Multi-component maintenance grouping optimization based on stochastic dependency

Abstract

Maintenance activities often require an identical preparatory work. Therefore, a joint execution of such maintenance activities may save a substantial cost. In this work, we consider the problem of optimizing the total maintenance cost of a multi-component repairable system by grouping of components and carrying out maintenance activities on group(s) of components of a complex system. More specifically, we propose a maintenance grouping cost optimization model based on the stochastic dependency as well as economic dependency among components in a system. The stochastic dependency modeling is done using Bayesian network by considering the failure probability of components as a measure of failure interactions among components. Penalty functions are formulated due to the shift of individual optimal maintenance time of components to find the optimum joint maintenance interval and associated cost benefits. Finally, a case study on a diesel engine of a diesel power plant involving 10 components (components of diesel engine, air intake system, and turbocharger) is presented to illustrate the proposed approach.