Spring Seminar Series - February 3, 2005
University of Minnesota
School of Statistics
College of Liberal Arts
SELC: Sequential Elimination of Level Combinations by Means of Modified
Genetic Algorithms
Abhyuday Mandal
School of Industrial and Systems Engineering
Georgia Institute of Technology
Thursday, February 3, 2005
3:30 PM, 115
Ford Hall
Minneapolis, East Bank Campus
Social at 3:00 PM, 300 Ford Hall
Abstract
To
search for an optimum in a large search space, Wu, Mao, Ma (1990)
suggested the SEL-method to find an optimal setting. Genetic algorithms
(GAs) can be used to improve upon this method. To make the search
procedure more efficient, new ideas of forbidden array and weighted
mutation are introduced. Relaxing the condition of orthogonality, GAs
are able to accommodate a variety of design points which allows more
flexibility and enhances the chance of getting the best setting in
fewer runs, particularly in the presence of interactions. The search
procedure is enriched by a Bayesian method for identifying the
important main effects and two-factor interactions. Illustration is
given with the optimization of three functions, one of which is from
Shekel's family.