials, such as 3D and 4D printing; 2) computational methods and simulation techniques to investigate architectured metamaterials; 3) wave propagation in metamaterials; 4) case studies and applications demonstrating the effectiveness of the integration of architectured metamaterials into engineering solutions; 5) local and nonlocal constitutive modelling approaches; 6) parametric and topological optimization methods for material design; 7) multi-field problems that involve coupled physical phenomena; 8) homogenization techniques exploited to characterize the macroscopic behavior of these complex materials under both static and dynamic conditions.
Keywords:
Architected materials, complex materials, Composite Structures, Design, Eigenvalue problem, Homogenization, Multiscale, Nonlocal modeling
Non-destructive testing (NDT) and structural health monitoring (SHM) are critical for ensuring quality in both the manufacturing and operational phases of various structures. This mini-symposium aims to present and discuss recent advancements in various NDT and SHM technologies (Visual, ultrasonic, thermographic, eddy current, acoustic emission, magnetic, etc.). It includes but not limited to: i) Computational modelling methods; ii) Novel methods, approaches, and software integrated with advanced sensor technologies; iii) Signal processing algorithms, feature extraction, and damage indicators; and iv) AI-based methods for enhancing the effectiveness and efficacy of defect detection in NDT and SHM.
Keywords:
AI, modeling and simulation, NDT, SHM, Signal Processing
The design of engineering structures is increasingly challenged by the need to ensure performance under uncertain and evolving conditions. From climate-induced hazards and seismic events to material variability and modeling inaccuracies, the sources of uncertainty in structural behavior are broad and complex. Traditional deterministic design methods often fall short in addressing these challenges sufficiently. In response, the concept of robust designâcreating structures that maintain functionality across a wide range of scenariosâhas emerged as a critical objective in structural engineering.
This mini symposium, "Robust Design of Structures" aims to bring together researchers and practitioners working across domains such as structural engineering, computational science, architecture, and computational design to explore how modern tools and interdisciplinary approaches can lead to safer and more resilient built environment. The symposium encourages contributions that address the theoretical foundations, computational strategies, and applications of robust design of load-bearing structures.
Key topics of interest include, but are not limited to:
⢠modeling and simulation of robust structures
⢠form finding and optimization for robust design
⢠basic concepts in structural mechanics
⢠distributed redundancy
⢠sensitivity analysis
⢠progressive collapse, risk, and fault tree analysis
⢠reliability and fail-safe design of structures
⢠uncertainty quantification for robust design
Keywords:
Optimization, Redundancy, Robust Design, Structural Mechanics
Functionally Graded Materials (FGMs) are advanced composites characterized by a continuously varying microstructure, able to mitigate abrupt interfaces, providing a smooth gradient of properties and mitigating stress concentrations typical of traditional laminates [1].
Initially designed to reduce the thermal stresses arising from the high temperatures of the metal and ceramic interfaces in a space shuttle project [2], their ability to tailor materials mixtureâs distributions according to specific operating requirements, without the disadvantages shown by traditional laminates, was rapidly noticed in different Science and Engineering fields [3].
The use of optimization techniques and more recently of machine learning algorithms is contributing in a significant manner to obtain enhanced and more representative models of the complex relationships between FGMsâ composition, their properties and performance at different scale levels.
This mini-symposium aims to highlight the recent advances in the broad scope of the modelling, analysis and optimization of functionally graded composites and structures.
Keywords:
Machine learning algorithms, Materials and structures modelling, Passive and active materials, Structural optimization, Functionally graded composite materials
This minisymposium convenes researchers and practitioners advancing supplemental energy-dissipation technologies that transform structural dynamics into controllable, low-damage response. We welcome contributions that tie first-principles mechanics to deployable engineering across building and bridge systems, in both new design and retrofit.
Device families of interest include metallic yielding fuses (ADAS/TADAS, slit and plate dampers, replaceable links), frictional interfaces, viscous fluid dampers, viscoelastic and fractional-order media, tuned mass/liquid/inerter devices and hybrid variants, negative-stiffness and self-centering solutions, semi-active magneto-rheological dampers, fully active control devices, and rocking or base-isolated systems with supplemental damping. Topics span constitutive characterization (rate/temperature dependence, low-cycle fatigue, degradation and aging), deviceâstructure interaction, connection and anchorage mechanics, and system integration with RC, steel, timber, and precast assemblages.
Methodologies may include shake-table and quasi-static cyclic testing at component and subassembly scales; hybrid simulation and real-time cyber-physical testing; full-field measurement (DIC), system identification, and model updating; high-fidelity finite-element and fluidâstructure models; validated macro-elements and surrogate/reduced-order models; multi-objective optimization (performance, cost, embodied carbon); and rigorous uncertainty quantification. Application-oriented studies are encouraged on performance-based and resilience-oriented design, probabilistic demand and fragility modeling calibrated to experiments and field observations, code development and calibration, and post-event inspection, repairability, and rapid functional recovery. Multi-hazard contextsânear-fault pulses, aftershock sequences, windâseismic interaction, thermal/environmental effectsâare especially welcome. The sessionâs goal is to distill mechanistic insight into trustworthy design rules and details that shorten downtime and elevate community resilience.
