Machine Learning-Based Optimization of Temperature, Speed, and Layer Thickness in FDM 3D Printing

Fused Deposition Modeling (FDM) is a widely used additive manufacturing technique due to its low cost and ease of operation; however, product quality strongly depends on process parameters such as extrusion temperature, printing speed, and layer thickness. Traditional trial-and-error optimization methods are time-consuming and lack generalizability. Recent advances in machine learning (ML) enable predictive and adaptive optimization of FDM parameters, offering improved control and reproducibility. This review summarizes the application of supervised learning, deep learning, and reinforcement learning for multi-objective optimization of key FDM parameters. It further highlights the positive effects of ML-based optimization on mechanical properties, dimensional accuracy, surface quality, and process sustainability. Finally, the integration of ML with IoT-enabled smart printers and digital twin technologies is discussed, demonstrating the potential for intelligent and autonomous FDM manufacturing systems.