Optimal Preventive Maintenance Scheduling Based on Reliability-Centered Maintenance (RCM) for Main Production Machines

Abstract

This study develops an integrated Reliability-Centered Maintenance (RCM) framework combining Failure Mode and Effects Analysis (FMEA), Weibull reliability modeling, and cost-based preventive-maintenance interval optimization for critical production machinery. Using 48 corrective-maintenance records collected between March 2021 and December 2024 from five critical components, namely the Hydraulic Pump Unit, Barrel Heater Band, Screw and Barrel Assembly, Mold Clamping Unit, and Cooling Water Pump, components were prioritized through composite criticality scoring together with FMEA-based Risk Priority Number and Action Priority assessment. Two-parameter Weibull analysis, validated by Kolmogorov-Smirnov tests, confirmed that all components exhibit wear-out behavior with shape parameters greater than one. The RCM logic classified every component under scheduled restoration or discard, and an age-replacement cost model determined reliability-constrained optimal preventive-maintenance intervals. Results show the optimal intervals are fifty-five to seventy-one percent shorter than existing calendar-based intervals, indicating systematic under-maintenance in current practice. Four of the five components achieved substantial improvements, with cost reductions of 34.9 to 52.6 percent and failure-event reductions of 61.2 to 79.5 percent, while the Barrel Heater Band, having the lowest shape parameter, showed a cost increase despite higher reliability. Sensitivity analysis confirmed the robustness of optimal intervals to cost variations but revealed heightened sensitivity to reliability-target changes for components with near-random failure behavior. This framework offers manufacturers a quantitative, data-driven approach for translating historical failure records into economically justified, component-specific maintenance intervals without requiring continuous sensor infrastructure.