|
Activity Number:
|
383
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 4, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Physical and Engineering Sciences
|
| Abstract - #304073 |
|
Title:
|
The Use of T-Squared Control Charts for Process-Driven Studies
|
|
Author(s):
|
Maria Weese*+
|
|
Companies:
|
The University of Tennessee
|
|
Address:
|
341 Stokely Management Center, Knoxville, TN, 37996-0532,
|
|
Keywords:
|
multivariate control chart ; process study
|
|
Abstract:
|
Univariate statistical control charts have been demonstrated to be an invaluable exploratory data analysis tool for the study of industrial processes. Multivariate statistical control charts also have a place in an industrial setting; however, their use as an exploratory tool has not been sufficiently studied. We first define a "process-driven" study as a meaningful collection and organization of data around specific facets of the process to be studied. We then make the distinction between a "data-driven" study (Phase I and Phase II paradigm) and a "process-driven" study and contrast the use of the T-squared in both studies. Both process data and simulated data are used as illustrations for our contrast.
|
