International Journal of Scientific Research and Engineering Development

Industrial equipment failures account for billions of dollars in unplanned downtime losses annually, yet the majority of manufacturing enterprises still depend on reactive or time-based scheduled maintenance strategies that offer no real-time fault anticipation capability. This paper presents a Predictive Maintenance Framework (PMF), a comprehensive intelligent system that continuously monitors the health of industrial machinery through distributed IoT sensor networks, applies machine learning algorithms to raw sensor streams to recognise fault patterns, and generates proactive maintenance alerts before equipment failures occur. The framework is engineered with a React.js-based operator dashboard, a Python FastAPI dataprocessing backend, and a time-series database (InfluxDB) for high-frequency sensor data persistence, following a scalable edge-cloud hybrid architecture. A defining innovation of this work is the direct integration of artificial intelligence into maintenance operations: a multi-model anomaly detection engine analyses vibration, temperature, pressure, and acoustic emission data to identify developing faults with high precision, while a natural language diagnostic assistant allows maintenance engineers to query the equipment's health status in a conversational manner. Remaining Useful Life (RUL) estimation is maintained through a Long Short-Term Memory (LSTM) neural network trained on historical degradation patterns. Role-based access control enforces security across three operational roles — System Administrator, Maintenance Engineer, and Equipment Operator — through JWT-based authentication. The system additionally supports automated maintenance work order generation, equipment history logging, real-time threshold alerting, and multi-plant dashboard visibility. Evaluation confirms a 95.6% fault detection accuracy, a 78% reduction in unplanned downtime incidents, and a System Usability Scale score of 84.1, collectively validating the framework as a deployable, enterprise-grade solution for industrial predictive maintenance intelligence.

Dr.A.Karunamurthy, S.Rithikvasan,"Predictive Maintenance Using Artificial Intelligence and IoT Sensors: A Smart Industrial Framework for Real-Time Fault Detection and Equipment Health Monitoring" International Journal of Scientific Research and Engineering Development, V9(3): Page(2082-2091) May-June 2026. ISSN: 2581-7175. www.ijsred.com. Published by Scientific and Academic Research Publishing.