Reliability sampling plans for a lognormal distribution under progressive first-failure censoring with cost constraint

摘要

In this paper, we establish reliability sampling plans for a two-parameter lognormal distribution when it is known that samples are progressive first-failure censored. An EM algorithm is developed to obtain maximum likelihood estimates of unknown parameters. Reliability sampling plans are obtained using two different criteria, namely minimizing the expected test time and minimizing the determinant of the variance covariance matrix of the maximum likelihood estimates, under a restriction on the total cost associated with the experiment and for given specifications on the operating characteristic curve. Finally, a numerical study is performed and specific comments are given.