Improvement of genetic algorithm using artificial bee colony

Abstract

Genetic algorithm (GA) is a part of evolutionary computing that simulates the theory of evolution and natural selection, where this technique depends on a heuristic random search. This algorithm reflects the operation of natural selection, where the fittest individuals are chosen for reproduction so that they produce offspring of the next generation. This paper proposes a method to improve GA using artificial bee colony (GABC). This proposed algorithm was applied to random number generation (RNG), and travelling salesman problem (TSP). The proposed method used to generate initial populations for GA rather than the random generation that used in traditional GA. The results of testing on RNG show that the proposed GABC was better than traditional GA in the mean iteration and the execution time. The results of testing TSP show the superiority of GABC on the traditional GA. The superiority of the GABC is clear in terms of the percentage of error rate, the average length route, and obtaining the shortest route. The programming language Python3 was used in programming the proposed methods.