Goodness-of-fit tests for system components with random redundancy level

Component redundancy is a well-known tool for improving the reliability of a system. In this article, given the probability model for the level of redundancy, N, and the repeated observations on the life time of a parallel system with N i.i.d. components, each of which follows the life cumulative distribution function (c.d.f.) F, a class of goodness-of-fit tests for testing the null hypothesis that F is F (a specified c.d.f.) based on the spacings is studied.