Maintenance cost evaluation for heterogeneous complex systems under continuous monitoring. (August 2020)

• Record Type: Journal Article
• Authors: Castro, Inma T.; Basten, Rob J.I.; van Houtum, Geert-Jan
• Is Part Of: Reliability engineering & system safety. Volume 200(2020)
• Publication Date: 2020-08
• Subjects: Condition-based maintenance -- Opportunistic maintenance -- Complex systems -- Semi-regenerative process; Reliability (Engineering) -- Periodicals; System safety -- Periodicals; Industrial safety -- Periodicals; Fiabilité -- Périodiques; Sécurité des systèmes -- Périodiques; Sécurité du travail -- Périodiques; 620.00452
• DOI: 10.1016/j.ress.2019.106745
• Languages: English
• ISSNs: 0951-8320

Abstract

Highlights: Probabilistic modelling of the maintenance for heterogeneous systems. Evaluation of the expected cost rate for heterogeneous systems. Comparison of two techniques of evaluation: Renewal techniques and semi-regenerative techniques. Abstract: The maintenance strategy for a complex system consisting of both non-monitored and monitored components is analyzed in this paper. Non-monitored components can only be maintained correctively upon failure. Monitored components are monitored continuously and are maintained when they become too degraded, i.e., when its degradation level hits a threshold. For this complex system, an opportunistic maintenance strategy is implemented, meaning that a maintenance intervention for a component can be used as an opportunity for preventive maintenance of monitored components: If the degradation level of a monitored component exceeds a preventive threshold at the time of another maintenance intervention, this component is maintained preventively. By performing these maintenance actions, different costs are incurred. The main purpose of this paper is to evaluate the expected cost rate of the system. To that end, two methods are compared: renewal and semi-regenerative techniques. Using renewal techniques, the evaluation of the expected cost rate of this maintenance strategy is time consuming, especially for a large number of monitored components. However, using semi-regenerative techniques the required computation time is drastically shortened: For a system with ten monitored components, the computation time to evaluate the optimal maintenance strategy goes from more than a day to few seconds. Numerical examples are given to illustrate the results. The conclusion is that for a large number of monitored components, semi-regenerative techniques are more appropriate to evaluate the expected cost rate in terms of computation time.