Novel techniques for enhancing performance of genetic algorithms

Article type: Other

Authors: Sivaraj, R.

Affiliations: Velalar College of Engineering and Technology, Erode, Tamil Nadu, India. E-mail: rsivarajcse@gmail.com

Abstract: Genetic algorithms are commonly used in many types of applications. Yet they suffer from longer execution time and premature convergence. To improve both these factors, the research work proposes two new procedural modifications in the basic genetic algorithm procedure and a new population initialization mechanism. The proposed algorithms are implemented in two types of real world problems of different sizes and the results confirm the superiority of the algorithms over existing ones both in terms of execution time and optimality.

Keywords: Clustering genetic algorithm, gene grouping genetic algorithm, selective initialization

DOI: 10.3233/AIC-150661

Journal: AI Communications, vol. 29, no. 4, pp. 545-546, 2016