Analysis of a simple step-stress life test with a random stress-change time

Chengjie Xiong, Kejun Zhu, Ming Ji

Research output: Contribution to journalArticlepeer-review

15 Scopus citations

Abstract

This paper studies a variation of a simple step-stress life testing in which the stress change time is random, and the test is subject to type II censoring. We assume that only two order statistics from the test are observed. The first observed order statistic is the stress change time from a low level stress to a high level stress during the testing, and the second observed order statistic is the final failure time when the test is censored. We first present the joint probability distribution of the two order statistics observed from the simple step-stress accelerated life test. Maximum likelihood estimates, and the method of moment estimates for model parameters based on the joint distribution are considered. We also present the exact confidence interval estimates for the model parameters based on various pivotal quantities, and demonstrate the estimation procedure by a simulated example.

Original languageEnglish
Pages (from-to)67-74
Number of pages8
JournalIEEE Transactions on Reliability
Volume55
Issue number1
DOIs
StatePublished - Mar 2006

Keywords

  • Confidence interval
  • Cumulative exposure model
  • Exponential distribution
  • Marginal distribution
  • Maximum likelihood estimate
  • Method of moment estimate
  • Order statistics
  • Pivotal quantity
  • Type II censoring

Fingerprint

Dive into the research topics of 'Analysis of a simple step-stress life test with a random stress-change time'. Together they form a unique fingerprint.

Cite this