Understanding metro origin-destination passenger flow dynamics: a novel entropy and machine learning perspective
Document Type
Journal Article
Publication Date
2026
Subject Area
place - urban, mode - subway/metro, infrastructure - station, land use - impacts, ridership - behaviour, land use - transit oriented development
Keywords
Metro passenger flows, OD pairs, entropy-based classification, built environment, machine learning, transit-oriented development
Abstract
This study explores how the built environment around origin and destination (OD) metro stations impacts passenger flow, using an entropy-based method to classify OD pairs into stable, unstable, and chaotic types based on daily flow rankings. Advanced machine learning is employed to explore the relationship between station environments and OD flows. Findings reveal the built environment’s varying influence. In stable OD flows, livelihood and economic services at the destination station are the primary factors driving passenger movement. In unstable OD flows, a mix of economic and recreation & health services near the origin station creates more fluctuating travel patterns. For chaotic OD flows, the presence of recreation & health services at both the origin and destination stations shapes spontaneous and unpredictable travel behaviors. Based on these findings, this study provides insights essential for guiding transit-oriented development (TOD) and underscore the importance of considering passenger flow dynamics in urban planning.
Rights
Permission to publish the abstract has been given by Taylor&Francis, copyright remains with them.
Recommended Citation
Sun, S., & Xing, Y. (2026). Understanding metro origin-destination passenger flow dynamics: a novel entropy and machine learning perspective. Transportation Planning and Technology, 49(5), 1028-1053.
