|
Activity Number:
|
362
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 6, 2008 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #302274 |
|
Title:
|
Bayesian Optimal Single Arrays for Robust Parameter Design
|
|
Author(s):
|
Lulu Kang*+ and Roshan J. Vengazhiyil
|
|
Companies:
|
Georgia Institute of Technology and Georgia Institute of Technology
|
|
Address:
|
School of Industrial and Systems Engineering, Atlanta, GA, 30332,
|
|
Keywords:
|
Design of experiment ; Exchange algorithms ; Quality improvement ; Variation reduction
|
|
Abstract:
|
We suggest a new optimal design criterion for robust parameter design experiment. This criterion is built on a Bayesian framework which has incorporated the hierarchical ordering principle. Compared with some existing design criteria, this criterion still follows the hierarchical ordering principle and fully focuses on variation reduction. We also develop a greedy-exchange algorithm to search for the optimal design. Our proposed method is very general and not restricted to a specific kind of designs such as orthogonal arrays. It can be extended to more complicated situations when there are mixed-level qualitative and quantitative factors, or even internal noise factors.
|
