Course

Unit 1

Introduction to AI in Manufacturing: Overview of AI and its applications in manufacturing, Challenges and opportunities of AI adoption in manufacturing, Case studies highlighting AI use cases in different manufacturing sectors. Fundamentals of Machine Learning: Basic concepts of machine learning, supervised, unsupervised, reinforcement learning, model evaluation metrics. Challenges of Black-Box AI Models: Lack of transparency and interpretability, Risks associated with black-box models in manufacturing, Regulatory requirements, and ethical considerations.

Unit 2

Explainable AI Techniques: Interpretable models (e.g., Decision Trees, linear models), Post-hoc explanations – SHapley Additive exPlanations (SHAP), Local Interpretable Model-agnostic Explanations (LIME), Rule-based systems, Model distillation, and simplification techniques. Interpretability in Deep Learning: Challenges of interpreting deep neural networks, Techniques for explaining deep learning models – layer-wise relevance propagation, Visualization methods for understanding deep learning processes. Applications of XAI in Manufacturing: Predictive maintenance and fault detection, Quality control and defect detection, Process optimization and scheduling, and Supply chain management.

Unit 3

Case Studies and Practical Applications: Hands-on exercises with XAI libraries in Python, analysis of real-world manufacturing datasets, and application of XAI techniques to solve manufacturing problems. Evaluation and Trade- offs: Quantitative and qualitative evaluation of XAI techniques, Trade-offs between model performance and interpretability, Strategies for selecting appropriate XAI methods based on application requirements, Future Directions and Emerging Trends: Advances in XAI research for manufacturing, Integration of XAI with autonomous systems and robotics, Ethical considerations and responsible AI practices in manufacturing.

Course Objectives

• Make students’ understand the Explainable AI (XAI) techniques and their application in Students will learn the fundamentals of AI, explore its relevance in manufacturing processes,
• Familiarize on methods for ensuring transparency, interpretability, and accountability in AI models used in manufacturing environments.

Prerequisites

• Basic understanding of statistics, Familiarity with programming languages (e.g., Python) and basic knowledge of machine learning concepts

Course Outcomes

Skills Acquired

Implementation of AI concepts for the performance enhancement of manufacturing systems.

1. Molnar, “Interpretable machine learning. A Guide for Making Black Box Models Explainable”, 2019
2. Uday Kamath, and John Liu, “Explainable Artificial Intelligence: An Introduction to Interpretable Machine Learning”, Springer, 2021, ISBN 9783030833558
3. Leonida Gianfagna and Antonio Di Cecco, “Explainable AI with Python”, Springer International Publishing, First Edition, 2021
4. Denis Rothman, “Hands-On Explainable AI (XAI) with Python”, Packt Publishing, First Edition, 2020