Minimization of Crossover Operator in Genetic Algorithm

Abstract

Crossover is a genetic operator used in genetic algorithms (GA) for varying the functionalities of chromosomes from one generation to the other. It is the base of any GA approach. It has a vital role in obtaining the optimal solution. The selection of appropriate crossover and how it is used on the individuals is highly significant to get better fitness value. This paper focuses on crossover operator and presents a technique to minimize it in GA such that the optimal solutions are not compensated. In this work, the attempt has been made on reducing the number of crossover operation such that the new generation contains the best traits of the old ones and at the same time the crossover overhead is reduced. As an analysis of the proposed technique, the same has been simulated using MATLAB and the effect has been investigated on the popular traveling salesman problem. The results obtained are satisfactory.