🤖 AI Summary
Complex industrial process data hinder fault detection model performance and interpretability, impeding reliable decision-making. To address this, we propose the first interpretable fault detection framework integrating SHAP values with causal graph modeling: SHAP quantifies feature contributions, while multi-algorithm causal discovery constructs a directed acyclic graph (DAG) to explicitly encode fault propagation mechanisms. Evaluated on the Tennessee Eastman Process benchmark, our method significantly improves fault identification accuracy and precisely identifies key causal drivers—such as cooling and separation systems—enabling root-cause attribution. Crucially, the learned causal structure aligns closely with SHAP-based feature importance rankings. This work achieves synergistic enhancement of predictive performance and causal interpretability, establishing a novel diagnostic paradigm for industrial intelligent maintenance that balances accuracy with operational transparency.
📝 Abstract
Industrial processes generate complex data that challenge fault detection systems, often yielding opaque or underwhelming results despite advanced machine learning techniques. This study tackles such difficulties using the Tennessee Eastman Process, a well-established benchmark known for its intricate dynamics, to develop an innovative fault detection framework. Initial attempts with standard models revealed limitations in both performance and interpretability, prompting a shift toward a more tractable approach. By employing SHAP (SHapley Additive exPlanations), we transform the problem into a more manageable and transparent form, pinpointing the most critical process features driving fault predictions. This reduction in complexity unlocks the ability to apply causal analysis through Directed Acyclic Graphs, generated by multiple algorithms, to uncover the underlying mechanisms of fault propagation. The resulting causal structures align strikingly with SHAP findings, consistently highlighting key process elements-like cooling and separation systems-as pivotal to fault development. Together, these methods not only enhance detection accuracy but also provide operators with clear, actionable insights into fault origins, a synergy that, to our knowledge, has not been previously explored in this context. This dual approach bridges predictive power with causal understanding, offering a robust tool for monitoring complex manufacturing environments and paving the way for smarter, more interpretable fault detection in industrial systems.