Abstract
Until now, in the literature, a variety of acceptance reliability sampling plans have been developed based on different life test plans. In most of the reliability sampling plans, the decision procedures to accept or reject the corresponding lot are developed based on the lifetimes of the items observed on tests, or the number of failures observed during a pre-specified testing time. However, frequently, the items are subject to degradation phenomena and, in these cases, the observed degradation level of the item can be used as a decision statistic. In this paper, we develop a variables acceptance sampling plan based on the information on the degradation process of the items, assuming that the degradation process follows the inverse Gaussian process. It is shown that the developed sampling plan improves the reliability performance of the items conditional on the acceptance in the test and that the lifetimes of items after the reliability sampling test are stochastically larger than those before the test. A study comparing the proposed degradation-based sampling plan with the conventional sampling plan which is based on a life test is also performed.
Original language | English |
---|---|
Pages (from-to) | 393-409 |
Number of pages | 17 |
Journal | Journal of Applied Statistics |
Volume | 48 |
Issue number | 3 |
DOIs | |
State | Published - 2021 |
Bibliographical note
Publisher Copyright:© 2020 Informa UK Limited, trading as Taylor & Francis Group.
Keywords
- Variables sampling plan
- degradation test
- inverse Gaussian process
- mixture distribution
- stochastic ordering