Departure efficiency evaluation of a comprehensive transport hub based on Wi-Fi probe data and a multilayer hybrid model

Abstract

Evaluation of the passenger departure efficiency of a comprehensive transport hub is essential for traffic managers. Through the evaluation, security risks in the hub can be found in time to ensure the safe departure of passengers. The attention of existing studies has focused on the analysis of the overall situation of the hub, and the quantitative description of departure status in different connection areas inside the hub is insufficient. In this study, a multilayer hybrid model based on an analytic hierarchy process and entropy weight method was established. The data collected using Wi-Fi probe technology were clustered by a K-means algorithm. The first level of the model was divided according to the connection areas of the passenger hub, and the second level was based on the number of stranded people, wait time and departure time in each connection area. It was found that the SP index has the greatest impact on departure efficiency. In addition, the impact of passenger flow aggregation on each connection area is different, and the management department should treat it accordingly. The applicability of the proposed multilayer hybrid model was verified in the example of the Chongqing north railway station.