AI-Driven Predictive Maintenance Framework for Industrial Induction Motors Using Vibration and Motor Current Signature Analysis

Industrial induction motors are critical assets in manufacturing industries, and their unexpected failures can result in significant production losses, increased maintenance costs, and reduced operational efficiency. This study presents an AI-driven predictive maintenance framework for induction motors using vibration analysis and Motor Current Signature Analysis (MCSA). Experimental data were collected under healthy, bearing fault, rotor fault, and stator fault conditions, followed by signal processing and feature extraction techniques. Machine learning algorithms including Support Vector Machine (SVM), Random Forest (RF), Artificial Neural Network (ANN), and K-Nearest Neighbour (KNN) were employed for fault classification and condition monitoring. The results demonstrated that the Random Forest model achieved the highest classification accuracy of 97.8%, enabling reliable fault detection and early maintenance decision-making. Vibration analysis effectively identified mechanical faults, while MCSA successfully detected electrical abnormalities. The proposed framework improved diagnostic accuracy, equipment reliability, and maintenance planning efficiency. The study highlights the potential of integrating artificial intelligence with condition monitoring technologies to support predictive maintenance and smart manufacturing environments.