Pheromone-based crossover operator of genetic algorithm for the traveling salesman problem

A new pheromone-based crossover operator of genetic algorithm for the traveling salesman problem was proposed, and the working process of the operator was analyzed when constructing offspring. When constructing offspring, the proposed operator utilizes both local and global irdormation. The local information includes edge lengths and adjacency relations, while the global information is stored as pheromone trails. The proposed operator was tested in a pure genetic algorithm framwork using the TSP benchmark instances. Experimental results show its better performance in both of speed and accuracy than other well known crossover operators.