|
Activity Number:
|
268
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, August 8, 2006 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Quality and Productivity
|
| Abstract - #305738 |
|
Title:
|
Distribution-Free Multivariate Process Control Based on Log-Linear Modeling
|
|
Author(s):
|
Peihua Qiu*+
|
|
Companies:
|
University of Minnesota
|
|
Address:
|
313 Ford Hall, Minneapolis, MN, 55455,
|
|
Keywords:
|
discrete measurements ; log-linear modeling ; multivariate distribution ; non-Gaussian data ; nonparametric procedures ; transformations
|
|
Abstract:
|
This paper considers statistical process control (SPC) when the process measurement is multivariate. Most existing multivariate SPC procedures assume the in-control distribution of the process measurement is known and it is a Gaussian distribution, which may not hold in applications. We demonstrate that results from conventional multivariate SPC procedures often are unreliable when the data are non-Gaussian. We suggest a methodology for estimating the in-control measurement distribution when a set of in-control data is available, which is based on log-linear modeling and takes into account the association structure of the measurement components. Based on the estimated in-control distribution, a CUSUM procedure for Phase II SPC also is suggested. This procedure does not depend on the Gaussian distribution assumption and thus is appropriate for most multivariate SPC problems.
|
- 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 2006 program |
