Selective Maintenance Modeling for a Multi-State System Considering Human Reliability

In an actual industrial environment, many systems need to perform multiple missions in succession. In order to ensure the success of next mission, the components require maintenance in maintenance breaks. In this case, there may not be enough time and cost to repair every component. Therefore, selective maintenance is required between two consecutive missions. In a multi-state system, each component also has multiple working states, which makes many maintenance options during maintenance break possible. But different maintenance workers will affect the states of components after maintenance. The purpose of this paper is to calculate the human error rate for maintenance of multi-state systems and give the reliability of performing next mission. For the selective maintenance of multi-state systems considering human reliability, with constraints of maintenance time and cost, an optimization model with the goal of maximum system reliability is presented, and the optimal selective maintenance strategy is given by genetic algorithm (GA). Finally, an example is presented to illustrate the effect of human reliability on the final maintenance strategy.