Keywords:
Buckling-Restrained Braces, Nonlinear Dynamic Analysis, Structural Seismic Resilience, Supplemental Damping Devices, Nonlinear Finite Element Modeling, Seismic Energy Dissipation
2000
Verification and Validation, Uncertainty Quantification and Error Estimation
Modeling and simulation play a critical role in contemporary engineering practices, often informing decision-making processes. Modeling and simulation are frequently used in several physics disciplines, including computational fluid dynamics (CFD), computational solid mechanics (CSM), and computational electromagnetics (CEM). Consequently, it is essential to assess the credibility of the simulation results. To this end, the activities of verification, validation, and uncertainty quantification (VVUQ) are central to assessing the credibility of the modeling and simulation outcomes. Several relevant organizations, including the American Society of Mechanical Engineers (ASME)[1-6], the American Institute of Aeronautics and Astronautics (AIAA)[7], and the International Association for the Engineering Modelling, Analysis and Simulation Community (NAFEMS) [8], have published comprehensive guidelines and motivations for VVUQ.
The proposed minisymposium welcomes presentations on different aspects of verification, validation, and uncertainty quantification, which include, but are not limited to the following:
⢠Standard and novel code-verification, solution-verification, validation, and uncertainty-quantification techniques.
⢠Descriptions and explanations of published VVUQ standards and guides.
⢠Applications of VVUQ techniques to practical engineering problems.
Keywords:
Uncertainty Quantification, Validation, Verification
Advances in physics-based modelling are responsible for the generation of massive datasets containing rich information about the physical systems they describe. Efforts in Uncertainty Quantification (UQ), once an emerging area but now a core discipline of computational mechanics, serve to further enrich these datasets by endowing the simulation results with probabilistic information describing the effects of parameter variations, uncertainties in model-form, and/or their connection to and validation against physical experiments.
This MS aims to:
1. Highlight novel efforts to (A) Harness the rich datasets afforded by potentially multi-scale, multi- physics simulations for the purposes of uncertainty quantification; and (B) Develop physics-based stochastic models, solvers, and methodologies for identification, forward propagation, and validation.
2. Address modelling problems at multiple length-scales, ranging from the atomistic level to the component level, for a broad class of materials (including metals, metallic alloys, composites, polymers, and ceramics).
This includes, but is not limited to, efforts that:
⢠Merge machine learning techniques with physics-based models.
⢠Develop physics-based stochastic models and low dimensional representations of very high dimensional systems for the purposes of uncertainty quantification.
⢠Extract usable/actionable information from large, complex datasets generated by physics-based simulations.
⢠Develop active learning algorithms that exploit simulation data to inform iterative/adaptive UQ efforts.
⢠Develop stochastic solvers and sampling algorithms.
⢠Interpolate high-dimensional data for high-fidelity surrogate model development.
⢠Learn the intrinsic structure of physics-based simulation data to better understand model-form and its sensitivity.
⢠Develop new methodologies for model identification.
⢠Assess similarities/differences/sensitivities of physics-based models and validate them against experimental data.
This MS aims to span across applications of mechanics, with an emphasis placed on methodological developments that can be applied to physical systems of all types.
Keywords:
Computational Materials Science, Physics-Based Data-Driven Modeling, Uncertainty Quantification, Computational Engineering
Recent developments in scientific machine learning and AI are reshaping the landscape of predictive computational modeling by enabling the fusion of heterogeneous data sources, including experimental measurements, imaging, and high-fidelity simulations. These integrative approaches are critical for advancing model credibility, informing high-consequence decision-making processes in complex physical systems, such as digital twins, optimal design under uncertainty, and experimental design.
This mini-symposium will highlight novel research at the intersection of uncertainty quantification and machine learning, with an emphasis on scalable, trustworthy, and reliable predictive modeling frameworks. It will provide a forum for discussing foundational advances, algorithmic innovations, and application-driven studies that leverage uncertainty-aware machine learning modeling. Contributions addressing theoretical, computational, and software aspects are welcome. Topics of interest include (but are not limited to):
⢠Uncertainty quantification (UQ) techniques in deep learning models
⢠Model selection and credibility assessment
⢠Optimal design, control, and decision-making under uncertainty
⢠Operator learning and neural surrogates
⢠Physics-constrained generative models and latent representations
⢠Scalable UQ and inference algorithms for high-dimensional systems
⢠Discovery of governing equations and interpretable model structures from data
⢠Software frameworks for large-scale Bayesian inference and UQ
⢠Applications in materials, biomedicine, climate, aerospace, and energy systems
Keywords:
AI, Predictive science
Variability and heterogeneity are intrinsic to many advanced engineering materialsâranging from sedimentary rocks and cementitious composites to fiber- and nano-reinforced metamaterials, porous media, and biological tissues. Accurately quantifying and propagating these uncertainties is critical for reliable computational design, robust material characterization, and meaningful interpretation of experimental data. This Minisymposium brings together recent theoretical advances and computational strategies in Uncertainty Quantification (UQ) for random heterogeneous materials, encompassing both forward propagation of uncertainties through computational models and statistical inverse identification of probabilistic material models.
