We study in this paper the possibility of contribution in solving the vehicle routing problem (VRP) by using the improved ant colony system ( IACS) , which is one of the optimization problems that, because of its Real Life applications, has attracted a lot of attention at the present time. It is a problem of the NP-hard type. However, because of the complication of polynomial time there is still no algorithm providing us with the optimal solution of this problem. All the used algorithms give solutions that are close to the optimal one . We present the improved ant colony system algorithm that, based on ant colony system algorithm, possesses a new state transition rule, a new pheromone updating rule and diverse local search approaches . The experimental results of the proposed ( IACS) algorithm compared with the results of well-known standard tests show that our IACS yields better solutions than the other ant algorithms in the literature and is competitive with other meta-heuristic approaches in terms of quality(run time and number of good solutions ).