Maintenance policies for deteriorating systems subject to non-self-announcing failures

In stochastically failing systems, maintenance activities are crucial to improve system availability performance, particularly, when the failures are non-self-announcing in which failures can only be detected through inspections. In this study, we comparatively evaluate various maintenance policies for systems subject to continuous time Markovian deterioration which may result in non-self-announcing failures. For each policy, the decision maker inspects the system periodically at the decision epochs, identifies the current state; good, poor, or failed and chooses an available action; do-nothing, repair, or replace. By the help of renewal theory, the objective is to minimize the expected long-run cost rate which consists of various components such as inspection, downtime, and maintenance costs. We provide a numerical example to analyse the effect of various cost parameters on the optimum inspection period and policy.