Machine and Deep Learning Approaches for Advanced Process Monitoring in Quality 5.0

Quality 5.0 represents the evolution of quality management by integrating automated systems, advanced data-driven methodologies, and human expertise to proactively improve manufacturing efficiency and decision-making. Modern industrial operations rely heavily on sensors that continuously produce high-dimensional data, frequently surpassing the analytical capacities of traditional quality monitoring tools. Classical monitoring techniques, originally designed for simple, low-dimensional datasets, face significant limitations when confronted with the complexity and scale of contemporary manufacturing data. Machine learning (ML) and deep learning (DL) approaches offer powerful alternatives capable of managing these vast datasets, detecting subtle and complex patterns, and performing effective unsupervised monitoring. This chapter discusses how ML and DL techniques are specifically utilized within the Quality 5.0 framework to overcome the limitations of conventional methods. It emphasizes essential considerations such as model selection, data preprocessing methods, and strategies for real-time integration into manufacturing processes. The chapter further examines how various ML models address critical challenges in quality monitoring, highlighting the roles of supervised, unsupervised, and self-supervised learning paradigms. Practical implementation strategies are illustrated through case studies, demonstrating how these advanced technologies enhance real-time decision-making, operational optimization, and system adaptability. Ultimately, this chapter emphasizes the synergy of intelligent analytics and human judgment in achieving proactive, robust, and sustainable quality control within Industry 5.0.