To solve the problem of crowded departure and long waiting times for passengers in terminal buildings, a generalized stochastic Petri net (GSPN) model was proposed under the consideration of check-in and security check services as a whole to find the process bottleneck. Then, an optimization method was proposed to address the identified bottleneck. Meanwhile, a comparative simulation was conducted by using AnyLogic simulation software before and after optimization. The experimental results show that after optimization, the number of passengers passing per unit hour has increased by 9.88%, the average stay time for check-in and security checks in the terminal has decreased by 6.96%.This method can effectively solves the bottleneck problem of passenger departure in the terminal and improve the throughput of the terminal.