|
Activity Number:
|
122
|
|
Type:
|
Invited
|
|
Date/Time:
|
Monday, August 7, 2006 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Quality and Productivity
|
| Abstract - #304964 |
|
Title:
|
The Effect of Cost and Other Constraints on Optimal Split-Plot Designs
|
|
Author(s):
|
Alejandro Heredia-Langner*+ and Timothy Robinson and Christine M. Anderson-Cook and Douglas C. Montgomery
|
|
Companies:
|
Pacific Northwest National Laboratory and University of Wyoming and Los Alamos National Laboratory and Arizona State University
|
|
Address:
|
P.O. Box 999, MSIN K5-12, Richland, WA, 99352,
|
|
Keywords:
|
split-plot designs ; genetic algorithms ; optimization
|
|
Abstract:
|
The need to use split-plot experimental designs arises when one or more factors in an experiment are so difficult or costly to change that complete randomization in run order is not practical. For this reason, split-plot designs need to be constructed using optimization algorithms that take into consideration the restrictions in randomization and any other conditions (such as model order) particular to the problem. We present a case study that shows how optimal split-plot designs change under a variety of practical conditions.
|