Contributions are invited (but not limited) to the following areas:
⢠Stochastic modeling of material heterogeneity: representation of random fields and microstructures, data-informed microstructure descriptors, multiscale and scale-bridging approaches.
⢠Forward UQ for constitutive behavior: propagation of uncertainties in linear and nonlinear laws (elasticity, plasticity, damage, fracture and other material nonlinearities) under (quasi-)static or dynamic loading.
⢠Statistical inverse problems and model calibration: Bayesian parameter identification of probabilistic models; stochastic optimization and Markov-Chain Monte Carlo (MCMC) methods with experimental-data considerations.
⢠Surrogate and reduced-order modeling: data-driven, physics-informed and probabilistic-learning approaches to accelerate UQ workflows; error estimation, adaptivity, and reliability assessment of surrogates; integration with high-fidelity simulations.
⢠Optimization under uncertainty: material and structural design with performance, reliability or robustness criteria; multi-objective and risk-aware optimization strategies.
This Minisymposium aims to (i) showcase advances in UQ methodologies specifically tailored to random heterogeneous materials and their computational modeling, (ii) highlight challenges in modeling, simulating and identifying complex material behavior under uncertainty, and (iii) foster interaction among theoreticians, computational scientists, and experimentalists to advance uncertainty-aware material design and analysis. By bridging together experts in stochastic modeling, numerical simulation, and data-driven methods, we seek to chart new directions for reliable and efficient UQ in materials science and computational mechanics.
Keywords:
Forward & Inverse UQ, Machine/Probabilistic Learning, Optimization Under Uncertainty, Random Linear and Nonlinear Heterogeneous Materials, Statistical Inverse Problems, Statistical Surrogate Modeling, Uncertainty Quantification
Computational fluid dynamics is used across a wide range of fields, including civil and aerospace engineering, where strict regulatory requirements must be met. Traditionally, certification has relied on extensive wind tunnel testing programs, which are both costly and time-consuming. With continuous advances in computing power and model fidelity, simulation accuracy has significantly improved over the past few decades, and the CFD community is now exploring the possibility of complementing and reducing, the need for wind tunnel testing. To achieve these ambitious goals, it is essential to develop robust and reliable uncertainty quantification methods particularly those capable of addressing epistemic uncertainties (i.e. caused by lack of knowledge), such as those arising from turbulence closure models. Furthermore, the issue of uncertainty in wind tunnel testing itself raises an important question: how accurate must a model be?
This mini-symposium aims to contribute to the establishment of a certification-by-analysis process. It brings together specialists from both computational and experimental backgrounds, offering diverse perspectives on the topic. Participants will have the opportunity to present and discuss their work, and to initiate or strengthen collaborations in this important field.
Keywords:
Computational Fluid Dynamics, Certification by Analysis
Reliability analysis is essential for the development, design and assessment of engineering systems under uncertainty. Challenges in computing the probability of failure are associated with non-linear system behaviour, large numbers of uncertain parameters and failure/rare events inducing multiple, disconnected failure domains. We invite talks discussing efficient computational methods for simulating rare events and quantifying failure probabilities based on sampling, surrogate modelling and machine learning as well as approximation approaches. Relevant applications of these techniques are in the assessment of static and dynamical engineering systems, reliability-based design optimization, reliability-oriented sensitivity analysis, Bayesian updating of failure probabilities with real-time data and applications in digital twin models.
Keywords:
Bayesian updating, rare events, reliability sensitivity analysis, reliability-based optimization, surrogate modeling, digital twins, Reliability analysis
The qualitative and quantitative analysis of dynamical systems is crucial for understanding numerous real-world systems and phenomena, from weather prognosis and biological networks, to medical and engineering applications. However, the ubiquitous presence of uncertainty often renders the analysis of dynamical systems a difficult task. Therefore, methods for uncertainty quantification (UQ) in dynamical systems are evolving rapidly, with a particular emphasis on data-driven and machine learning approaches in recent years. At the same time, numerous challenges remain, particularly in the nonlinear setting. Exemplary challenges include, quantifying uncertainties on bifurcations and limit cycles; computationally managing high-dimensional uncertain states and inputs; deriving suitable sensitivity analysis metrics.
This minisymposium aims to cover methodological work in the vast field of UQ in dynamical system analysis with a focus on data driven methods, related (but not limited) to:
⢠UQ for nonlinear dynamics (e.g., bifurcations, limit cycles)
⢠Data-driven forward and inverse UQ for dynamical systems
⢠Combinations of Fourier analysis with UQ methods
⢠Surrogate and reduced order modeling (incl. machine learning) for dynamical systems
⢠Sensitivity analysis for dynamical systems
REFERENCES
[1] Partovizadeh, A., SchĂśps, S., and Loukrezis, D. "Fourier-enhanced reduced-order surrogate modeling for uncertainty quantification in electric machine design." Engineering with Computers (2025): 1-21.
