Implementation of Machine Learning for Enhancing Lean Manufacturing Practices for Metal Additive Manufacturing

Abstract

Metal additive manufacturing (AM), particularly laser powder bed fusion (LPBF), has emerged as a promising technology for rapidly producing intricate parts while minimizing material waste. However, the widespread adoption of AM has been hindered by the lack of adequate quality control measures. To address this challenge, a large number of machine learning (ML) applications have been proposed to improve the quality and productivity of AM processes. This study proposes the Lean concept as a guiding framework for classifying ML applications according to the Lean principles they support. Through a comprehensive review of literature studies, the research demonstrates the efficacy of this holistic approach, emphasizing ML's contributions to the Lean principles and the derived benefits to refine metal AM practices, improve efficiency, foster continuous improvement in LPBF, and finally bring value to the customer. The obtained results are particularly important for manufacturing engineers, quality control specialists, and decision-makers in the AM industry, as they provide actionable insights for enhancing process reliability, reducing waste, and achieving higher productivity.