Condition-based maintenance strategies for a partially observable deteriorating system

This paper discusses the problem of the optimization of condition based maintenance policy for a stochastic deteriorating system in presence of covariates. The deterioration is modeled by a monotone stochastic process. The process of covariates is assumed to be a finite-state Markov chain. A model similar to the proportional hazards model is used to represent the influence of the covariates. In the framework of a non-monotone system, we derive the optimal maintenance threshold and optimal inspection period to minimize the expected average maintenance cost. Comparison of the expected average cost under different conditions of covariates and different maintenance policies is given by numerical results of Monte Carlo simulation.