On the efficiency of crossover operators in genetic algorithms with binary representation

摘要: Genetic Algorithm (GA) represents robust, adaptive method successfully applied to various optimization problems. To evaluate the performance of genetic algorithm, it is common use some kind test functions. However, "no free lunch" theorem states not possible find perfect, universal solver algorithm. necessary characterize type problems for which that algorithm suitable. That would allow conclusions about based on class a problem. In crossover operator has an invaluable role. better understand in whole, role operator. The purpose this paper compare larger set operators same and their's efficiency. Results presented here confirm uniform two-point give best results but also show interesting comparisons between less used like segmented or half-uniform crossover.