|
Activity Number:
|
142
|
|
Type:
|
Invited
|
|
Date/Time:
|
Monday, August 3, 2009 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Physical and Engineering Sciences
|
| Abstract - #303183 |
|
Title:
|
Comparing Computer-Generated Designs: Are G-Optimal Designs the Best Choice?
|
|
Author(s):
|
Douglas C. Montgomery*+
|
|
Companies:
|
Arizona State University
|
|
Address:
|
Department of Industrial Engineering, Tempe, AZ, 85287-5906,
|
|
Keywords:
|
Optimam designs ; G-optimality
|
|
Abstract:
|
A new algorithm for generating near G-optimal designs for second order models over cuboidal regions is presented. Designs created using this new method match or exceed the G-efficiency of previously reported designs. A new graphical tool, the Variance Ratio Fraction of the Design Space Plot, for comparing designs to a reference design is also introduced. Stability of prediction variance in the region of interest is a desirable property that G-optimal designs offer. However, low values for the prediction variance are useful throughout the design region not just near the boundary. We demonstrate that there are cases where the G-optimal design does minimize the maximum variance but is not competitive in prediction variance over a large proportion of the experimental region.
|
- 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 2009 program |
