Reliability modeling of repairable system based on a stochastic model

Reliability modeling is an important part of reliability research, which has important guiding significance for evaluating system reliability index and optimizing preventive maintenance strategy. Maintenance behaviors, such as maintenance and replacement of system components, etc. have great influence on the reliability of repairable system. But the commonly used reliability distribution models ignore the repair history, which is not in conformity with the actual engineering application. In this paper, a reliability model based on failure time and maintenance effect was established by using log-linear baseline intensity function. The proposed model overcome the lack of Weibull distribution and Weibull process that only consider the change trend of failure time and ignore the importance of repair history. Then, the proposed model was a three parameter model, and the model parameter estimation was studied by using maximum likelihood method. Finally, the reliability model was applied in the reliability analysis of a repairable Numerical Control (NC) system. The validity of the proposed model and its parameter estimation method were verified. Compared with the Weibull distribution and Weibull process, the proposed model has obvious superiority, and fits the cumulative number of failure curve very well.