Genetic algorithms for MD-optimal follow-up designs

Authors:
Chuen-Lung Chen;Rong-Ho Lin;Jianping Zhang
Affiliations:
Department of Management Information Systems, National Chengchi University, No. 64, Section 2, Chihnan Road, Taipei, Taiwan, ROC;Department of Business Management, National Taipei University of Technology, No. 1, Section 3, Chung-Hsiao E. Rd., Taipei, Taiwan, ROC;Temple-Inland Forest Product Corporation, 700 N. Temple Dr., Diboll, TX
Venue:
Computers and Operations Research
Year:
2003

Citing 6
Cited 0

Optimization of control parameters for genetic algorithms

IEEE Transactions on Systems, Man and Cybernetics
Time trends and factorial experiments

Technometrics
Connections between two-level designs of resolutions III and V

Technometrics
Adaptation in natural and artificial systems

Adaptation in natural and artificial systems
Genetic algorithms applied to the continuous flow shop problem

Computers and Industrial Engineering
Follow-up designs to resolve confounding in multifactor experiments

Technometrics

Quantified Score

Hi-index	0.01

Visualization

Abstract

The 2k-p fractional factorial design is the most widely used technique for industrial experimentation. This is because it can significantly reduce the number of experimental runs so that the application of experimental design to problems with a large number of factors becomes possible. However, the application of this technique usually causes the loss of important information. That is, some effects of the experiment may confound with each other and cannot be clearly identified. The follow-up design is a tool used to untangle the confounded effects produced in the initial experiment. In this research, a heuristic based on an effective evolutionary algorithm, Genetic Algorithms, has been developed to generate the optimal follow-up design. The heuristic has been applied in two common test examples. The result showed that the heuristic could simply find optimal follow-up designs, and dominate the existing algorithm.