Abstract
When using acceleration in a degradation test, additional parameters are needed to incorporate the accelerating variables into the degradation process, requiring more statistical information to achieve the same level of estimation precision. When the increase of statistical information due to acceleration fails to compensate the information consumption caused by the additional parameters, acceleration is statistically inefficient. This paper identifies situations where acceleration is unnecessary in a degradation test when common stochastic process models are used, including the Wiener, gamma and inverse Gaussian (IG) processes. An acceleration relation index is introduced to unify different kinds of acceleration relations. It is shown that when this acceleration relation index is greater or equal to one, acceleration is always unnecessary. Otherwise, the necessity of acceleration depends on values of the model parameters and the acceleration relation index. The procedure to identify the necessity of acceleration is illustrated by a numerical example.
Original language | English |
---|---|
Title of host publication | 2016 International Conference on Industrial Engineering and Engineering Management, IEEM 2016 |
Publisher | IEEE Computer Society |
Pages | 546-550 |
Number of pages | 5 |
Volume | 2016-December |
ISBN (Electronic) | 9781509036653 |
DOIs | |
Publication status | Published - 27 Dec 2016 |
Event | 2016 International Conference on Industrial Engineering and Engineering Management, IEEM 2016 - Bali, Indonesia Duration: 4 Dec 2016 → 7 Dec 2016 |
Conference
Conference | 2016 International Conference on Industrial Engineering and Engineering Management, IEEM 2016 |
---|---|
Country/Territory | Indonesia |
City | Bali |
Period | 4/12/16 → 7/12/16 |
Keywords
- Acceleration Relation Index
- Degradation Tests
- Necessity of Acceleration
- Stochastic Process Models
ASJC Scopus subject areas
- Business, Management and Accounting (miscellaneous)
- Industrial and Manufacturing Engineering
- Safety, Risk, Reliability and Quality