Reliability and condition-based maintenance modeling for systems operating under performance-based contracting

• Published in: Computers & industrial engineering Vol. 142; p. 106344
• Main Authors: Wang, Yukun, Liu, Yiliu, Chen, Junyan, Li, Xiaopeng
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.04.2020
• Subjects: Condition-based maintenance; Degradation; Inverse Gaussian process; Performance-based contracting; Shock; Degradation; Shock; Condition-based maintenance; Inverse Gaussian process; Performance-based contracting
• ISSN: 0360-8352; 1879-0550
• DOI: 10.1016/j.cie.2020.106344

•Two-stage degradation-based and shock-based failure processes are established.•System state transition and unit-specific heterogeneity are incorporated.•Reliability and CBM models are formulated.•Optimal maintenance policy is obtained under performance-based contracting.•Applicability of the models is demonstrated through a numerical study. Performance-based contracting motivates service providers to implement effective maintenance policies so as to boost profits and improve system performance at a lower cost. This paper deals with reliability and condition-based maintenance modeling for single-unit systems operating under performance-based contracting. For a system subject to degradation and sudden shocks, there involves three states: normal, degraded and failed. Once entered into the degraded state, the system deteriorates faster and becomes more susceptible to shocks. The degradation conforms to a two-stage inverse Gaussian process with random effects characterizing unit-specific heterogeneity in the population. Furthermore, the arrival of sudden shocks follows a doubly stochastic Poisson process. A reliability model is developed based on degradation-based and shock-based failures modeling. Afterwards, the long-run maintenance cost rate and the average system availability are evaluated. Optimal inspection-based preventive replacement policy is obtained by maximizing the expected profit rate to the service provider. Finally, a numerical example along with sensitivity analysis of model parameters is presented to demonstrate the applicability and solution procedure of the proposed models.