A Systematic Literature Review on Machine Learning Approaches for Quality Monitoring and Control Systems for Welding Processes

Welding is an indispensable manufacturing process widely used across different commercial industries. Embedded monitoring and control systems are necessary to make this process more maintainable, error-free, and cost-effective. This paper reviews literature about machine learning approaches for improving embedded quality monitoring and control systems for welding processes. Based on the literature included in international peer-reviewed journals and conferences, we build a comprehensive overview of pre-process, in-process, and post-process approaches for established welding techniques. Furthermore, we motivate future research for technologies that enable detecting defects in the manufacturing environment.