Online Program Home
My Program

Abstract Details

Activity Number: 29
Type: Contributed
Date/Time: Sunday, July 31, 2016 : 2:00 PM to 3:50 PM
Sponsor: Quality and Productivity Section
Abstract #320121
Title: Statistical Lifetime Inference Based on Skew-Normal Accelerated Destructive Degradation Test Model
Author(s): Chih-Chun Tsai* and Chien-Tai Lin
Companies: Tamkang University and Tamkang University
Keywords: accelerated destructive degradation tests ; expectation-maximization algorithm ; highly reliable products ; skew-normal distribution ; model mis-specification
Abstract:

The accelerated destructive degradation test (ADDT) method provides an effective way to assess the reliability information of highly reliable products whose quality characteristics degrade over time, and can be taken only once on each tested unit during the measurement process. Conventionally, engineers assume that the measurement error follows the normal distribution. However, degradation models based on this normality assumption often do not apply in practical applications. To relax the normality assumption, the skew-normal distribution is adopted in this study because it preserves the advantages of the normal distribution with the additional benefit of flexibility with regard to skewness and kurtosis. Here, motivated by polymer data, we propose a skew-normal nonlinear ADDT model, and derive the analytical expressions for the product's lifetime distribution along with its corresponding th percentile. Then, the polymer data are used to illustrate the advantages gained by the proposed model. Finally, we addressed analytically the effects of model mis-specification when the skewness of measurement error are mistakenly treated, and the obtained results reveal that the impact from the


Authors who are presenting talks have a * after their name.

Back to the full JSM 2016 program

 
 
Copyright © American Statistical Association