Genetic algorithms for MD-optimal follow-up designs

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.

[1]  H. Wynn.  Results in the Theory and Construction of D‐Optimum Experimental Designs , 1972 .

[2]  R. D. Meyer,et al.  Finding the Active Factors in Fractionated Screening Experiments , 1993 .

[3]  B. Margolin.  Results on Factorial Designs of Resolution IV for the 2 n and 2 n 3 m Series , 1969 .

[4]  B. Margolin.  Resolution Iv Fractional Factorial Designs , 1969 .

[5]  George E. P. Box,et al.  Follow-up designs to resolve confounding in multifactor experiments , 1996 .

[6]  Peter W. M. John,et al.  Time trends and factorial experiments , 1990 .

[7]  D. Montgomery,et al.  Foldovers of 2k-p Resolution IV Experimental Designs , 1996 .

[8]  Chuen-Lung Chen,et al.  Genetic algorithms applied to the continuous flow shop problem , 1996 .

[9]  Dennis K. J. Lin,et al.  Connections between two-level designs of resolutions III and V , 1990 .

[10]  A. Atkinson,et al.  Optimal design : Experiments for discriminating between several models , 1975 .

[11]  D. Cox,et al.  Planning Experiments for Discriminating between Models , 1974 .

[12]  Chuen-Lung Chen,et al.  An application of genetic algorithms for flow shop problems , 1995 .

[13]  Volker Nissen,et al.  Evolutionary Algorithms in Management Applications , 1995 .

[14]  A. Atkinson,et al.  The design of experiments for discriminating between two rival models , 1975 .

[15]  C. Daniel.  Sequences of Fractional Replicates in the 2 p–q Series , 1962 .

[16]  John H. Holland,et al.  Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .

[17]  Volker Nissen,et al.  An Overview of Evolutionary Algorithms in Management Applications , 1995 .

[18]  John J. Grefenstette,et al.  Optimization of Control Parameters for Genetic Algorithms , 1986, IEEE Transactions on Systems, Man, and Cybernetics.

[19]  William J. Hill,et al.  Discrimination Among Mechanistic Models , 1967 .

[20]  Sidney Addelman,et al.  trans-Dimethanolbis(1,1,1-trifluoro-5,5-dimethylhexane-2,4-dionato)zinc(II) , 2008, Acta crystallographica. Section E, Structure reports online.