Machine learning and artificial intelligence in polymer matrix composite materials: A review

This work reveals the transformative impact of artificial intelligence (AI) and machine learning (ML) techniques in the design, analysis, and performance prediction of composite materials. The review we examined systematically evaluated current AI and ML approaches used in composite material research published between 2005-2024. The main application areas of these approaches are determined as mechanical property prediction, process optimization and structural health monitoring. Analyses have shown that a wide range of models, including Artificial Neural Networks (ANN), Support Vector Machines, Decision Trees and hybrid models, have been successfully applied in composite systems. According to the findings of the study, ML models offer remarkable effectiveness in predicting key mechanical properties such as tensile strength, impact resistance and fatigue behavior, providing accuracy of over 90% in many studies compared to traditional methods. However, limited data access, low interpretability of models and high computational costs are emerging as the main obstacles to the widespread use of these methods on an industrial scale. Future research is expected to focus on developing explainable AI approaches, integrating real-time sensor data into models, and supporting sustainable material design strategies. By integrating data-based methods with materials science, it is concluded that artificial intelligence and machine learning techniques have an important transformative potential in terms of accelerating innovation, reducing costs and improving performance in the development of next-generation composite structures.