Predictive Modeling and Optimization of Additive Manufacturing and Laser Cladding Processes Using Machine Learning and Artificial Intelligence

Rapid advances in additive manufacturing (AM) and laser cladding enable precise fabrication and repair of complex parts with high material efficiency. These processes involve intricate thermophysical phenomena—such as phase changes, rapid melting/solidification, and non-linear material–process interactions—that are challenging to predict accurately using conventional analytical or physics-based models. Artificial intelligence (AI) and machine learning (ML) offer powerful solutions by enabling surrogate models that predict melt pool dimensions, microstructure, defect formation, and mechanical properties with minimal computational cost. Techniques such as deep learning, reinforcement learning (RL), and evolutionary algorithms are increasingly applied for multi-objective process optimization, improving quality while reducing waste and experimentation. This chapter reviews AI/ML algorithms (regression, decision trees, neural networks, and genetic algorithms), data sources (sensor, simulation, and experimental), and applications in intelligent control, monitoring, and quality prediction. It also presents case studies on AI-enabled digital twins and real-time feedback while addressing challenges like data scarcity, model generalizability, and interpretability for Industry 4.0 adoption.