At present, most taxi systems at airport hubs generally have the problems of unreasonable vehicle resource allocation with lower level-of-service and higher operational cost. In order to improve the level of taxi transfer service and increase the operational efficiency of taxi system at airport hubs, we focus on optimizing the resource allocation of the taxi departure system at airport hubs. Firstly, the optimization model of passenger departure process in taxi pick-up area is proposed, and a genetic algorithm is designed to solve the model. In addition, aiming at the problem of insufficient taxi capacity resources, we formulate the taxi dispatching model for airport hubs to obtain the taxi dispatching plan under the optimal passenger departure scheme. Finally, taking the Beijing Capital International Airport as an example, we carry out a simulation study by the proposed models. The optimization results show that the taxi pick-up area can meet the travel needs of passengers in peak hours, and the total cost is reduced by 25.2%. The taxi dispatching model can dynamically dispatch taxis to enter or leave the storage pool according to the number of taxis in the storage pool and the supplementary arrival rate of taxis. Our research results can provide the reference for optimizing the taxi system at other large-scale passenger transportation hubs.