A Solution to Traveling Salesman Problem Using Hybrid Genetic Algorithm

To overcome the shortcomings of Simple Genetic Algorithm (SGA) in solution of Traveling Salesman Problem (TSP), a Hybrid Genetic Algorithm (HGA) was developed, in which three improving operators such as greedy search, sub-tour exhaustive searching, and 2-Opt were incorporated into the iteration process of simple genetic algorithm. These extra improving operators were inserted immediately after three global operators of SGA, and expected to replace an old individual with a new one whenever the new individual is fitter than the old one. Two simulation examples showed the effectiveness of the hybrid method as well as the improvement upon the quality for solution of moderate sized traveling salesman problems. The hybrid algorithm has potential application in many fields such as military, traffic, logistics, and so on.