Reliability Evaluation for the Exponential Load-Sharing System Based on System Accelerated Life Testing Data

In this article, the reliability assessment methods for the exponential load-sharing system is proposed based on the system accelerated life testing data under the equal load-sharing rule. The exponential distribution is used to model the baseline distributions of the component lifetimes in load-sharing systems. The relationship between the load levels and the component lifetimes is assumed to be log-linear. Accurate confidence intervals for the model parameters, system reliability and mean lifetime at the initial stress level are proposed. The prediction intervals are also derived for the future failure time and the remaining useful life of the system at the initial stress level. The performance of the proposed procedures is evaluated by using Monte Carlo simulation. The simulation results show that the proposed procedures outperform the large sample normal approximation method in terms of the coverage probability and average interval length, in particular for small sample sizes. Finally, a numerical example is used to illustrate the developed procedures.