Exchange algorithms for constructing model-robust experimental designs

Byran J. Smucker, Enrique Del Castillo, James Landis Rosenberger

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

Optimal experimental design procedures, utilizing criteria such as D-optimality, are useful for producing designs for quantitative responses, often under nonstandard conditions such as constrained design spaces. However, these methods require a priori knowledge of the exact form of the response function, an often unrealistic assumption. Model-robust designs are those that, from our perspective, are efficient with respect to a set of possible models. In this paper, we develop a model-robust technique motivated by a connection to multiresponse D-optimal design. This link spawns a generalization of the modified Fedorov exchange algorithm, which is then used to construct exact model-robust designs. We also study the effectiveness of designs robust for a small set of models compared with designs that account for much larger sets. We give several examples and compare our designs with two model-robust procedures in the literature.

Original languageEnglish (US)
Pages (from-to)28-42
Number of pages15
JournalJournal of Quality Technology
Volume43
Issue number1
DOIs
StatePublished - Jan 1 2011

Fingerprint

Design of experiments
Experimental design
Robust design

All Science Journal Classification (ASJC) codes

  • Safety, Risk, Reliability and Quality
  • Strategy and Management
  • Management Science and Operations Research
  • Industrial and Manufacturing Engineering

Cite this

Smucker, Byran J. ; Del Castillo, Enrique ; Rosenberger, James Landis. / Exchange algorithms for constructing model-robust experimental designs. In: Journal of Quality Technology. 2011 ; Vol. 43, No. 1. pp. 28-42.
@article{35a15463483b4649970c19513c017640,
title = "Exchange algorithms for constructing model-robust experimental designs",
abstract = "Optimal experimental design procedures, utilizing criteria such as D-optimality, are useful for producing designs for quantitative responses, often under nonstandard conditions such as constrained design spaces. However, these methods require a priori knowledge of the exact form of the response function, an often unrealistic assumption. Model-robust designs are those that, from our perspective, are efficient with respect to a set of possible models. In this paper, we develop a model-robust technique motivated by a connection to multiresponse D-optimal design. This link spawns a generalization of the modified Fedorov exchange algorithm, which is then used to construct exact model-robust designs. We also study the effectiveness of designs robust for a small set of models compared with designs that account for much larger sets. We give several examples and compare our designs with two model-robust procedures in the literature.",
author = "Smucker, {Byran J.} and {Del Castillo}, Enrique and Rosenberger, {James Landis}",
year = "2011",
month = "1",
day = "1",
doi = "10.1080/00224065.2011.11917843",
language = "English (US)",
volume = "43",
pages = "28--42",
journal = "Journal of Quality Technology",
issn = "0022-4065",
publisher = "American Society for Quality",
number = "1",

}

Exchange algorithms for constructing model-robust experimental designs. / Smucker, Byran J.; Del Castillo, Enrique; Rosenberger, James Landis.

In: Journal of Quality Technology, Vol. 43, No. 1, 01.01.2011, p. 28-42.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Exchange algorithms for constructing model-robust experimental designs

AU - Smucker, Byran J.

AU - Del Castillo, Enrique

AU - Rosenberger, James Landis

PY - 2011/1/1

Y1 - 2011/1/1

N2 - Optimal experimental design procedures, utilizing criteria such as D-optimality, are useful for producing designs for quantitative responses, often under nonstandard conditions such as constrained design spaces. However, these methods require a priori knowledge of the exact form of the response function, an often unrealistic assumption. Model-robust designs are those that, from our perspective, are efficient with respect to a set of possible models. In this paper, we develop a model-robust technique motivated by a connection to multiresponse D-optimal design. This link spawns a generalization of the modified Fedorov exchange algorithm, which is then used to construct exact model-robust designs. We also study the effectiveness of designs robust for a small set of models compared with designs that account for much larger sets. We give several examples and compare our designs with two model-robust procedures in the literature.

AB - Optimal experimental design procedures, utilizing criteria such as D-optimality, are useful for producing designs for quantitative responses, often under nonstandard conditions such as constrained design spaces. However, these methods require a priori knowledge of the exact form of the response function, an often unrealistic assumption. Model-robust designs are those that, from our perspective, are efficient with respect to a set of possible models. In this paper, we develop a model-robust technique motivated by a connection to multiresponse D-optimal design. This link spawns a generalization of the modified Fedorov exchange algorithm, which is then used to construct exact model-robust designs. We also study the effectiveness of designs robust for a small set of models compared with designs that account for much larger sets. We give several examples and compare our designs with two model-robust procedures in the literature.

UR - http://www.scopus.com/inward/record.url?scp=79951882464&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79951882464&partnerID=8YFLogxK

U2 - 10.1080/00224065.2011.11917843

DO - 10.1080/00224065.2011.11917843

M3 - Article

AN - SCOPUS:79951882464

VL - 43

SP - 28

EP - 42

JO - Journal of Quality Technology

JF - Journal of Quality Technology

SN - 0022-4065

IS - 1

ER -