MACHINE LEARNING-BASED DEFECT PREDICTION AND PARAMETER OPTIMIZATION IN METAL ADDITIVE MANUFACTURING

Abstract

Metal additive manufacturing has revolutionized industrial production by enabling complex geometries, reduced material waste, and rapid prototyping. However, defects such as porosity, cracks, lack of fusion, and dimensional inaccuracies continue to limit part quality and production reliability. This paper proposes a machine learning-based framework for defect prediction and process parameter optimization in metal additive manufacturing by integrating Industrial Internet of Things (IIoT), multi-sensor data acquisition, cloud-based analytics, digital twin technology, predictive modeling, and adaptive process control. The proposed methodology continuously analyzes manufacturing data, predicts defect formation, optimizes laser power, scan speed, hatch spacing, and layer thickness, and supports real-time quality improvement. Experimental evaluation demonstrates improvements in defect prediction accuracy, dimensional precision, surface quality, process stability, production efficiency, and material utilization. The proposed framework provides a scalable and intelligent solution for smart metal additive manufacturing, supporting Industry 4.0 implementation and sustainable digital manufacturing. Keywords— Metal Additive Manufacturing, Machine Learning, Defect Prediction, Process Optimization, Digital Twin, IIoT, Smart Manufacturing, Industry 4.0.