EXPLAINABLE DATA-DRIVEN FAULT DIAGNOSIS USING IOT SENSOR FUSION AND INDUSTRIAL DIGITAL TWINS

Abstract

Industrial equipment reliability is a critical requirement for modern smart manufacturing systems, where unexpected failures can lead to production losses, safety risks, and increased operational costs. Traditional fault diagnosis methods often depend on manual inspection and predefined rules, limiting their ability to analyze complex equipment behavior and large-scale sensor data. This paper proposes an Explainable Data-Driven Fault Diagnosis Framework Using IoT Sensor Fusion and Industrial Digital Twins by integrating Industrial Internet of Things (IIoT), multi-sensor data fusion, machine learning, explainable artificial intelligence (XAI), and Digital Twin technology. The proposed framework continuously collects equipment data from heterogeneous IoT sensors, performs intelligent feature extraction, identifies abnormal conditions, classifies faults, and provides interpretable diagnostic explanations. Digital Twins enable real-time synchronization between physical assets and virtual models for improved fault analysis and prediction. Experimental evaluation demonstrates improvements in fault detection accuracy, diagnostic transparency, response efficiency, equipment reliability, and predictive maintenance capabilities. The proposed framework provides a trustworthy and scalable solution for Industry 4.0 intelligent fault diagnosis.