A hybrid preventive maintenance model for systems with partially observable degradation. (6th February 2020)

• Record Type: Journal Article
• Authors: Finkelstein, Maxim; Cha, Ji Hwan; Levitin, Gregory
• Is Part Of: IMA journal of management mathematics. Volume 31:Number 3(2020)
• Page Start: 345
• Page End: 365
• Publication Date: 2020-02-06
• Subjects: maintenance -- shot-noise process -- Poisson process -- age replacement -- failure (hazard) rate process; Management -- Mathematical models -- Periodicals; Management science -- Mathematical models -- Periodicals; Business mathematics -- Periodicals; 650.01513
• DOI: 10.1093/imaman/dpz018
• Languages: English
• ISSNs: 1471-678X

Abstract

Abstract: A new model of hybrid preventive maintenance of systems with partially observable degradation is developed. This model combines condition-based maintenance with age replacement maintenance in the proposed, specific way. A system, subject to a shock process, is replaced on failure or at some time ${T}_S$ if the number of shocks experienced by this time is greater than or equal to m or at time $T>{T}_S$ otherwise, whichever occurs first. Each shock increases the failure rate of the system at the random time of its occurrence, thus forming a corresponding shot-noise process. The real deterioration of the system is partially observed via observation of the shock process at time ${T}_S$ . The corresponding optimization problem is solved and a detailed numerical example demonstrates that the long-run cost rate for the proposed optimal hybrid strategy is smaller than that for the standard optimal age replacement policy.