Optimization of Dynamic Preventive Maintenance for Multicomponent Systems

In safety-critical Cyber Physical Systems (CPS), effective maintenance is of utmost importance for the operation of equipment. Periodic maintenance can prevent or delay the occurrence of fault to a certain extent, but it inevitably leads to either “over-maintenance” or “non-timely maintenance”, which results in a waste of resources or safety risks in the system. For the purpose of balancing maintenance cost and reliability, this study proposes an optimization model for the dynamic maintenance of multi-component systems. In this model, we consider synthetically the influence of maintenance time and cost, the effect of failure rate of each component on overall maintenance cost, and the effect of incomplete maintenance on component failure rate; we take the overall system maintenance cost as objective function. This model calculates the optimal maintenance times and intervals in the life span of the whole system, under the constraint of minimum reliability requirements. Our simulation results in the context of metro equipment maintenance show that the proposed model is effective and efficient in optimizing maintenance strategies.