Alternatives to resolution III regular fractional factorial designs for 9 - 14 factors in 16 runs

Bradley Jones, Shilpa Madhavan Shinde, Douglas Montgomery

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Resolution III regular fractional factorial designs for 9-14 factors in 16 runs are standard designs for factor screening in industrial experimentation because of their economical run size. However, for all these designs, the main effects are completely confounded with some two-factor interaction(s), so experimenters must frequently either augment the original fractional factorial with new runs or rely on assumptions or external information to resolve ambiguities in interpretation.We propose nonregular orthogonal fractions in 16 runs for these situations that have no complete confounding of two-factor interactions and main effects. These designs allow for the unambiguous estimation of the main effects provided that there are only a few two-factor interactions. We show how these designs can be selected from the nonisomorphic 16-run orthogonal fractions and discuss potential analysis methods.

Original languageEnglish (US)
Pages (from-to)50-58
Number of pages9
JournalApplied Stochastic Models in Business and Industry
Volume31
Issue number1
DOIs
StatePublished - Jan 1 2015

Fingerprint

Fractional Factorial Design
Main Effect
Alternatives
Design for All
Fractional Factorial
Interaction Effects
Confounding
Interaction
Experimentation
Screening
Resolve
Factorial design
Factors
Design

Keywords

  • Aliases
  • Design resolution
  • Nonregular designs
  • Optimal designs
  • Resolution iii designs
  • Screening experiments

ASJC Scopus subject areas

  • Business, Management and Accounting(all)
  • Modeling and Simulation
  • Management Science and Operations Research

Cite this

Alternatives to resolution III regular fractional factorial designs for 9 - 14 factors in 16 runs. / Jones, Bradley; Shinde, Shilpa Madhavan; Montgomery, Douglas.

In: Applied Stochastic Models in Business and Industry, Vol. 31, No. 1, 01.01.2015, p. 50-58.

Research output: Contribution to journalArticle

@article{5ff4755b779b477687811662c27f5cd3,
title = "Alternatives to resolution III regular fractional factorial designs for 9 - 14 factors in 16 runs",
abstract = "Resolution III regular fractional factorial designs for 9-14 factors in 16 runs are standard designs for factor screening in industrial experimentation because of their economical run size. However, for all these designs, the main effects are completely confounded with some two-factor interaction(s), so experimenters must frequently either augment the original fractional factorial with new runs or rely on assumptions or external information to resolve ambiguities in interpretation.We propose nonregular orthogonal fractions in 16 runs for these situations that have no complete confounding of two-factor interactions and main effects. These designs allow for the unambiguous estimation of the main effects provided that there are only a few two-factor interactions. We show how these designs can be selected from the nonisomorphic 16-run orthogonal fractions and discuss potential analysis methods.",
keywords = "Aliases, Design resolution, Nonregular designs, Optimal designs, Resolution iii designs, Screening experiments",
author = "Bradley Jones and Shinde, {Shilpa Madhavan} and Douglas Montgomery",
year = "2015",
month = "1",
day = "1",
doi = "10.1002/asmb.2093",
language = "English (US)",
volume = "31",
pages = "50--58",
journal = "Applied Stochastic Models in Business and Industry",
issn = "1524-1904",
publisher = "John Wiley and Sons Ltd",
number = "1",

}

TY - JOUR

T1 - Alternatives to resolution III regular fractional factorial designs for 9 - 14 factors in 16 runs

AU - Jones, Bradley

AU - Shinde, Shilpa Madhavan

AU - Montgomery, Douglas

PY - 2015/1/1

Y1 - 2015/1/1

N2 - Resolution III regular fractional factorial designs for 9-14 factors in 16 runs are standard designs for factor screening in industrial experimentation because of their economical run size. However, for all these designs, the main effects are completely confounded with some two-factor interaction(s), so experimenters must frequently either augment the original fractional factorial with new runs or rely on assumptions or external information to resolve ambiguities in interpretation.We propose nonregular orthogonal fractions in 16 runs for these situations that have no complete confounding of two-factor interactions and main effects. These designs allow for the unambiguous estimation of the main effects provided that there are only a few two-factor interactions. We show how these designs can be selected from the nonisomorphic 16-run orthogonal fractions and discuss potential analysis methods.

AB - Resolution III regular fractional factorial designs for 9-14 factors in 16 runs are standard designs for factor screening in industrial experimentation because of their economical run size. However, for all these designs, the main effects are completely confounded with some two-factor interaction(s), so experimenters must frequently either augment the original fractional factorial with new runs or rely on assumptions or external information to resolve ambiguities in interpretation.We propose nonregular orthogonal fractions in 16 runs for these situations that have no complete confounding of two-factor interactions and main effects. These designs allow for the unambiguous estimation of the main effects provided that there are only a few two-factor interactions. We show how these designs can be selected from the nonisomorphic 16-run orthogonal fractions and discuss potential analysis methods.

KW - Aliases

KW - Design resolution

KW - Nonregular designs

KW - Optimal designs

KW - Resolution iii designs

KW - Screening experiments

UR - http://www.scopus.com/inward/record.url?scp=84923586959&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84923586959&partnerID=8YFLogxK

U2 - 10.1002/asmb.2093

DO - 10.1002/asmb.2093

M3 - Article

AN - SCOPUS:84923586959

VL - 31

SP - 50

EP - 58

JO - Applied Stochastic Models in Business and Industry

JF - Applied Stochastic Models in Business and Industry

SN - 1524-1904

IS - 1

ER -