Explainability for Fault Detection System in Chemical Processes

2026-02-18

AI Summary

This study addresses the lack of interpretability in fault detection for chemical processes by proposing a diagnostic framework that integrates a high-accuracy LSTM classifier with explainable artificial intelligence (XAI) techniques. It presents the first systematic comparison of Integrated Gradients (IG) and SHAP in interpreting faults within complex nonlinear chemical processes, specifically using the Tennessee Eastman benchmark. The evaluation demonstrates that SHAP more accurately identifies root causes and key contributing variables. The resulting model-agnostic explainability framework not only effectively localizes faulty subsystems but also exhibits strong generalization across diverse operational scenarios. This approach provides reliable and transparent decision support for safety monitoring in industrial processes, enhancing both trustworthiness and practical utility in real-world applications.

Technology Category

• Machine Learning: Transparent, Interpretable, Explainable ML
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Humans and AI: Explainable AI (XAI) for Human Understanding

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

In this work, we apply and compare two state-of-the-art eXplainability Artificial Intelligence (XAI) methods, the Integrated Gradients (IG) and the SHapley Additive exPlanations (SHAP), that explain the fault diagnosis decisions of a highly accurate Long Short-Time Memory (LSTM) classifier. The classifier is trained to detect faults in a benchmark non-linear chemical process, the Tennessee Eastman Process (TEP). It is highlighted how XAI methods can help identify the subsystem of the process where the fault occurred. Using our knowledge of the process, we note that in most cases the same features are indicated as the most important for the decision, while insome cases the SHAP method seems to be more informative and closer to the root cause of the fault. Finally, since the used XAI methods are model-agnostic, the proposed approach is not limited to the specific process and can also be used in similar problems.

Problem

• Explainability
• Fault Detection
• Chemical Processes
• XAI
• Tennessee Eastman Process

Innovation

• Explainable AI
• Fault Detection
• LSTM
• SHAP
• Integrated Gradients