A passenger flow optimization model based on improved ant colony optimization (ACO) algorithm is proposed for the passenger flow pressure faced by urban rail transit systems during peak hours. The model takes platform passing capacity, train carrying capacity and platform carrying capacity as constraints, and aims to reduce the total passenger waiting time and improve the operational efficiency. By constructing the passenger flow control model, the relationship between passenger flow and capacity constraints is analyzed, and a new passenger flow control model is proposed. The improved ACO algorithm improves the convergence speed and the quality of the solution of the algorithm through a dynamically weighted pheromone update strategy. The case study results show that compared with the existing algorithms, the proposed algorithm performs better in terms of searching ability and convergence and is able to effectively reduce the number of waiting passengers, lower the train carrying rate and increase the boarding rate. This study is of significance for optimizing passenger flow management in urban rail transit systems and provides new perspectives and methods for improving passenger travel experience.