Preventive Maintenance Optimization of Aperiodic Multiple Component-Reassignments

This article proposes a new aperiodic preventive maintenance policy, which involves multiple reassignments of components in a system. Considering a multicomponent system, the components deteriorate according to heterogeneous stochastic processes because the components undertake different workloads and environmental stresses. Component reassignment (CR) is an action that reassigns the components to positions during system operation in order to improve overall system performance. The assignments of multiple CRs are mutually dependent decisions, and the execution times to conduct these CRs are also decision variables. All these decisions synthetically impact the system maintenance cost. The reliability functions and failure rates of components under the multiple CRs as well as the side effect of CR action on reducing the component reliability are formulated, using statistical virtual ages that establish links between decisions of consecutive CRs. To optimize the multi-CR based maintenance policy, a binary mixed integer nonlinear programming model is established with the objective of minimizing expected annual system maintenance cost that arises from the CRs, system replacement, and minimal repairs for emergency component failures. The optimal number of CRs is analytically shown to be finite and can be obtained by solving a series of optimization models. This article proposes two matheuristic approaches, an integrative construction approach and a sequential construction approach, to solve the models. Numerical experiments show the application of the proposed model and solution approaches in maintenance policy scheduling.

Automated summary

This article proposes a new aperiodic preventive maintenance policy based on component reassignment in a multicomponent system. A binary mixed integer nonlinear programming model is established to minimize the expected annual system maintenance cost. Two matheuristic approaches, an integrative construction approach and a sequential construction approach, are proposed to solve the model. Numerical experiments demonstrate the effectiveness of the proposed model and solution approaches in maintenance policy scheduling.