Adequate station classification is of great importance for the study of passenger flow characteristics and the development of land around stations. To address the problem that it is difficult to accurately classify new stations before the opening of new lines due to unknown passenger flow data, this paper proposes a classification method for new line stations based on the mapping relationship between the built environment and subway passenger flow. First, we cluster existing stations and calculate the passenger flow characteristic intervals of each category of stations. Random forests are used to screen important built environment factors that affect the classification of categories. Second, based on the above characteristics of passenger flow and built environment, the mapping relationship between them is determined by logarithmic transformation fitting. Finally, we calculate the unknown passenger flow characteristics of new line stations according to the mapping relationship and built environment characteristics, and then realize the classification of new line stations. An example of the Beijing subway system is taken to verify the effectiveness of the method. The results show that the method can realize the early classification of station types before the opening of new lines, and has good classification effect and applicability.
A Classification Method for New Line Stations Based on the Mapping Relationship Between Subway Passenger Flow and Built Environment
24.09.2023
1218116 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Spatially Varying Relation between Built Environment and Station-Level Subway Passenger-Distance
DOAJ | 2022
|Simulation Optimization of Holiday Large Passenger Flow Organization in Subway Stations
Springer Verlag | 2024
|Passenger Flow Detection in Subway Stations Based on Improved You Only Look Once Algorithm
Transportation Research Record | 2023
|A Parallel Passenger Flow Management System for Outside Subway Stations Based on ACP Approach
British Library Conference Proceedings | 2018
|