[2] de Jong, L., Clasen, P., MĂźller, M., RĂśmer, U. "Uncertainty analysis of limit cycle oscillations in nonlinear dynamical systems with the Fourier generalized Polynomial Chaos expansion." Journal of Sound and Vibration 607 (2025): 119017.
[3] Lux, K., Ashwin, P., Wood, R., and Kuehn, C. "Assessing the impact of parametric uncertainty on tipping points of the Atlantic meridional overturning circulation", Environ. Res. Lett. (2022) 17 075002.
Keywords:
Dynamical Systems , Sensitivity Analysis, Uncertainty Quantification
Optimization of system planning, design, manufacturing, controls, and operations are critical for ensuring that engineering systems maximize performance. However, each stage of a systemâs lifecycle is rife with uncertainty. These uncertainties can arise from uncontrolled factors such as manufacturing precision and operating conditions, leading to aleatoric uncertainty. These uncertainties can also arise because the models used for optimization have unknown parameters that require calibration, leading to the epistemic uncertainty.
As a result, optimization of engineering systems requires quantifying and accounting for a variety of (wanted or unwanted) outcomes in the presence of uncertainties. Consideration of this uncertainty is vital to ensure safety; safeguard against costly design alterations late in the design cycle; minimize the probability of operational failures; and maximize the likelihood of mission success. The traditional and widely used approach is to add safety or robustness margins to compensate for uncertainties after a deterministic optimization is performed. This produces a sense of security but is at best an imprecise recognition of the potential outcomes and results in overly conservative designs that can limit performance. Properly accounting for risk during the optimization could allow for more efficient and well-performing designs.
In this minisymposium, contributors will survey the state of the art in optimization under uncertainty for complex systems, ranging from methodological developments to challenging applications. This will cover recent developments focused on multifidelity and surrogate approaches, along with their associated efforts for efficient statistical sampling, stochastic and deterministic optimization algorithms, optimal transport theory and applications, etc. Applications are welcome that focus on design, manufacturing, trajectory optimization, and any other optimization-based tasks throughout an engineered system's lifecycle.
Keywords:
multi-fidelity methods, multi-step optimization, optimal experimental design, Optimization Under Uncertainty, risk-based design, robust optimization, surrogate models
Sensitivity analysis is a cornerstone of scientific modeling and engineering design, offering systematic ways to quantify how variations in model inputs influence outputs. It plays a crucial role in improving model understanding, guiding data collection, enhancing robustness, and supporting decision-making under uncertainty. In the current era of digital twins and scientific machine learning, the importance of sensitivity analysis has become even more pronounced. As models become increasingly complex and data-driven, the analysis offers a principled framework for combining data with mechanistic models, tackling challenges such as identifiability, model selection, and interpretability. Different methods enable rigorous assessment of how data inform model predictions, helping to bridge the gap between theory and experiment while promoting transparency and reliability in predictive modeling.
In this mini-symposium, we aim to showcase recent advances and applications of (global) sensitivity analysis in science and engineering. We welcome contributions ranging from theoretical developments and computational methods to applied case studies. In particular, we highlight (without restricting the scope to) three emerging focus areas for sensitivity analysis:
1. Sensitivity analysis with given data.
2. Identifiability and parameter estimation.
3. Explainability in scientific machine learning.
We invite researchers from academia and industry to contribute to this discussion, fostering dialogue between developers and practitioners on the evolving role of sensitivity analysis in contemporary computational challenges.
Keywords:
explainability, global sensitivity analysis, identifiability, uncertainty quantification (UQ)
Uncertainty Quantification (UQ) plays a pivotal role in computational modeling, enabling robust predictions and informed decision-making in the presence of uncertain input data. However, the high computational cost associated with repeated evaluations of complex models poses a significant challenge. Surrogate models offer a powerful solution to this issue, providing efficient and reliable approximations of full-order models. This minisymposium welcomes contributions on a broad spectrum of surrogate approaches with good error control and convergence guarantees and their use in UQ, including polynomial-based methods (e.g., stochastic Galerkin and collocation), rational approximations, Gaussian process regression and kernel methods, reduced basis techniques and dynamical low-rank approximations. Emphasis is placed to multilevel and multifidelity surrogate methods that leverage hierarchies of models or resolutions to further reduce computational effort while maintaining predictive accuracy. Applications are drawn primarily from computational mechanics (fluids and solids), quantum mechanics, and materials science.
Keywords:
multifidelity surrogates, multilevel surrogates, Uncertainty Quantification
This minisymposium focuses on uncertainty quantification for engineering systems, a domain where the computational expense of high-fidelity simulations poses significant challenges. Traditional methods as standard Monte Carlo simulation, Polynomial Chaos and Gaussian Process Regression often become prohibitively costly. This necessitates the development and application of innovative and efficient methodologies.
