|
Activity Number:
|
48
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 8, 2004 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Section on Physical and Engineering Sciences
|
| Abstract - #300788 |
|
Title:
|
Nonparametric Failure Distribution Analysis for Accelerated Degradation Test
|
|
Author(s):
|
Shuen-Lin Jeng*+
|
|
Companies:
|
Tunghai University
|
|
Address:
|
No. 181, Sec. 3, Taichung-Kan Rd. , Taichung, 407-04, Taiwan
|
|
Keywords:
|
accelerated degradation test ; nonparametric failure distribution ; pseudo-failure time ; random parameter degradation model
|
|
Abstract:
|
A typical procedure of a life test is to treat the survive observations as censored data. This way, there is an enormous information loss on estimating the failure distribution when the test units have high reliability. Accelerated degradation data could provide information of failure distribution for units that have not failed in the normal usage condition when the degradation measurements are closely related to the failure mechanism. We will first review the existence procedures on finding the failure distribution from degradation data. When the degradation model is nonlinear and several parameters of the model are random, the failure distribution is hard to be calculated and is usually evaluated with numerical methods. An easier but approximately correct way is to use "pseudo-failure times" which is generated from the degradation model. Following the idea of using pseudo-failure times, we suggest a nonparametric procedure to obtain the failure distribution and give comparisons with previous results.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2004 program |
