The manufacturing industry is rapidly evolving through Industry 4.0 technologies, where Artificial Intelligence (AI) enables predictive maintenance, quality control, and process optimization. Yet, the widespread use of “black-box” models limits trust, interpretability, and human collaboration. Explainable Artificial Intelligence (XAI) addresses this challenge by providing transparency and accountability, aligning AI-driven decisions with the principles of Operational Excellence.

This thesis aims to apply and evaluate XAI techniques in industrial contexts to enhance decision-making and operational efficiency. The research will focus on implementing interpretable machine learning models (such as SHAP, LIME, etc.) within selected manufacturing processes. Their impact on performance, explainability, and user trust will be assessed through tailored KPIs.

This thesis offers opportunities to work abroad, as the project will be conducted in collaboration with a University in Germany. This international experience will provide access to industrial data, exposure to applied research, and a valuable professional network within the Italo-German industrial landscape. By bridging data science, industrial engineering, and management, this thesis will contribute to a new paradigm of transparent and human-centered AI for manufacturing.