|
Activity Number:
|
203
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Physical and Engineering Sciences
|
| Abstract - #308997 |
|
Title:
|
Accelerated Destructive Degradation Test Planning
|
|
Author(s):
|
Ying Shi*+ and Luis Escobar and William Meeker
|
|
Companies:
|
Iowa State University and Louisiana State University and Iowa State University
|
|
Address:
|
119 Snedecor Hall, Ames, IA, 50010,
|
|
Keywords:
|
reliability ; large sample approximate variance ; optimum ADDT plan ; compromise ADDT plan ; Monte Carlo simulation
|
|
Abstract:
|
Accelerated Destructive Degradation Test (ADDT) is used to estimate the time at which a fixed percentage of product have a strength less than the critical degradation level when operating at normal temperature conditions. An ADDT plan specifies a set of test stress levels of temperature and evaluation time and the units' corresponding allocations to each test level. A class of degradation models is used with specified planning values for the parameters and plausible distribution for the model variability. This talk describes methods to find good ADDT plans. We show how to obtain an optimum plan to minimize the large sample approximate variance of the maximum likelihood (ML) estimate of a specified quantity. We then propose a more useful compromise plan. Also Monte Carlo simulations are used to evaluate ADDT plans. The methods are illustrated with an application for an adhesive bond.
|