A particular emphasis of this minisymposium is the advancement of surrogate models. Surrogates approximate the input-output relation or other functional dependencies within a model using inexpensive, easy-to-evaluate functions. They are derived from the underlying governing equations or from a very limited number of high-fidelity simulations. Consequently, they allow to reduce the computational burden for uncertainty quantification.
This minisymposium aims to showcase the latest advancements in surrogate modelling for computationally expensive engineering simulations. In addition, the application to certification, reliability analysis and parameter identification is of interest.
Possible topics for contributions of the minisymposium are:
- Advances in surrogate modelling
- Polynomial-based surrogates
- Advanced Gaussian process regressions
- Operator-Learning
- Multi-fidelity surrogates
- Adaptive surrogates and active learning strategies
Keywords:
Reduced-Order Models, Surrogate Model, Uncertainty Quantification
Computational modeling and design require decision-making in the presence of uncertainties due to variations in material and geometric parameters, external loads, and environmental conditions. In general, data and information are characterized by aleatory uncertainty (natural variability) and epistemic uncertainty (lack of knowledge, and imprecision). For a realistic risk and structural safety assessment, both types of uncertainty characteristics have to be considered using probabilistic (random variables) and non-traditional methods (e.g., interval /fuzzy variables, and polymorphic / hybrid / mixed uncertainty models). The numerical implementation requires a computationally expensive nested loop algorithm. Therefore, the computaional design of complex structures accounting for uncertainties is quite challenging and needs ongoing research. This minisymposium addresses recent developments and current challenges in uncertainty quantification in computational mechanics. Areas of interest include, but are not limited to:
⢠Uncertainty quantification using probabilistic approaches (random variables) and
non-traditional methods (e.g., interval /fuzzy variables, and polymorphic / hybrid /
mixed uncertainty models)
⢠Random fields and processes
⢠Stochastic finite element methods
⢠Structural optimization considering uncertainties
⢠Resilience-based design
⢠Sensitivity analysis
⢠Risk analysis
⢠Surrogate modeling strategies
⢠Sampling methods
⢠AI methods and high-performance computing for UQ
⢠Applications of UQ, e.g., in solid mechanics, material modeling, stability analysis,
structural dynamics, multi-scale and multi-physics simulation
Keywords:
imprecision, probability, reliability, surrogate models, optimization, uncertainty
Reliable modeling and prediction of the response of engineering systems under uncertainty are fundamental for their design, operation, and safety assessment. However, this process is challenged in real-world applications by complex system behavior, limited and noisy data, and the high dimensionality of uncertain model parameters. Addressing these challenges requires advanced methods for quantifying uncertainties and calibrating models based on observed data. This mini-symposium focuses on recent developments in Bayesian analysis and model calibration for solving inverse problems in engineering systems under uncertainty. It aims to bring together researchers working on theoretical developments and practical applications of Bayesian methods in uncertainty quantification, propagation, and decision-making. We invite contributions that explore theoretical or computational aspects, including sampling strategies and approximation techniques for solving Bayesian inverse problems, as well as their application to uncertainty quantification in physics-based and data-driven models. Applications to digital twins, system identification, and predictive modeling in structural systems are particularly encouraged.
Keywords:
Approximation Techniques, Bayesian Analysis, Inverse Problems, Model Calibration, Sampling Strategies, Uncertainty Quantification
This minisymposium addresses generative methodologies for the automated creation of physics-based models and digital twins, with applications in early-stage engineering design, simulation-driven development, and process automation. Such methods accelerate model development by leveraging diverse information sources, including (un)structured data, symbolic or mathematical descriptions, heterogeneous sensor or process data, and domain-specific knowledge, while maintaining physical consistency, interpretability, and ideally treating (epistemic) uncertainty.
The motivation for this minisymposium stems from the growing need to develop models in settings where traditional manual modeling is slow and unflexible. Generative approaches can integrate knowledge from physics, data, and prior models into unified, adaptive representations for fast and reliable model generation.
A central theme is AI-driven generative models' rigorous verification and validation (V&V) to ensure their reliability in industrially relevant and safety-critical contexts. Verification should confirm physical consistency, numerical robustness, and the absence of hidden failure modes, while validation -guided by frameworks such as VV10 or VV20- should demonstrate predictive capability against trusted experimental or high-fidelity simulation benchmarks. We particularly encourage methods that integrate V&V and explainable AI into the generative process, making model behavior transparent, clarifying the role of physical constraints and data, and enabling continuous assessment of fidelity and reliability throughout the model lifecycle.
Relevant application areas range from medicine and biomechanics to industrial systems, including discrete and continuous manufacturing, large-scale multi-physics simulations (e.g., FEM or CFD), and scenarios aligned with industry standards such as FMI/FMU-based co-simulation. Contributions may also address integration with legacy simulation environments, such as industrial software, system simulation platforms, operational digital twins, or multi-physics codes, to enhance the interoperability and practical deployment of generative methods.
