Maintenance analysis and optimization via statistical model checking: Evaluating a train pneumatic compressor

TitleMaintenance analysis and optimization via statistical model checking: Evaluating a train pneumatic compressor
Publication TypeConference Paper
Year of Publication2016
AuthorsRuijters E.JJ, Guck D., Drolenga P., Peters M., Stoelinga M.IA
Conference NameProceedings of the 13th International Conference on Quantitative Evaluation of SysTems, QEST 2016, Québec City, Canada
Date PublishedAugust
PublisherSpringer Verlag
Conference LocationBerlin
KeywordsFault Maintenance Trees, Fault Tree Analysis, Reliability analysis

Maintenance is crucial to ensuring and improving system dependability: By performing timely inspections, repairs, and renewals the lifespan and reliability of systems can be significantly improved. Good maintenance planning, however, has to balance these improvements against the downsides of maintenance, such as costs and planned downtime. In this paper, we study the effect of different maintenance strategies on a pneumatic compressor used in trains. This compressor is critical to the operation of the train, and a failure can lead to a lengthy and expensive disruption. Within the rolling stock maintenance company NedTrain, we have modelled this compressor as a fault maintenance tree (FMT), i.e. a fault tree augmented with maintenance aspects. We show how this FMT naturally models complex maintenance plans including condition-based maintenance with regular inspections. The FMT is analysed using statistical model checking, which allows us to obtain several key performance indicators such as the system reliability, number of failures, and required unscheduled maintenance. Our analysis demonstrates that FMTs can be used to model the compressor, a practical system used in industry, including its maintenance policy. We validate this model against experiences in the field, compute the importance of performing minor services at a reasonable frequency, and find that the currently scheduled overhaul may not be cost-effective.