Abstract: Aiming at the problems of delayed material distribution, low load rate of distribution vehicles and high routing cost of distribution in discrete manufacturing plant, an integrated strategy of material distribution and residual material recovery with soft curve time window was proposed. The path optimization model with the minimum total distribution cost considering the distribution vehicles, the route length and penalty cost of violating time window was constructed. At the same time, based on the shortcomings of ant colony algorithm in this kind of problems, a path search strategy combining randomness and determinism was proposed, and genetic algorithm and two-element optimization algorithm were introduced into the process of ant optimization to enrich the diversity of ant population and speep up the ants optimization and improve the solution quality. The example results show that compared with the traditional ant colony algorithm, the improved ant colony algorithm has significant optimization in the path cost, penalty cost and vehicle assignment cost, which verifies the feasibility and effectiveness of the algorithm.