基于相关性的系统预防性维修任务成组优化

Optimal groupingpreventive maintenance of a system under dependencies

针对传统维修相关性分类方法未全面考虑维修相关性影响,提出从时间相关性、故障相关性、结构相关性和功能相关性4个方面描述多部件系统的维修相关性,分析了各类相关性对预防性维修成组的影响.在此基础上,将系统预防性维修费用分为不可分摊部分与可分摊部分,以系统维修成本最小化为优化目标建立系统预防性维修任务成组优化模型,并利用生物地理学优化算法确定最佳的系统预防性成组维修方案.模型中将直接维修成组和间接维修成组有机结合起来,实现两种预防性成组策略的优势互补,从而有效地降低系统预防性维修费用.最后通过算例,将模型优化结果与直接维修成组方式的优化结果进行了比较,表明模型的合理性.

In view of incomplete consideration of maintenance correlation with traditional methods,this paper puts forward a new idea to judge maintenance correlation in a multi-unit system from four aspects such as time correlation,fault correlation,structural correlation and function correlation.Meanwhile,this paper presents an analysis of the impact caused by maintenance correlation on grouping preventive maintenance.On this basis,preventive maintenance costs can be divided into shared parts and non-shared parts.The optimization model of grouping preventive maintenance is provided to minimze the system maintenance cost during required available time.Applying biogeography-based optimization,the model can choose the best grouping preventive maintenance plan about correlating components in the system.This model combined direct group maintenance and indirect group maintenance in order to realize two complementary advantages of two preventive group strategies.Thus,it can reduce system preventive maintenance costs effectively.Finally,the optimizing model results and optimization results by direct maintenance group way are compared via a case study,so as to indicate the model is reasonable.