A new Bayesian reliability assessment scheme based on D-S evidence theory

The sizes of experiment samples are usually small in the reliability evaluation of comprehensive new products. In this case, Bayesian statistical method is generally more suitable than the classical methods and has thus been widely used in the process of reliability assessment. In order to express and apply the uncertain prior information better, we present a new Bayesian reliability assessment scheme based on the D-S evidence theory. The estimation for reliability index of exponential distributions is studied for the non-replacement fixed failure number test samples. At the same time, we propose two reliability assessment methods. One is a comprehensive evaluation by integrating the posterior distributions of reliability indices. The other is one by integrating the posterior confidence intervals. Numerical examples show that the proposed methods are prudent and reasonable.