EXPLAINABLE ARTIFICIAL INTELLIGENCE FOR PROCESS DEFECT DIAGNOSIS IN ADDITIVE MANUFACTURING

Abstract

Additive manufacturing has become a key enabling technology for Industry 4.0 by supporting the production of complex components with high design flexibility and reduced material waste. However, manufacturing defects such as porosity, lack of fusion, cracking, residual stresses, and dimensional inaccuracies continue to affect product quality and process reliability. This paper proposes an Explainable Artificial Intelligence (XAI)-based framework for process defect diagnosis in additive manufacturing by integrating machine learning, explainable AI, Industrial Internet of Things (IIoT), multi-sensor data fusion, cloud computing, digital twin technology, predictive analytics, and adaptive process control. The proposed methodology continuously monitors manufacturing operations, diagnoses process defects, explains prediction outcomes, and recommends optimized manufacturing parameters for quality improvement. Experimental evaluation demonstrates improvements in defect diagnosis accuracy, model interpretability, production stability, equipment utilization, surface quality, and manufacturing efficiency. The proposed framework provides a scalable and intelligent solution for transparent, trustworthy, and sustainable additive manufacturing under Industry 4.0 environments. Keywords— Explainable Artificial Intelligence, Additive Manufacturing, Defect Diagnosis, Machine Learning, Digital Twin, IIoT, Smart Manufacturing, Industry 4.0.