Many structural components employed in the transportation industry rely on composite laminates, which are subjected to complex loading conditions. Advanced simulations at the meso-scale, i.e., at the level of the ply, allow for accurate predictions of the progressive failure of composite structures. These ingredients make this kind of models particularly interesting for the virtual testing of composite structure, which supports the design and optimization of structural details, towards more efficient vehicles and aircrafts. At the same time, the ply properties depend on their microstructure, i.e., on the fibres and matrix interactions. To account for these, high-fidelity micro-mechanical numerical models of composite materials can provide important insights on the micro-mechanisms that lead to the onset and evolution of fracture in composite materials [1]. This kind of model is not suitable for the direct finite element simulation of structural details, due to their computational cost. In this contribution, a first scale transition is performed by using micro-mechanical models to determine homogenised ply properties and build databases of homogenised responses for data-driven surrogate models [2], which quickly predict the ply properties from micro-descriptors. The second-scale transition is accomplished by using such properties in coupon (meso-scale) simulations. Similar approach is taken: finite element simulations are used to build databases of coupon strengths, related to laminate descriptors and ply properties, which are employed to train surrogate models linking the meso (ply) to the macro (coupon)-scale. The combination of both surrogate models yields a multi-scale one, linking the microstructure to the laminate strength. This work has received funding from the European Union’s Horizon Europe research and innovation programme under Grant Agreement No. 101056682 (DIDEAROT project). Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union. Neither the European Union nor the granting authority can be held responsible for them. REFERENCES [1] Vallmajó O, Arteiro A, Guerrero JM, Melro AR, Pupurs A, Turon A. Micromechanical analysis of composite materials considering material variability and microvoids. Int J Mech Sci. 2024;263:108781. [2] Mirkhalaf M, Rocha I. Micromechanics-based deep-learning for composites: Challenges and future perspectives. Eur J Mech - A/Solids. 2024;105:10524