RCA: A route city attraction model for air passengers

Human movement pattern is a research hotspot of social computing and has practical values in various fields, such as traffic planning. Previous studies mainly focus on the travel activities of human beings on the ground rather than those in the air. In this paper, we use the reservation records of air passengers to explore air passengers' movement characteristics. After analyzing the effect of the route-trip length on the throughput, we find that most passengers eventually return to their original departure city and that the mobility of air passengers is not related to the route length. Based on these characteristics, we present a route city attraction (RCA) model, in which GDP or population is considered for the calculation of the attraction. The sub models of our RCA model show the better prediction performance of throughput than the radiation model and the gravity model.