期刊論文
| 學年 | 110 |
|---|---|
| 學期 | 2 |
| 出版（發表）日期 | 2022-03-01 |
| 作品名稱 | Analysis for constant-stress model on multicomponent system from generalized inverted exponential distribution with stress dependent parameters |
| 作品名稱（其他語言） | |
| 著者 | Wang, L.; Wu, S.-J.; Zhang, C.; Dey, S.; Tripathi, Y. M |
| 單位 | |
| 出版者 | |
| 著錄名稱、卷期、頁數 | Mathematics and Computers in Simulation 193, p.301-316 |
| 摘要 | In this paper, inference of multicomponent system is presented under constant-stress accelerated life test. When the lifetime of the components of the multicomponent system follows a generalized inverted exponential distribution (GIED), different from standard extrapolation approach where only the scale parameter depends on the stress conditions, a life-stress model is proposed assuming that both parameters of the GIED are nonconstant and depend on the stress. The model parameters are estimated along with the existence and uniqueness via maximum likelihood method, and the survival function of the multicomponent system is extrapolated at normal use condition. The approximate confidence intervals are further constructed using the asymptotic distribution theory and delta technique. Furthermore, another alternative generalized estimates are also constructed by using proposed pivotal quantities for comparison. In addition, likelihood ratio testing is presented as a complementary by comparing the life-stress models with nonconstant and constant parameters. Finally, simulation studies and a real data example are carried out for illustrations, and the results indicates that the proposed generalized approach is superior to conventional likelihood estimation. |
| 關鍵字 | Accelerated life test;Multicomponent system;Generalized inverted exponential distribution;Maximum likelihood estimation;Generalized estimation |
| 語言 | en_US |
| ISSN | 1872-7166 |
| 期刊性質 | 國外 |
| 收錄於 | SCI Scopus |
| 產學合作 | |
| 通訊作者 | Liang Wang |
| 審稿制度 | 是 |
| 國別 | NLD |
| 公開徵稿 | |
| 出版型式 | ,電子版,紙本 |
| 相關連結 |
機構典藏連結 ( http://tkuir.lib.tku.edu.tw:8080/dspace/handle/987654321/121590 ) |