Ranking and selection procedures for gamma populations with type-Ⅱ censored data

Abstract

Ranking and selection procedures for gamma populations are developed with respect to the largest scale paramenter under Type-II censoring. The shape parameters of gamma distributions are assumed to be known and equal for all the populations. Based upon common sample size and censoring number, four procedures are proposed based on the TTOT(Total Time On Test) statistic. To derive expressions for the probability of correct selection and the infimum of this probability for each procedure, a two-moment chi-square approximation is used since the exact distribution of the TTOT statistic is not known exactly.