Abstract:

Accurately describing the mapping relationship between built environment and urban rail transit passenger flow is an important prerequisite for passenger flow demand forecast. In view of the problems of incomplete and multi-dimensional sparse data of OD between stations, this paper proposes a research method of mapping relationship between built environment and passenger flow at the zone level. Firstly, leveraging the characteristics of natural geography and passenger flow directionality, a two-tier clustering method is devised to aggregate passenger flow demand at the zone level. Subsequently, an built environment indicator system is formulated, encompassing two dimensions: the attraction capacity of O/D zones and the accessibility characteristics of OD pairs. Thirdly, a methodology based on the Gradient Boosting Decision Tree (GBDT) model is introduced to characterize the relationship between built environment features and passenger flow, delving into the influence intensity and threshold values of individual factors on passenger flow. Finally, an empirical analysis is carried out on Beijing subway. The results show that the mapping relationship between built environment and passenger flow at zone-to-zone level has spatial and temporal heterogeneity, nonlinear characteristics and threshold effects. The zoning-based research perspective effectively addresses issues of data sparsity, leading to a 7.4% improvement in prediction accuracy. OD impedance emerges as the primary feature influencing passenger flow, accounting for up to 38.4% of the explanatory power, while demographic and economic characteristics serve as secondary factors, exhibiting significant threshold effects. Consequently, in the process of urban rail transit planning, primary attention should be given to optimizing network topology and enhancing transportation accessibility, followed by a thorough consideration of the impact of regional economic activities. This research provides quantitative analysis tools for urban planners, which can help planners determine the effective range and adjustment space of built environment indicators, thereby provide references for improving the operation efficiency of rail transit.

Key words: urban rail traffic, built environment, gradient Boosting decision tree, passenger flow at zone-to-zone level, nonlinear relationship