Improved ACO Algorithm with Pheromone Correction Strategy for the Traveling Salesman Problem

A new, improved ant colony optimization algorithm with novel pheromone correction strategy is introduced. It is implemented and tested on the traveling salesman problem. Algorithm modification is based on undesirability of some elements of the current best found solution. The pheromone values for highly undesirable links are significantly lowered by this a posteriori heuristic. This new hybridized algorithm with the strategy for avoiding stagnation by leaving local optima was tested on standard benchmark problems from the TSPLIB library and superiority of our method to the basic ant colony optimization and also to the particle swarm optimization is shown. The best found solutions are improved, as well as the mean values for multiple runs. The computation cost increase for our modification is negligible.