Maintenance Strategy Selection and Its Impact on the Optimization of Life Cycle Costs in Industrial Systems

This paper proposes to examine the impact of corrective, preventive, and predictive maintenance strategies on the Life Cycle Cost (LCC) of industrial equipment. A baseline assessment of current Reliability, Availability, and Maintainability (RAM) and LCC was performed using real maintenance and failure data, supported by a cost-based model covering acquisition, operation, downtime, and repair costs. To enhance decision-making, a Random Forest model was applied as a machine learning tool to identify the most appropriate maintenance strategy for each subsystem or critical spare part. By combining RAM indicators with cost data, the model enabled the optimal assignment of maintenance types, linking technical performance to economic outcomes. Results show that transitioning from corrective to preventive maintenance improves Mean Time Between Failures (MTBF) by 15% and reduces Mean Time to Repair (MTTR) by 10%, lowering LCC by 8–10%, whereas moving further to predictive maintenance enhances MTBF by 20% and decreases MTTR by 15%, achieving up to an 18% additional LCC reduction. Although preventive and predictive strategies require higher initial investments, they significantly reduce downtime and unplanned failures compared with corrective approaches. Overall, this work highlights the value of integrating RAM–LCC analysis with machine learning to guide maintenance strategy selection and demonstrates a practical path toward improving reliability, cost efficiency, and sustainability in Industry 4.0.