Technologies of interest include physics-informed and probabilistic machine learning, diffusion models, Bayesian model discovery, physics-constrained optimization, symbolic and (knowledge) graph-based analysis, large language models with integrated reasoning, causal modeling, and uncertainty-aware inference.
Keywords:
explainable AI, generative models, Physics-based modeling, uncertainty quantification (UQ), verification and validation
High-fidelity scale-resolving simulations of turbulent flows have an utmost importance for understanding the flow physics and achieving optimal engineering designs. Such simulation approaches which include DNS, LES, and hybrid RANS-LES [1] require (prohibitively) large computational resources. Moreover, their resulting quantities of interest are uncertain up to some extent due to various sources. Therefore, not only the accurate quantification of uncertainties for such simulations is vital, but also cost-effective techniques must be considered when addressing outer-loop problems where several flow realisations are required.
This minisymposium aims at gathering experts in the theoretical development and application of uncertainty quantification (UQ) and data-driven approaches for scale-resolving simulations of turbulent flows. The topics of interest include, but are not limited to, forward and inverse UQ problems, error estimation, data-driven optimization, multi-fidelity/multi-level models, sensitivity analysis, predictive scientific machine learning models, reduced-order and surrogate models [2]. A particular focus will be on the strategies capable of making the overall computational cost of the data-driven methods applied to scale-resolving turbulence simulations more affordable while retaining high accuracy.
Keywords:
Data-driven models, Multi-fidelity models, Uncertainty quantification (UQ), Machine Learning, Turbulence simulation
Computational models are frequently used to make predictions affecting high-consequence engineering design and policy decisions. However, simplifying assumptions (e.g., constitutive models, neglecting interactions) are often made to maintain computational tractability or due to lack of adequate information about the modeled phenomenon. These assumptions, if incorrect, can lead to model-form errors that impact model reliability. Validation and model-form uncertainty (MFU) approaches seek to quantify and assess the reliability of model predictions in the presence of model-form errors. Validation methods and processes probe the adequacy of a model for a target use case by measuring its agreement with data. MFU approaches quantify uncertainty in model predictions arising from simplifying assumptions and model-form errors. Despite advances in validation and MFU methods, quantifying model prediction reliability for predictions that extrapolate beyond available data, is an ongoing research challenge. This minisymposium presents a range of approaches to assess and quantify model trustworthiness, including: novel validation methods, processes, and metrics, especially those supporting model extrapolation beyond available data; methods to quantify model-form uncertainty, e.g., model-form error and model discrepancy approaches; and methods accounting for model misspecification in inference, design optimization, and other outer-loop processes.
Keywords:
model discrepancy, model reliability, Decision-making under uncertainty, Model-form error, Validation
The accurate prediction of hypersonic flow physics requires rigorous Verification, Validation, and Uncertainty Quantification (VVUQ) to establish credibility and guide model improvements. This minisymposium will bring together researchers and practitioners to address emerging challenges in VVUQ for computational fluid dynamics (CFD) modeling of hypersonic regimes, including strong shockâboundary layer interactions, high-temperature gas and gas-surface interaction effects, and transition to turbulence. Topics will include numerical verification, experimental validation strategies, assessment of model-form uncertainty, and statistical inference methods tailored to extreme flow conditions. The goal of this minisymposium is to foster interdisciplinary discussion on best practices, innovative methodologies, and pathways to integrate VVUQ into the design and analysis of hypersonic systems. The research topics of this minisymposium are focused on Bayesian inference and data assimilation for hypersonic model calibration, code verification and solution verification for hypersonic CFD, optimal design of experiments, uncertainty propagation in coupled fluidâthermalâstructural simulations, and VVUQ workflows for complex multi-physics simulations. Emerging topics that use AI and Machine Learning to advance research in these areas are also welcome.
Keywords:
a posteriori error estimation, AI-Powered Methods, Bayesian updating, model calibration, uncertainty quantification (UQ), verification and validation
2300
Other
This minisymposium brings together researchers and practitioners applying computational mechanics to tackle critical challenges in sustainable development. From climate change mitigation to renewable energy systems, from atmospheric protection to inclusive infrastructure design, computational mechanics has a pivotal role to play in enabling a just and sustainable future.
We invite contributions showcasing how computational techniques are being applied to real-world sustainability problems, such as noise mitigation in offshore wind farms, gas leak prevention in pipelines, or geomechanical modeling for infrastructure resilience. A central goal of this symposium is to bridge sectors, bringing together academia, industry, and government, and to foster interdisciplinary dialogue among engineers, environmental scientists, data scientists, and policy experts.
We particularly encourage submissions that:
⢠Address concrete sustainability problems through computational modeling
⢠Integrate environmental, social, or ethical dimensions into engineering workflows
⢠Present collaborative case studies involving academic, industrial, or governmental stakeholders
⢠Explore novel methodologies in optimization, multiscale analysis, and physics-based modeling
Recognizing that sustainable development requires transdisciplinary and cross-sector solutions, this minisymposium will serve as a platform not only to share technical advances but also to promote new collaborations between diverse communities committed to global impact.
