The dispatching of airport ground support vehicles is a critical component of airport ground operations. To enhance ground support efficiency, a two-stage model for airport ground support vehicle scheduling was proposed, addressing changes in flight schedules and unplanned additional demands. An improved genetic algorithm was designed to solve the problem. Using actual operational data from Nanjing Lukou International Airport, simulation experiments were conducted with shuttle buses and refueling trucks under sequential constraints as examples. The experimental results indicate that, compared with first-come-first-served and traditional genetic algorithm approaches, the proposed method reduces the number of vehicles used by 40.0% and 33.3%, respectively, and decreases the total support time by 41.4% and 16.1%, respectively. Compared with static scheduling results, the support service time increases by 3.1%, which is still within a reasonable range, while other metrics show significant improvements. These findings demonstrate that the dynamic scheduling model for airport ground support vehicles can improve ground operation scheduling efficiency and reduce flight delay rates.