An Empirical Study of the Effect of Parameter Combination on the Performance of Genetic Algorithms

An Empirical Study of the Effect of Parameter Combination on the Performance of Genetic Algorithms

Pi-Sheng Deng (Department of Computer Information Systems, California State University, Stanislaus, Turlock, CA, USA)
Copyright: © 2013 |Pages: 13
DOI: 10.4018/ijrat.2013070104
OnDemand PDF Download:
No Current Special Offers


Performance of genetic algorithms is affected not only by each genetic operator, but also by the interaction among genetic operators. Research on this issue still fails to converge to any conclusion. In this paper, the author focuses mainly on investigating, through a series of systematic experiments, the effects of different combinations of parameter settings for genetic operators on the performance of the author’s GA-based batch selection system, and compare the research results with the claims made by previous research. One of the major findings of the author’s research is that the crossover rate is not as a determinant factor as the population size or the mutation rate in affecting a GA’s performance. This paper intends to serve as an inquiry into the research of useful design guidelines for parameterizing GA-based systems.
Article Preview

2. Parameter Configuration For Genetic Operators

GAs solve problems by maintaining and stochastically modifying a population of candidate solutions through the application of genetic operators in generating optimal solutions. During this process, a GA explores multiple potentially promising regions in the solution space at the same time, and switches stochastically from one region to another for performance improvement.

It is commonly believed that crossover is the major operator of GAs, with mutation for preventing the population from early convergence to a certain solution before an extensive exploration of other candidate solutions is made (Goldberg, 1989; Holland, 1992a). According to Holland (1992a), it is the crossover operator that enables GAs to focus on the most promising regions in a solution space, and mutation alone does not advance the search for a solution (Goldberg, 1989). Spears (1993) also maintains that crossover is a more robust constructor of new candidate solutions than mutation. In addition, in Herrera & Lozano’s (2000) gradual distributed GA model, mutation was embedded in the crossover operator.

Complete Article List

Search this Journal:
Open Access Articles
Volume 6: 2 Issues (2018)
Volume 5: 2 Issues (2017)
Volume 4: 2 Issues (2016)
Volume 3: 2 Issues (2015)
Volume 2: 2 Issues (2014)
Volume 1: 2 Issues (2013)
View Complete Journal Contents Listing