Keywords:
2030 Agenda, Large-Scale Problems, Sustainable Development, Sustainability
This minisymposium is dedicated to the flexoelectric effect in electroactive materials, including rigid dielectrics, ferroelectrics, semiconductors, soft polymers, thin films, heterostructures, biological membranes, liquid crystals, and 2D materials, among others. The goal of the minisymposium is to cover a wide spectrum of recent research developments in computational, theoretical, and experimental approaches associated with flexoelectric materials, including (but not limited to):
Continuum modeling (large deformations, constitutive models, micromorphic approaches, surface effects, boundary conditions)
Computational modeling (mixed finite elements, immersed boundary method, isogeometric analysis, meshless discretizations, stabilization techniques, computational quantum mechanics)
Nonlinear rod or shell modeling
Couplings to other physics (piezoelectricity, thermal effects, magnetics, flexo-phototronics, fracture, friction, damage)
Material and structural instabilities (buckling, wrinkling, rippling)
Multi-scale modeling (FE2, homogenization, metamaterials)
Characterization (experimental methods, ab-initio methods, quantum mechanics)
Design and optimization (shape optimization, topology optimization)
Manufacturing techniques (3D printing, lithography, thermal curing, coating)
Applications (soft robotics, sensing, actuation, energy harvesting, catalysis, diodes, transistors)
Keywords:
Electromechanics, Flexoelectricity, MEMS MEMS, Nanotechnology
The realistic prediction of material behavior using computer-aided methods is essential for many engineering applications. This involves the formulation and algorithmic implementation of suitable mathematical constitutive models but, just as importantly, their calibration and validation based on experimental characterization data. Classically, the data required for the identification of model parameters is obtained through relatively simple (uniaxial; measurement of homogeneous gauge length response), standardized, macroscale experiments. More recently, the emergence of ever more complex material systems, but simultaneously much higher-fidelity experimental techniques, has offered novel opportunities, but also posed new challenges for experimentalist and material modelers alike. In the field of experimental methods such developments have included the use of specimens with heterogeneous stress and strain fields, the measurement of strain fields on the specimen surfaces by digital image correlation or the acquisition of microstructural data, e.g., by stacked microscope images or micro-CT scans. Current research is focused on integrating experiments and simulations, in a way that one discipline can inform the other. Examples include data-driven and hybrid modeling approaches and, conversely, the simulation-supported design and interpretation of experiments. Modern approaches further aim to enable ontology-based interoperability across heterogeneous data, in which information from different experimental techniques but also from numerical experiments in virtual laboratories is able to co-exist and mutually inform. This interdisciplinary Minisymposium brings together experts interested in advancing research at the intersection of experimental and computational mechanics.
Keywords:
Constitutive Modeling, Numerical Experiments, Experimental Characterization
When methods in computational mechanics leave the laboratory and enter industrial workflows, the decisive question is no longer theoretical convergence but wall-clock time: how quickly can a software deliver a result of prescribed accuracy? In practice, however, performance figures are still reported in incompatible waysâif at allâmaking it almost impossible to compare, say, different finite-element flavours, AI surrogates, GPU kernels or emerging quantum solvers on equal footing. The situation is further aggravated for transient problems: while much of todayâs research publicises ever-larger or complex simulations on heroic HPC machines, genuinely time-dependent analyses at the scaling limit receive far less attention, and progress in their time-to-solution is rarely documented. This minisymposium therefore launches a community-wide âtime-to-solutionâ challenge that puts absolute speed for transient problems centre stage while remaining method-agnostic and scientifically rigorous.
Keywords:
Bechmarks, Transient Problems
This mini-symposium invites participants to share simple analytical models that can accompany or strengthen, from a theoretical or analytical perspective, known or unknown solutions to problems of nonlinear analysis applied to the natural, social, or engineering sciences, without excluding any real-life problems that can be addressed through nonlinear analysis.
Keywords:
Complex cratering dynamics, Compressible flow models, Mechanistic Models
3000
Special Tracks: Teaching, Learning and Open-Source Software
This mini-symposium aims to bring together developers and users of open-source scientific software libraries for solid mechanics, fluid mechanics, and multiphysics applications. It will encompass discussions on software design and performance, which includes topics such as data structures, the implementation of approximation methods, adaptive hp-refinement, and high-performance computing. Additionally, the mini-symposium will delve into linear solvers for large systems of equations, meshing tools, and various pre- and post-processing tools. Intended as a forum for developers, contributors, and users of open software libraries, the mini-symposium will facilitate the sharing of knowledge and experience about modern infrastructures, emerging technologies, and algorithms. It will also cover performance profiling, testing, and development. The symposium will provide insights into the process of building and managing communities of developers and users, drawing from experiences with successful and widely-used open-source codes [1-5]: FEniCS project, deal.II, MFEM, PETSc, MoFEM, and others.
