Abstract
Generalized gamma distribution includes many useful lifetime distributions for analyzing lifetime data in reliability and survival studies. In the context of one-shot device testing, it is dicult to collect sucient lifetime information on the one-shot devices, due to the destructive nature of one-shot devices that either left- or right-censored data are collected. In modern life-tests, test devices are subjected to conditions in excess of its normal operation condition in order to induce more failures within a relatively short period of time. Such life-tests are called accelerated life-tests and commonly used for collecting lifetime data. In this paper, we discuss the analysis of one-shot device testing data under accelerated life-tests based on generalized gamma distributions. Both maximum likelihood and least-squares approaches are developed to find the estimates of the model parameters. Furthermore, the estimation on the reliability at a specic mission time as well as on the mean lifetime of the devices are also developed. Both approaches are then compared through comprehensive simulation studies. The results show that both approaches are quite satisfactory in terms of biases, root mean square errors, and numbers of cases of convergence. In general, the maximum likelihood approach is comparably stable to nd the estimates. Copyright © 2018 Ling.
Original language | English |
---|---|
Article number | 124 |
Journal | International Journal of Applied & Experimental Mathematics |
Volume | 3 |
Issue number | 1 |
Publication status | Published - 2018 |
Citation
Ling, M. H. (2018). A comparison of estimation methods for generalized gamma distribution with one-shot device testing data. International Journal of Applied & Experimental Mathematics, 3. Retrieved from https://www.graphyonline.com/journal/journal_article_inpress.php?journalid=IJAEMKeywords
- Accelerated life-tests
- One-shot devices
- Generalized gamma distribution
- Fisher scoring
- Least-squares