A Review of Artificial Intelligence in Composite Finite Element Impact Damage Modelling: A Five-Role Lifecycle Framework

Abstract Finite Element (FE) simulation of impact damage in fibre-reinforced composite laminates is constrained by high computational cost, many difficult-to-calibrate material parameters and the absence of experimental evidence for different composite structures. Artificial Intelligence (AI) and Machine Learning (ML) methods are increasingly used to address these limitations. This review paper organises the field using a five-role lifecycle framework: Constitutive and Damage Modelling, Parameter Calibration, Computational Acceleration, Verification and Validation and Design Optimisation. The framework maps how each stage of the FE workflow can support the AI and ML approach and identifies the connections between the different roles. For each role, the current state of the art is assessed, gaps in knowledge are identified and where appropriate, example studies are highlighted.