[1] AlnĂŚs, M. et al. "The FEniCS project version 1.5.", Arch. Num. Softw. 3.100:9-23
(2015).
[2] Bangerth, W. et al. "Deal.II - a general-purpose object-oriented finite element library."
ACM Transactions on Mathematical Software (TOMS) 33.4:24 (2007).
[3] Anderson, R. et al. "MFEM: A modular finite element methods library." Computers &
Mathematics with Applications, 81 (2021).
[4] Balay, S. et al. "PETSc users manual revision 3.7. No. ANL--95/11 Rev. 3.7" Argonne
National Lab, Argonne, IL (United States) (2016).
[5] Kaczmarczyk, Ĺ. et al. "MoFEM: An open source, parallel finite element library". Journal
of Open Source Software, 5(45):1441 (2020).
Keywords:
algorithms, fluid mechanics, high performance computing, multiphysics, open-source software, solid mechanics
This mini-symposium aims to provide a focused platform for presenting recent advances and fostering discussion around computational multiphysics using open-source software frameworks. Emphasis is placed on contributions that not only demonstrate progress in physical modeling and numerical algorithms but also provide insights into the design and implementation of robust, extensible, and high-performance software infrastructures.
A central objective of the mini-symposium is to showcase developments in established open-source multiphysics platformsâsuch as OpenFOAM, Alya, Sparselizard, Yales2 among othersâand to provide a forum for the communities behind these tools to exchange experiences and ideas. Topics of interest include, but are not limited to, novel coupling strategies for multiphysics problems, algorithmic improvements in time integration and solver technology, and methodological advances in partitioned and monolithic approaches. Particular attention will be given to the challenges of maintaining accuracy, stability, and scalability in strongly coupled systems.
In addition to numerical methods, the mini-symposium invites contributions on software engineering practices that enable maintainable and reusable code bases. This includes discussions on modular software architectures, data structures optimized for coupled solvers, integration workflows, and the role of community governance in sustainable development of scientific software. Contributions highlighting the use of open standards, reproducible research workflows, and interfaces for user extensibility are especially welcome.
Application-oriented presentations are encouraged, particularly where they highlight the use of open-source multiphysics tools in complex, real-world scenarios. Examples may include fluidâstructure interaction, conjugate heat transfer, magneto-hydrodynamics, or coupled processes in biomechanics, geophysics, or energy systems. Submissions that illustrate full simulation pipelinesâfrom geometry and meshing to solver execution and data analysisâare of particular interest when they provide insight into practical bottlenecks or innovative solutions.
By bringing together developers, methodologists, and advanced users, this mini-symposium seeks to promote exchange across communities and strengthen collaboration in the open-source multiphysics ecosystem. The session aims to advance shared understanding of current capabilities and identify future directions in the dev
Keywords:
Coupling, Open-Source Software, Computational Multiphysics, High-Performance Computing (HPC)
Motivation. First-year courses in Computational Mechanics, CSE, Engineering Mechanics, Applied Mathematics, and related STEM fields remain critical gatewaysâand often gatekeepersâfor diverse student populations transitioning from secondary school to university. Misalignment between studentsâ prior preparation and curricular expectations continues to impede student success. Additionally, there is often a disconnect between the mathematical concepts emphasized in lower-division mathematics courses and those actually used in engineering and the sciences. At the same time, a rapidly evolving methodological landscapeâincluding active learning, project- and problem-based learning, and inclusive teaching and assessmentâoffers concrete avenues to redesign these courses.
Scope. This mini-symposium brings together two complementary perspectives on STEM education. First, we invite contributions from undergraduate STEM education researchers (with a strong welcome to undergraduate mathematics education researchers) and from instructors who systematically study their practice. Topics include the transition from secondary school to university in STEM, research-informed innovations in teaching and assessmentâparticularly in Computational Mechanics and Mathematicsâand studies on the (mis)alignment between first-year mathematics curricula and their application in engineering and the sciences.
Second, we welcome contributions focused on the teaching and learning of Computational Mechanics, CSE, Engineering Mechanics, and Applied Mathematics at both the undergraduate and graduate levels. These may include research-informed course design, novel teaching strategies, and reflective studies on pedagogy, assessment, and student engagement in technical disciplines.
By linking these two domains, the mini-symposium aims to foster cross-disciplinary dialogue between education researchers and instructors in computational and engineering sciencesâcreating space to discuss teaching innovations across the full undergraduate and graduate spectrum.
Topics include:
⢠Transition from high school to university in mathematics and STEM courses
⢠Research on active, blended, and flipped learning in large-enrollment courses
⢠Teaching practices in Computational Mechanics, CSE, Engineering Mechanics, and Applied Mathematics
⢠Inclusion of non-traditional and historically underrepresented students; self-efficacy and identity in STEM
⢠Cross-disciplinary coordination of first-year STEM
Keywords:
active learning, AI in education, assessment for learning, curriculum alignment, inclusion and equity, self-efficacy, transition to university, Undergraduate and graduate education in STEM
