Optimal design of a condition-based maintenance model

Abstract

System performance and associated gain can be improved by using efficient maintenance policies. The latest developments in maintenance models include condition-based maintenance strategies. This paper presents a condition-based maintenance model for a deteriorating system. Deterioration can be defined as a process where the important parameters of a system gradually worsen, and if left unattended, the process leads to deterioration failure. In this paper, we consider a discrete stage deterioration, where the first stage is a good stage and the last stage is the failed stage. Failure of the system can be identified immediately, and the system is restored through corrective maintenance. The system is subjected to periodic inspection that identifies the condition of deterioration. After an inspection, based on the degree of deterioration (system condition), a preventive maintenance is performed or no action is taken. Both corrective and preventive maintenance bring the system to an ''as good as new" stage. Using Markov chains, this paper presents closed-form analytical solutions for the performance measures of the model. This paper also presents algorithms to find the optimal model parameters that maximize the system availability.