A demonstration of the utility of fractional experimental design for finding optimal genetic algorithm parameter settings

D.J. Stewardson, R.I. Whitfield

Research output: Contribution to journalArticle

3 Citations (Scopus)
34 Downloads (Pure)

Abstract

This paper demonstrates that the use of sparse experimental design in the development of the structure for genetic algorithms, and hence other computer programs, is a particularly effective and efficient strategy. Despite widespread knowledge of the existence of these systematic experimental plans, they have seen limited application in the investigation of advanced computer programs. This paper attempts to address this missed opportunity and encourage others to take advantage of the power of these plans. Using data generated from a full factorial experimental design, involving 27 experimental runs that was used to assess the optimum operating settings of the parameters of a special genetic algorithm (GA), we show that similar results could have been obtained using as few as nine runs. The GA was used to find minimum cost schedules for a complex component assembly operation with many sub-processes.
Original languageEnglish
Pages (from-to)132-138
Number of pages6
JournalJournal of the Operational Research Society
Volume55
Issue number2
DOIs
Publication statusPublished - 2004

Fingerprint

Design of experiments
Demonstrations
Genetic algorithms
Computer program listings
Genetic algorithm
Experimental design
Costs
Schedule

Keywords

  • genetic algorithms
  • scheduling
  • sequential experimentation
  • regression
  • optimization
  • design engineering

Cite this

@article{25048d877a8843c1ae3ea39eb6e440b4,
title = "A demonstration of the utility of fractional experimental design for finding optimal genetic algorithm parameter settings",
abstract = "This paper demonstrates that the use of sparse experimental design in the development of the structure for genetic algorithms, and hence other computer programs, is a particularly effective and efficient strategy. Despite widespread knowledge of the existence of these systematic experimental plans, they have seen limited application in the investigation of advanced computer programs. This paper attempts to address this missed opportunity and encourage others to take advantage of the power of these plans. Using data generated from a full factorial experimental design, involving 27 experimental runs that was used to assess the optimum operating settings of the parameters of a special genetic algorithm (GA), we show that similar results could have been obtained using as few as nine runs. The GA was used to find minimum cost schedules for a complex component assembly operation with many sub-processes.",
keywords = "genetic algorithms, scheduling, sequential experimentation, regression, optimization, design engineering",
author = "D.J. Stewardson and R.I. Whitfield",
year = "2004",
doi = "10.1057/palgrave.jors.2601703",
language = "English",
volume = "55",
pages = "132--138",
journal = "Journal of Operational Research Society",
issn = "0160-5682",
number = "2",

}

TY - JOUR

T1 - A demonstration of the utility of fractional experimental design for finding optimal genetic algorithm parameter settings

AU - Stewardson, D.J.

AU - Whitfield, R.I.

PY - 2004

Y1 - 2004

N2 - This paper demonstrates that the use of sparse experimental design in the development of the structure for genetic algorithms, and hence other computer programs, is a particularly effective and efficient strategy. Despite widespread knowledge of the existence of these systematic experimental plans, they have seen limited application in the investigation of advanced computer programs. This paper attempts to address this missed opportunity and encourage others to take advantage of the power of these plans. Using data generated from a full factorial experimental design, involving 27 experimental runs that was used to assess the optimum operating settings of the parameters of a special genetic algorithm (GA), we show that similar results could have been obtained using as few as nine runs. The GA was used to find minimum cost schedules for a complex component assembly operation with many sub-processes.

AB - This paper demonstrates that the use of sparse experimental design in the development of the structure for genetic algorithms, and hence other computer programs, is a particularly effective and efficient strategy. Despite widespread knowledge of the existence of these systematic experimental plans, they have seen limited application in the investigation of advanced computer programs. This paper attempts to address this missed opportunity and encourage others to take advantage of the power of these plans. Using data generated from a full factorial experimental design, involving 27 experimental runs that was used to assess the optimum operating settings of the parameters of a special genetic algorithm (GA), we show that similar results could have been obtained using as few as nine runs. The GA was used to find minimum cost schedules for a complex component assembly operation with many sub-processes.

KW - genetic algorithms

KW - scheduling

KW - sequential experimentation

KW - regression

KW - optimization

KW - design engineering

UR - http://www.cad.strath.ac.uk/~ianw/publications/resources/Ref24.pdf

UR - http://dx.doi.org/10.1057/palgrave.jors.2601703

U2 - 10.1057/palgrave.jors.2601703

DO - 10.1057/palgrave.jors.2601703

M3 - Article

VL - 55

SP - 132

EP - 138

JO - Journal of Operational Research Society

JF - Journal of Operational Research Society

SN - 0160-5682

IS - 2

ER -