Analysis of Functional Responses From Robust Design Studies
- Publication:
- Journal of Quality Technology
- Date:
- October 2002
- Issue:
- Volume 34 Issue 4
- Pages:
- pp. 355-370
- Author(s):
- Nair, Vijayan N., Taam, Winson, Ye, Kenny Q.
- Organization(s):
- University of Michigan, Ann Arbor, MI, Oakland University, Rochester, MI, SUNY, Stony Brook, NY
Abstract
[This abstract is based on the authors' abstract.]
Robust design studies with functional responses are being conducted with increasing frequency for the purpose of analyzing location and dispersion effects and to optimize performance over a range of input-output values. More general and flexible methods are considered for analyzing location and dispersion effects for robust design with dynamic characteristics than the signal-to-noise ratio analysis proposed by Taguchi and others. The methods are illustrated with three applications. Two demonstrate the usefulness of functional regression techniques for location and dispersion analysis, while the third illustrates a parametric analysis with two-stage modeling. A mean-variance analysis for random selection of noise settings and a control-by-noise interaction analysis for explicitly controlled noise factors are each considered.