Abstract: In order to better enhance the comprehensive efficiency and reduce the cost of food enterprises, for the integrated production and distribution scheduling problem (IPDS) considering a multi-plant and multi-customer scenario, a multi-agent simulation model with the optimization objectives of minimizing makespan and total cost is established, and an improved NSGA-II algorithm is proposed for solving the problem. Aiming at the two-stage integration characteristics of the problem, a two-dimensional coding structure is designed based on order allocation and processing sequence, and the simulation model acts as a decoder to resolve the behavior of each agent. In the improved NGSA-II algorithm, a combination of random and heuristic rules is used to generate the initial population, and four kinds of crossover and mutation operators are introduced according to the characteristics of the problem to enhance the genetic diversity in order to expand the search range of the algorithm; at the same time, a neighborhood search operator based on the key factories is designed to improve the local search capability of the algorithm. Finally, the test cases of a food enterprise are used to generate test cases for simulation experiments, and the experimental results show that the improved NSGA-II algorithm is superior to the existing multi-objective optimization algorithm, and the simulation model and the algorithm collaborate with the ability to optimize the diversity of the objectives, which has a broad application prospect in the production and distribution integrated scheduling.