基于随机冲击的视情维修和预防维修决策  

On-condition Maintenance and Preventive Maintenance Decision Based on Random Shocks

作　　者：张斌[1] 费文龙[1] 杨洲木[1] ZHANG Bin;FEI Wen-long;YANG Zhou-mu(College of Mathematics and Statistics,Nanjing University of Information Science&Technology,Nanjing 210044,China)

机构地区：[1]南京信息工程大学数学与统计学院,江苏南京210044

出　　处：《数理统计与管理》2020年第4期644-653,共10页Journal of Applied Statistics and Management

摘　　要：研究了系统受随机冲击影响发生失控和故障等两种风险情况下的视情维修和预防维修决策问题。假设系统因随机冲击产生的退化量是随机变量,退化量累积达到阈值时系统将失控,构建了系统失控时间的分布函数。系统在失控状态下能够继续运行,但收益将减小,发生故障的风险将增大。为了提高系统运行的效益,对系统的状态进行中期检查,根据系统的运行状态决定是否维修。讨论了系统在一个周期内可能存在的各种状态,综合考虑系统运行产生的收益及各类维修成本,提出了单位时间平均利润的数学模型。最后,通过仿真模拟,分析了模型参数对单位时间平均利润及决策变量的影响。In this paper,we focus on the decisions of on-condition maintenance and prevent maintenance for system with out of control which caused by random shocks and failure.The degradation of the system due to shocks is a random variable.The system will get out of control when the accumulation of degradation quantity reaches the threshold.The distribution function of out of control time is proposed.The system is able to continue operating in a state of out of control,but the benefits will be smaller and the risk of failure will be greater.In order to improve the running efficiency of system,the state of system is checked.The maintenance decision is according to the running state of the system.The possible states of the system in a period are discussed.A mathematical model of average profit per unit of time is proposed,which is based on the income generated by the operation of the system and various maintenance costs.Finally,the effects of process parameters on average profit per unit of time and the decision variable are studied by simulation.

关 键 词：随机冲击 失控 故障 维修决策 

分 类 号：O212[理学—概率论与数理统计] O213.1[理学—数学]