Nonparametric Estimation for accelerated life testing under imperfect repair.

This paper considers nonparametric estimation of lifetime distribution of a system subject to imperfect repair, based on data from constant stress accelerated life tests. By assuming as time transformation function relating stress to lifetime, a version of the inverse power law, a method of estimating the lifetime distribution at use condition stress has been recently proposed by Diana and Giordan. This method, based on the Brown-Proschan imperfect repair model, is nonparametric in that it does not make any assumptions about the underlying distribution of life length. Some simulations to understand if accelerated life tests can be used instead of normal tests evaluate the behaviour of the test procedure.