Multi-state Deteriorating System Dependability with Maintenance Using Monte Carlo Simulation

Abstract

In this paper a multi-state deteriorating system in which condition based preventive maintenance is performed in order to improve its availability and performance, is modeled and studied extensively. The system experiences various states of deterioration before a total failure occurs. At each deterioration state an inspection is carried out at constant intervals in order to check if the system should undergo any preventive maintenance, either minimal or major. Different predefined levels of deterioration are settled which indicate the states where minimal and major maintenance start to take place. When, despite maintenance the system fails, a repair procedure is triggered and after its completion, the system returns to its initial perfect state. The proposed model incorporates also imperfect maintenance, either minimal or major, failed maintenance and sudden failures that may occur mostly due to external factors at any deterioration state. A model with general sojourn time distribution is proposed and thus a semi-Markov process is used to model system's evolution in time. The transient availability, the reliability and the operational cost at any given time instant t are estimated using Monte Carlo simulation instead of analytical computation which in most cases is very difficult or even infeasible.