Abstract: To overcome premature searching by standard Particle Swarm Optimization(PSO) algorithm, an improved particle swarm optimization algorithm is proposed. In the new algorithm, different particles are assigned specific tasks. Better particles are given smaller inertial weights, while worse ones are given larger inertial weights. And the particle’s inertial weights are adaptively adjusted according to its fitness function. These strategies improve the PSO algorithm at the aspects of diversity and the balance of exploration and exploitation. This paper tests the algorithm with a Traveling Salesman Problem with 14 nodes. The result shows that the algorithm can break away from local minimum earlier and it has high convergence speed and convergence ratio.

Key words: Particle Swarm Optimization(PSO), Traveling Salesman Problem(TSP), inertia weight, premature convergence