Cooperative ant colony-genetic algorithm based on spark

Abstract

By taking full advantages of both the map and reduce function for the MapReduce parallel framework and the memory computation for the Spark platform, this paper designs and implements the algorithms for solving the traveling salesman problem based on ant colony algorithm on MapReduce framework and Spark platform. Next, adds the nearest neighbor selection strategy for choosing next city for the Spark platform ant colony algorithm, and combines it with genetic algorithm by using the optimal individual between ant colony algorithm and genetic algorithm, in order to update each other's best individual at the end of each iteration. Experimental results show that with the increase of ant colony size, compared to the stand-alone ant colony algorithm, MapReduce ant colony algorithm reflects the superiority of parallel computation; compared to the MapReduce ant colony algorithm, Spark platform ant colony algorithm reflects the superiority of memory computing. Cooperated with genetic algorithm, the solution has been improved significantly in its precision.