The Two Phase of the Bayes Statistical Model of Accelerated Test Censoring

DOI

10.2991/icence-16.2016.9How to use a DOI?

Keywords

Bayesian method; life prediction; distribution model

Abstract

In the Bayes method, the unknown parameters are given proper prior distribution, and the data are obtained. According to the Bayes theorem, the posterior distribution of the parameters is updated according to the data. In this paper we use Bayesian analysis and get the help of a priori information so that we can avoid the problem which the life can't be estimated caused by a small amount of failure data with the traditional methods. Limited hybrid conjugate prior distribution can represent phase stress prior distribution. We use this concept to make n test product units be put under the second phase stress (low and high) in turns to test. Then the test won't be stopped until we get a previously scheduled limited time T and proportion . Our aim is to minimize the asymptotic variance of product life's maximum likelihood estimate in hope of being used normally so as to expect to get the best method. What's more, we can find the best change time and proportion among stresses. That is to say, we can get the best life test method limited in Bayesian two phase typeI.

Copyright

© 2016, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Siquan Wang
PY  - 2016/09
DA  - 2016/09
TI  - The Two Phase of the Bayes Statistical Model of Accelerated Test Censoring
BT  - Proceedings of the 2nd International Conference on Electronics, Network and Computer Engineering (ICENCE 2016)
PB  - Atlantis Press
SP  - 43
EP  - 50
SN  - 2352-538X
UR  - https://doi.org/10.2991/icence-16.2016.9
DO  - 10.2991/icence-16.2016.9
ID  - Wang2016/09
ER  -