Abstract
Engineering systems have become increasingly complex, which demands the study of system-level maintenance strategy to be integrated with a realistic prediction of component-level performance. Most previous research of system maintenance models has focused only on the relationship between component failure rate and maintenance cost. For a system with degrading components, however, it is more intuitive to model the system’s health state together with component degradation processes. In this paper we propose a preventive maintenance strategy based on the simulation of component degradation processes and lifetime predictions. In addition, the proposed strategy utilizes the maintenance opportunities between tasks assigned by system. Specifically, the unit-time system maintenance cost is minimized with the consideration of interactions between component-level and system-level decision-makings. The approach we propose is more practical due to the simulation of the component degradation processes according to the real task profile of the system and the initial maintenance schedule can be re-adjusted in the event of an unexpected component failure. Therefore, it ensures an overall low maintenance cost and high system reliability over time. A case study of unmanned aerial vehicle illustrates the proposed method.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Additional information
Funding
Notes on contributors
Yaning Sun
Yaning Sun received the master degree at School of Reliability and Systems Engineering, Beihang University, Beijing, China. Her current research interests include life prediction and UAS.
Xiaohong Wang
Xiaohong Wang received the Ph.D. degree at Beihang University, Beijing, China. She is an associate professor with the School of Reliability and Systems Engineering, Beihang University, Beijing, China. Her main research interests include reliability and environment testing, accelerated testing and life prediction.
Lizhi Wang
Lizhi Wang received the Ph.D. degree at Beihang University, Beijing, China. He is an associate professor with the Unmanned System Institute, Beihang University, Beijing, China. His research interests include accelerated degradation testing, Bayesian evaluation, lifetime prediction, systems engineering, UAS.
Rong Pan
Rong Pan is an associate professor of Industrial Engineering in the School of Computing and Augmented Intelligence at Arizona State University. He received his doctoral degree in Industrial Engineering from Penn State University in 2002. His research interests include failure time data analysis, system reliability, design of experiments, multivariate statistical process control, timeseries analysis, and computational Bayesian methods.