An integrated model for predictive maintenance and inventory management under a reliability chance constraint

This paper proposes a new model that integrates opportunistic maintenance and routine maintenance to enhance the effectiveness of predictive maintenance and inventory management in complex manufacturing systems subject to a reliability chance constraint. It considers both hard and soft failure modes and their mutual dependence. When a machine experiences a hard failure, an opportunistic maintenance policy is utilized on the machine’s components. When the soft failure degradation level of a machine component surpasses a threshold, imperfect preventive maintenance or replacement maintenance is carried out. The choice of component supplier, including OEM and aftermarket suppliers, significantly impacts the joint decision model. To improve the model’s realism and applicability, a random variable representing supplier availability intervals is introduced, reflecting a more nuanced understanding of supply chain dynamics. We develop a simulation optimization method to determine the degradation thresholds for opportunistic and regular maintenance, the component inventory policy, and supplier selection. The objective is to minimize the total maintenance and inventory cost, while ensuring a high level of system reliability. The proposed algorithm effectively addresses the system reliability chance constraint by formulating a surrogate model of the quantile of system downtime. A numerical study is conducted to verify the efficacy of the proposed model and to demonstrate the efficiency of the solution method in finding the optimal feasible solution. Furthermore, the influence of critical factors in the model on the optimal policy is analyzed to derive useful managerial insights.