Integrating Machine Learning for Automated and Adaptive Quality Decisions in Manufacturing

Abstract

This paper explores the integration of machine learning (ML) techniques into manufacturing processes to enable automated and adaptive quality decision-making. By leveraging real-time data and advanced analytics, ML models can predict defects, optimize process parameters, and facilitate dynamic adjustments to maintain product quality. We review existing ML methodologies applicable to manufacturing quality control and propose a modular framework that combines predictive modeling with feedback loops for adaptive decision-making. Case studies demonstrating practical implementations highlight the benefits and challenges of this approach. Finally, we discuss future directions including edge computing integration and explainability in ML-driven quality systems. Our findings emphasize that ML-based adaptive systems significantly enhance manufacturing efficiency, reduce defects, and support continuous improvement. Machine learning (ML) is transforming quality control in manufacturing. Other areas, such as automotive, semiconductor, and pharmaceutical industries, have also been improved greatly, with 97 % of defect detection accuracy in automotive components and 12 % increase in yield in semiconductor manufacturing. ML can minimize flaws and enhance productivity as well as create adaptive responses to real-time decision making. The ML systems have feedback loops that constantly enhance production by minimizing downtime and improving the ultimate quality.