摘要

Estimating component reliability frequently requires the use of system life time data. Due to time and cost constraints, the true cause of system failure may be masked. This article discusses how to identify the true cause of system failure from masked data through an approximate grey prediction approach for small sample sizes. Two numerical examples are given to illustrate the application of our proposed approach. The results of parametric 1 estimations and reliability analysis have shown that our proposed method adequately identifies the failed components and results in improved estimates of component lifetimes.