Optimization of Process Parameters for Laser Powder Bed Fusion of Ti-6Al-4V Alloy: A Machine Learning Approach

• Machine learning models, particularly random forest, accurately predict density and microhardness of LPBF Ti-6Al-4V parts, with R² values above 0.95. • Multi-objective optimization using genetic algorithms identifies optimal process parameters that achieve near-full density (99.8%) and high microhardness (390 HV). • The proposed methodology reduces experimental effort by up to 70% compared to traditional design of experiments, significantly lowering cost and time. • The optimized parameters are validated experimentally, confirming the reliability of the machine learning approach for industrial application.