Explainable Artificial Intelligence in Rotating Machinery Fault Diagnosis: A Comprehensive Review and Emerging Trends

Abstract

Highlights This review systematically surveys industrial explainable AI (XAI) techniques for rotating machinery fault diagnosis, with a balanced coverage of ante-hoc and post-hoc interpretable methods. It identifies critical research challenges and real-world deployment bottlenecks, and outlines future prospects for constructing robust, explainable diagnostic frameworks. What are the main findings? Industrial explainability requires transparency, comprehensibility, and credibility, with explanations aligned with physical fault mechanisms. Ante hoc and post hoc methods are complementary, but industrial deployment remains limited by inadequate benchmarks, robustness, and actionable outputs. What are the implications of the main findings? Standardized evaluation frameworks and lightweight, robust explainable models are needed for complex industrial conditions. Integrating physical knowledge, causal inference, and digital twins can support trustworthy diagnosis and actionable maintenance decisions. Abstract Rotating machinery is essential to energy, aerospace, manufacturing, and other safety-critical industries. Although deep-learning-based fault diagnosis has achieved high accuracy, its black-box nature limits trust, verification, and industrial deployment. This review examines explainable artificial intelligence for rotating machinery fault diagnosis and classifies existing methods into ante hoc and post hoc approaches according to their integration with model architectures. Their physical interpretability, applicable fault scenarios, explanation quality, computational overhead, robustness, and edge-deployment potential are critically compared. Quantitative criteria, including fidelity, stability, robustness, localization, and physical consistency are discussed to support objective evaluation of explanations. The review further highlights the gap between laboratory validation and industrial operation, particularly under sensor degradation, electromagnetic interference, variable working conditions, limited computing resources, and scarce fault data. It also discusses how model-relative explanations can be mapped to calibrated vibration quantities, fault-characteristic frequencies, industrial diagnostic standards, and actionable maintenance decisions. The distinction between correlation-based attribution and causal root-cause analysis is clarified, together with the role of digital twins and human-in-the-loop decision support. Finally, future research priorities are identified in standardized benchmarking, robust lightweight models, causal reasoning, and human-centered industrial deployment.