出版日期：2019-09-01 00:00:00

著者：Wu Shuo-Jye; Huang Syuan-Rong

著錄名稱、卷期、頁數：Journal of Statistical Computation and Simulation 89(17), p.3241-3257

摘要：In this paper, we investigate some inference and design problems related to multiple constant-stress accelerated life test with progressive type-I interval censoring. A Weibull lifetime distribution at each stress-level combination is considered. The scale parameter of Weibull distribution is assumed to be a log-linear function of stresses. We obtain the estimates of the unknown parameters through the method of maximum likelihood, and also derive the Fisher's information matrix. The optimal number of test units, number of inspections, and length of the inspection interval are determined under D-optimality, T-optimality, and E-optimality criteria with cost constraint. An algorithm based on nonlinear mixed-integer programming is proposed to the optimal solution. The sensitivity of the optimal solution to changes in the values of the different parameters is studied.

關鍵字：D-optimality;E-optimality;maximum likelihood method;multiple constant stress;T-optimality;Weibull distribution

語言：en_US

期刊性質：國外

收錄於：SCI

通訊作者：Wu Shuo-Jye

審稿制度：是

國別：USA

出版型式：電子版,紙本