Condition-based maintenance for long-life assets with exposure to operational and environmental risks

• Published in: International journal of production economics Vol. 221; p. 107482
• Main Authors: Liang, Zhenglin, Liu, Bin, Xie, Min, Parlikad, Ajith Kumar
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.03.2020
• Subjects: Hypo-exponential distribution; Maintenance; Moment matching; Operational and environmental risk; Stochastic modeling; Operational and environmental risk; Maintenance; Stochastic modeling; Moment matching; Hypo-exponential distribution
• ISSN: 0925-5273; 1873-7579
• DOI: 10.1016/j.ijpe.2019.09.003

This paper presents a new condition-based maintenance (CBM) model for long-life assets to address the potential risk caused by the decline of the operating environment. Two types of maintenance are formulated in the CBM model. Minor maintenance can mitigate the operational and environmental risk, and major maintenance can eliminate the accumulated damage within the asset. A continuous-time semi-Markov chain (CTSMC) is used for modeling the aging of the asset as well as the stochastic decline of the operating environment. To optimize the CBM policy in a mathematically tractable manner, we introduce a hypo-exponential approximation approach to match the first four moments of the sojourn time distribution of CTSMC. This approach guarantees a minimum representation of the CTSMC with non-fictitious surrogated Markov chain. The model provides both good mathematical tractability and sufficient generalizability. The practical impact of this research is demonstrated by applying it to a real industrial case of concrete bridge maintenance. It is observed that this approach results in a CBM plan with a lower asset lifecycle cost compared to current techniques.