Interpretability for Industry 4.0 : Statistical and Machine Learning Approaches

Antonio Lepore is an Associate Professor of Statistics for Experimental and Technological Research (SECS-S/02) in the Department of Industrial Engineering of the University of Naples Federico II. His research interests and publications in international journals focus on the use of statistical methods for the analysis and monitoring of functional data aimed at the interpretation of complex data coming from high-frequency multi-sensor data acquisition systems. He is a member of the ENBIS (European Network for Business and Industrial Statistics) and SIS (the Italian Statistical Society). Biagio Palumbo is an Associate Professor of Statistics for Experimental and Technological Research (SECS-S/02) in the Department of Industrial Engineering of the University of Naples Federico II and President Elect of the European Network for Business and Industrial Statistics (ENBIS). His research interests are in interpretable statistical learning techniques for industrial engineering and, in particular, for the monitoring of complex data coming from high-frequency multi-sensor acquisition systems and for optimization of manufacturing processes. He is member of the Italian Statistical Society, the American Society for Quality (ASQ), and the Italian Association of Mechanical Technology.   Jean-Michel Poggi is a Professor of Statistics at Université Paris Cité and a member of the Lab. Maths Orsay (LMO) at Université Paris-Saclay, in France. His research interests are in nonparametric time series, wavelets, tree-based methods (CART, Random Forests, Boosting) and applied statistics. His work combines theoretical and practical contributions with industrial applications (mainly environment and energy) and software development. He is Associate Editor of three journals: the Journal of Statistical Software (JSS), Advances in Data Analysis and Classification (ADAC) and the Journal of Data Science, Statistics, and Visualisation (JDSSV). He is President of the European Network for Business and Industrial Statistics (ENBIS).  

- 1. Different Views of Interpretability. - 2. Model Interpretability, Explainability and Trust for Manufacturing 4.0. - 3. Interpretability via Random Forests. - 4. Interpretability in Generalized Additive Models.