A new crossover operator based on the rough set theory for genetic algorithms

The performance of genetic algorithms (GAs) is dependent on many factors. In this paper, we have isolated one factor: the crossover operator. Commonly used crossover operators such as one-point, two-point and uniform crossover operator are likely to destroy the information obtained previously because of their random choices of crossover points. To overcome this defect, RSO, a new adaptive crossover operator based on the rough set theory, is proposed. By using RSO, useful schemata can be found and have a higher probability of surviving recombination regardless of their defining length. In this paper, the mechanism of RSO is discussed and its performance is compared with two-point crossover operator on several typical function optimization problems. The experimental results show that the proposed operator is more efficient.