An imperfect age-based and condition-based opportunistic maintenance model for a two-unit series system

• Published in: Computers & industrial engineering Vol. 160; p. 107583
• Main Authors: Wang, Jingjing, Miao, Yonghao, Yi, Yinggang, Huang, Dagang
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.10.2021
• Subjects: Condition monitor; Economic dependence; Imperfect maintenance; Random repair time; The semi-Markov decision process; Economic dependence; Condition monitor; Random repair time; Imperfect maintenance; The semi-Markov decision process
• DOI: 10.1016/j.cie.2021.107583
• ISSN: 0360-8352

•An imperfect opportunistic maintenance policy for a two-unit system is proposed.•Deterioration levels and usage of units are utilized to repair the whole system.•Semi-Markov decision process is used to solve the optimal optimization problem.•Maintenance cost is minimized by optimizing the maintenance thresholds of units. Maintenance strategies have been studied extensively and applied successfully to reduce the operation costs of industrial assets. However, the study for heterogeneous two-unit systems is rarely based on the usage and condition information of units, simultaneously. This paper originally proposes a novel imperfect opportunistic maintenance model for a two-unit series system considering random repair time and two types of failures, where unit 1 and unit 2 are respectively subject to soft failure and hard failure. The system maintenance actions are performed not only based on the deterioration level of unit 1 but also on the usage of unit 2. To utilize the economic dependence between the two units, the opportunistic maintenance policy is adopted in the series system. When a unit fails or is being repaired, the other unit has an opportunity to be simultaneously maintained once it meets the requirement of opportunistic maintenance. Herein, opportunistic maintenance action is imperfect maintenance, which means that the deterioration level of unit 1 and the usage of unit 2 restore to a lower level than the as-good-as new status after repair. Nonetheless, once both units meet the requirement of replacement actions, they will be perfectly replaced by new ones. The objective is to minimize the system average maintenance cost by optimizing the thresholds of opportunistic maintenance and perfect replacement of two units. Finally, an illustrative example is presented to demonstrate the superiority of the proposed maintenance model.