The classical ant colony algorithm for vehicle routing problem with time windows (VRPTW) has problems of low efficiency, slow convergence and prematurity. And the discrete ant colony optimization (DACO) is proposed for these problem. It adopts the one-dimensional discrete coding that can make the data structure simpler and bring in faster convergence speed. In addition, self-convergence mode is used to calculate the optimal vehicle number rather than setting the optimal vehicle number at the beginning, which makes the algorithm more flexible and accelerates the convergence speed effectively. T...