Explainable machine learning models for defects detection in industrial processes

Abstract

In Industry 4.0, machine learning (ML) algorithms for non-linear pattern recognition are transforming defect detection in manufacturing, enabling enhanced quality monitoring, operational efficiency, and competitiveness. Explainable Artificial Intelligence (XAI) can facilitate understanding how the models make decisions and assist in tracking anomaly points. This research proposes an XAI framework for ML models in defect detection. This work applied the multinational industry's approach to the tire manufacturing process.