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 adaptivity—the ability of computational models to account for uncertain and evolving environments—positioning 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.