Abstract
Smart industrial systems must balance equipment reliability, energy use, maintenance cost, production continuity, and technical resources. This study proposes a multi-objective predictive maintenance and energy-aware resource optimization framework for industrial decision-making. The framework combines equipment health assessment, Remaining Useful Life (RUL) estimation, anomaly detection, failure probability, and maintenance prioritization. K-means clustering determines the Anomaly Level (AL), while Weibull survival analysis estimates Failure Probability (FP). These measures form a Hybrid Risk Index (HRI) for determining the Optimal Maintenance Point (OMP). NSGA-II evaluates maintenance timing, asset selection, and technician allocation under maintenance cost, energy, quality, and production objectives. Performance is compared with preventive maintenance, condition based maintenance, and reliability only optimization under changing industrial conditions.