A dynamic inspection and replacement policy for a two-unit production system subject to interdependence

• Published in: Applied Mathematical Modelling Vol. 103; pp. 221 - 237
• Main Authors: Zheng, Rui, Zhou, Yifan
• Format: Journal Article
• Language: English
• Published: New York Elsevier Inc 01.03.2022; Elsevier BV
• Subjects: Approximation; Component dependence; Condition monitoring; Dynamic programming; Dynamic replacement policy; Economic models; Fuel oils; Graphical representations; Inspection; Inventory management; Maintenance costs; Markov processes; Optimization; Production process; Proportional hazards model; Component dependence; Dynamic replacement policy; Dynamic programming; Production process; Proportional hazards model
• ISSN: 0307-904X; 1088-8691; 0307-904X
• DOI: 10.1016/j.apm.2021.10.028

•A dynamic inspection and replacement policy for two-unit production systems.•Integration of economic, stochastic, and structural dependencies into maintenance model.•SMDP formulation based on a matrix-based approximation method.•A policy-iteration based algorithm for solving the joint optimization problem.•The proposed policy outperforms the other widely used CBM policies. This paper proposes a dynamic inspection and replacement policy for a production system composed of two nonidentical units and an inventory system. Unit 1 deteriorates gradually and it is subject to soft failure. Unit 2 experiences hard failures that occur following a general lifetime distribution. The degradation processes of the two units are stochastically interdependent. Condition monitoring is performed to reveal the deterioration state of unit 1 when a certain number of production runs complete or when unit 2 fails, whichever comes first. At each inspection epoch, a decision is made on whether to replace unit 1 and/or unit 2. The objective is to jointly optimize the inspection interval and the replacement policy by minimizing the long-run average cost rate. The optimization problem is formulated in the semi-Markov decision process (SMDP) framework. Economic dependence and structural dependence are also integrated into the maintenance model. A matrix-based approximation method is employed to derive the SMDP quantities. The application of the proposed approach to a fuel oil supply system demonstrates the superiority of the proposed approach in saving operation and maintenance costs. The obtained graphical representation of the obtained policy is supportive during maintenance implementation.