This work suggested a new multi-ACO algorithm by introducing the basic idea in the particle swarm optimization(PSO) into solution information exchange between ant colonies.The new algorithm takes much less cost for exchanging solution information than those existing methods.The new algorithm was used to solve the VRPTW benchmark instances and was compared with one existing algorithm.The results show that the new algorithm out performs the existing methods.Exploiting the idea of individual moving in the swarm intelligence to design the rule of information exchange between ant colonies can improve the performance of multi-ACO algorithm.