Prediction of system reliability for multiple component repairs

Abstract

Optimal asset management in industries requires accurate reliability prediction of complex repairable systems. A Split System Approach (SSA) has previously been presented for predicting the reliability of complex systems with multiple Preventive Maintenance (PM) cycles over a long term horizon. However, the algorithms in that model were derived with an assumption that the same single component is always repaired in all PM actions. This paper extends the model to a scenario where a different single component is repaired each time. This extended model can be used to determine the remaining life of the system and to describe the changes in reliability with PM actions for this scenario. As a result, it can be used to support asset PM decision making over the operation and maintenance phase of the asset. Assets often have a number of vulnerable components, i.e., the lives of these components are much shorter than the lives of the rest of the system. An optimal time of sequential PM actions of these critical components can maximise the useful life of the asset effectively. The model developed in this paper can be used to determine this optimal PM strategy.