Abstract: An algorithm for predicting passenger flow parameter of comprehensive passenger transport hub was proposed. According to the study of spatial correlation degree between bottleneck and correlative points in comprehensive passenger transport hub, real passenger flow parameters of bottleneck and correlative points were used to achieve short-term prediction of passenger flow parameter in bottleneck. The results show that this algorithm can quickly respond to perturbation of passenger flow, and it has the capability of anti-interference in the passenger flow safety forewarning of high-speed railway comprehensive passenger transport hub.