Availability-oriented dynamic group maintenance scheduling under iteratively updated information

Availability serves a core indicator of asset health management, as it immediately revealing the capacity of preserving functional. This study investigates an intelligent availability optimization approach aimed at group maintenance of multi-component systems. In contrast to most existing statistic group maintenance models within pre-set horizons, this paper introduces a globally dynamic grouping framework, aimed at improving operational availability and decision-making robustness. The proposed framework retains the simplicity of conventional rolling horizon approaches in spare planning while offering fast response to real-time health evolution. By leveraging system status information, it facilitates unified cluster maintenance planning by considering the impact of shared execution time on the overall maintenance schedule. Numerical experiments demonstrate that, compared with conventional static approaches, the proposed dynamic group framework exhibits significantly superior performance in enhancing system availability and reducing idle time.