Research of Reliability Evaluation Method Based on Optimal Parameter Estimation  
Author Jianjun Ma
Co-Author(s) Songlin Zheng; Tie Chen; Jinzhi Feng; Lihui Zhao
Abstract In reliability evaluation, the obtained estimation is not unbiased estimation based on the traditional parameter estimation methods. Sometimes even if the estimation is unbiased but it is not the most effective, and the interval estimation of parameter is not the shortest. In this paper, we study the case of the first failure mileage for the vehicle engine. The parameter estimation methods of uniformly minimum variance unbiased estimate (UMVUE) and shortest interval estimation are proposed for reliability evaluation. The distribution of the first failure mileage is concluded, and the reliability function, confidence interval and mean residual life (MRL) function are obtained. Comparing with the traditional parameter estimation methods, the proposed methods can provide more accurate results. Thus, the proposed methods are the optimal parameter estimation method. The results provide theoretical supports for reliability design of the engine.
Keywords UMVUE, shortest confidence interval, reliability evaluation, first failure mileage
    Article #:  23-195
Proceedings of the 23rd ISSAT International Conference on Reliability and Quality in Design
August 3-5, 2017 - Chicago, Illinois, U.S.A.