Genetic Algorithm for TSP Problem

The TSP is a typical NP problem. Therefore it is very important to the optimization for solving TSP problem. The genetic algorithm is one of ideal methods in solving it. However, the traditional genetic algorithm has the problems of slow convergence speed and local optimum. Aiming at the existing problems, the paper improved various operations and operators of traditional genetic algorithm, so that a better solution was put forward to solve the problem of local optimum prematurely. After making the selection operation, the elitist strategy was introduced to ensure the quality of the population gene; using adaptive mutation operator in mutation operation to improve the mutation quality and search result of algorithm. Genetic algorithm has characteristics of accurate result and fast convergence in solving TSP problem.