Application of Artificial Intelligence in the Material Characterization of Various Types of Concrete: A Comprehensive Review

The integration of artificial intelligence (AI) into the field of civil engineering and materials science has catalyzed a change in basic assumptions in the characterization, design, and optimization of concrete. This comprehensive review paper synthesizes the current state-of-the-art applications of AI technologies, including machine learning (ML), deep learning (DL), artificial neural networks (ANNs), and large language models (LLMs), across a diverse spectrum of concrete types. These encompass conventional, high-performance (HPC), ultra-high-performance (UHPC), self-consolidating (SCC), rubberized, geopolymer, alkali-activated, pervious, and prestressed concrete, among others. The primary focus is on AI’s role in predicting key mechanical properties (compressive, flexural, and tensile strength), optimizing complex mix designs, simulating long-term structural behavior, and enhancing sustainability. The reviewed studies report varying levels of predictive performance depending on the concrete type, target property, dataset characteristics, and evaluation procedure. The reported performance is assessed using appropriate indicators, including the coefficient of determination (R2), root mean square error (RMSE), mean absolute error (MAE), and classification accuracy. Furthermore, it highlights the emerging trends of explainable AI (XAI) for model interpretability and the potential of LLMs in risk assessment and decision support. The reviewed studies demonstrate the potential of AI techniques to support concrete property prediction, mixture design, and material characterization under specific experimental and computational conditions. However, the reliability and broader applicability of these approaches depend on data quality, appropriate validation procedures, model interpretability, and experimental verification. Further research is needed to establish their generalizability and practical effectiveness across different concrete materials and engineering applications. It concludes by affirming that the continued evolution of AI promises to further redefine material characterization processes, paving the way for smarter, more resilient, and environmentally responsible infrastructure.