Start-Up Demonstration Tests Under Markov Dependence

Abstract

A start-up demonstration test is a mechanism that can be used to demonstrate the reliability of an equipment to the customer. In this paper, we study CSTF (consecutive successes total failures) and TSTF (total successes total failures) procedures assuming that the individual start-ups follow a first order Markov dependence structure. Explicit (nonrecursive) expressions for the distributions of the test lengths are provided. Maximum likelihood and moments estimators of the expected test length are obtained and some numerical results are presented for an illustration.