Industrial electric motor failures impose devastating economic burdens on manufacturing enterprises, with Fortune Global 500 companies losing 1.4 trillion annually toun planned downtime. This research presents a multi modal predictive maintenance framework integrating tri−axial accelerometers, current sensors, and acoustic monitoring for comprehensive motor health assessment across AC, DC, BLDC motors, and alternators. The system employs sophisticated frequency domain feature extraction with motor−specific fault frequency analysis and Random Forest classification optimized through Bayesian hyper parameter tuning. Experimental validation across 240 hours of operation on 12 diverse motor configurations demonstrates precision and 93.1 2. 1 - 4. 8 million annual savings for medium-sized facilities.