This study presents a hybrid genetic algorithm (GA) to optimize the periodic preventive maintenance model in a series-parallel system. The intrinsic properties of a repairable system, including the structure of reliability block diagrams, maintenance priority of components, and their maintenance periods, are considered in developing the proposed hybrid GA. The importance measure of components is employed to account for these properties, identify important components, and determine their maintenance priorities. The optimal maintenance periods of these important components are then determined to minimize total maintenance cost given the allowable worst reliability of a repairable system using the GA search mechanism. An elitist conservation strategy is applied to retain superior chromosomes in the iterative breeding process to accelerate the approach toward the global optimum. Furthermore, the response surface methodology is utilized to systematically determine crossover probability and mutation probability in the GA instead of using the conventional trial-and-error process. A case study demonstrates the effectiveness and practicality of the proposed hybrid GA in optimizing the periodic preventive maintenance model in a series-parallel system.

No abstract

No abstract

A new method to optimise the non-periodic preventive maintenance model of a series-parallel system is proposed. A two-stage algorithm that incorporates the failure limit policy to determine maintenance components, maintenance times, and total maintenance cost is suggested. When the reliability of the system reaches a threshold value, preventive maintenance is performed. The first stage identifies the parallel subsystem required to be maintained, while the second stage determines the component required to be maintained in the parallel sub-system. A unit-cost life index (UCL) has been developed to evaluate the extent to which maintaining a component extends the life of a system for the parallel subsystem. Three simulated cases demonstrate the effectiveness and the practicality of the proposed method in optimising the non-periodic preventive maintenance model of a series-parallel system.

No abstract