Vehicle routing problem with time windows (VRPTW) is a special kind of vehicle routing with adding time windows constraints and has a variety of applications in logistics. Many researchers have attacked the VRPTW by approximate solutions. Ant colony optimization (ACO) is a classical method to solve the VRPTW problem but the constraints of VRPTW are not used to consider customer selection. Most ACO-based optimization algorithms can suffer from the complexity of the VRPTW such as trapping in local optimum. In this paper, we present a novel ACO-based optimization method for VRPTW by using customer selection in order to decrease or solve the inefficiency of the customer selection of the ACO process. Moreover, we enhance performance searching of ACO in order to eliminate these small routes from the ACO process. Finally, we proposed the re-initialization technique in order to decrease or solve trapping in local optimum. Experiments conducted on fifty-six maps dataset have shown that the proposed method achieves encouraging performance compared to other ACOs.