Energy-Efficient Condition-Based Maintenance: A Smart Framework for Predictive Decision-Making in Industry 4.0

Abstract

This study presents an energy-aware Condition-Based Maintenance (CBM) framework for an SKF 6205 bearing in a motor-driven system, integrating Industry 4.0 technologies. Real-time data from ESP32-based IoT sensors enabled degradation modeling using a Gamma process and evaluation of energy efficiency. The degradation index (Xt), derived from tri-axial RMS vibration, identified a failure threshold of 41.5 g·min, with a CBM trigger set at 75% (31.13 g·min). An Energy Efficiency Indicator (EEI), defined as the ratio of power input to incremental degradation, highlighted performance drops near failure, validating the thresholds. Remaining Useful Life (RUL) was estimated using Maximum Likelihood Estimation on Gamma distribution parameters (α = 16.61, β = 0.000286). The proposed approach links energy efficiency with wear progression, enabling accurate, sustainable maintenance decisions in smart manufacturing environments.