Optimal condition-based preventive maintenance policy for balanced systems

• Published in: Reliability engineering & system safety Vol. 211; p. 107606
• Main Authors: Wang, Jingjing, Qiu, Qingan, Wang, Huanhuan, Lin, Cong
• Format: Journal Article
• Language: English
• Published: Barking Elsevier Ltd 01.07.2021; Elsevier BV
• Subjects: Balanced system; Condition monitoring technique; Deterioration; Discrete time systems; Maintenance costs; Markov processes; Optimization; Preventive maintenance; Reliability analysis; Reliability aspects; Reliability engineering; semi-Markov decision process; System reliability; Systems analysis; Balanced system; semi-Markov decision process; Preventive maintenance; Condition monitoring technique
• ISSN: 0951-8320; 1879-0836
• DOI: 10.1016/j.ress.2021.107606

•An optimal preventive maintenance policy is proposed for balanced systems.•The system state is evaluated based on the deterioration levels of all components.•Semi-Markov decision process is utilized to solve the optimal optimization problem. Reliability evaluation of balanced systems has been extensively studied in recent years. However, the research on maintenance optimization for balanced systems is still underexplored. This paper makes a novel contribution to existing literature by proposing a condition-based preventive maintenance (PM) policy for balanced systems with identical components. Each component is subject to continuous degradation and the states of all components are simultaneously detected at discrete time epochs. The system state is evaluated based on the deterioration levels of all components. The system is out of balance once the deterioration level of any component reaches a critical value or the difference between the deterioration levels of two components on the symmetric positions exceeds a specific threshold, whichever occurs first. In this research, PM activities are employed to avoid competing failures. The optimal PM thresholds are determined by minimizing the system maintenance cost within the framework of semi-Markov decision process. An illustrative example is provided to demonstrate the application of the proposed PM policy for balanced systems.