ANNUAL CONGRESS 2025

Exploring Interactions between Urban Rail Transit Network Characteristics and Station-Area Functional Morphology: A Typological Analysis and Flow Prediction of Shanghai Metro Based on Multi-Source Data

Not scheduled

20m

Yildiz Technical University, Istanbul

Oral Track 03 | MOBILITY

Speakers

Mr Yuhao SHI (College of Architecture and Urban Planning, Tongji University)Ms Jinghao HEI (College of Design and Innovation, Tongji University)Ms Jiaming Xing (College of Architecture and Urban Planning, Tongji University)Mr Yuxuan WANG (College of Architecture and Urban Planning, Tongji University)

Description

In the era of global crises, urban planning is confronted with multiple challenges, including climate change, social inequality, and resource scarcity. Urban rail transit networks, as critical infrastructure, significantly reshape urban spatial agglomeration patterns and morphological characteristics by guiding population flows (Yang et al., 2020). Consequently, they have become a vital lever for promoting urban sustainability and social inclusiveness (Xiao et al., 2021). In China, the transit-oriented development (TOD) model has emerged as a planning tool to address inefficient urban land use, guiding urban spatial expansion(Shen and Wu, 2020). As a result, the interaction between urban rail network structures and the functional differentiation of urban spaces needs a more comprehensive reassessment.
However, existing studies on typologies of station-area spaces often overlook the characteristics of the rail network itself (Feng et al., 2023), while passenger flow predictions tend to overemphasize the current network's flow patterns, neglecting the influence of land use and functional characteristics on transit usage (Zhang et al., 2024). To address these gaps, this study examines the Shanghai metro system from the dual perspectives of complex network analysis and functional morphological measurement. It aims to uncover the interaction mechanisms between station network characteristics and station-area functional morphology, employing typological induction to better capture real-world patterns and offering scientific and innovative insights for future urban transit planning through passenger flow predictions.
Our research unfolds in four stages:
1, We utilize complex network analysis to measure the network characteristics of all operational metro stations in Shanghai. Metrics such as betweenness centrality, closeness centrality, and eigenvector centrality are used to identify the structural roles and connectivity of stations within the transit network at the node level.
2, Station-area boundaries are defined using a 600-meter buffer or a 15-minute walking isochrone. Indicators across three dimensions—functionality, vitality, and environment—comprising 18 metrics in total, are measured to capture the spatial quality and functional characteristics of station areas at the place level. These two datasets form the foundation of our analysis.
3, We examine the interaction patterns between rail network characteristics and station-area functional morphology through descriptive statistics and correlation matrices. Results indicate that stations with high closeness centrality exhibit strong regional accessibility and overall transportation convenience. In contrast, stations with high betweenness centrality serve as bridges for long-distance flows, facilitating interactions between different urban functional zones. Using the K-means clustering method, we classify Shanghai metro stations into six types: central homogeneous, suburban terminal, urban residential, central employment, urban public service, and suburban central. Typical station case studies validate the practical effectiveness of these classifications.
4, Using operational stations as the training set and unbuilt or planned stations as the testing set, we apply various machine learning models to predict future passenger flows. The random forest model achieves the highest prediction accuracy. SHAP analysis identifies key factors influencing passenger flow, including social media activity in station areas, floor area ratio, and office rental levels. Predictions highlight that the northern part of Line 19 in the Lujiazui area is likely to become a new high-traffic transit hub.
In conclusion, this study, adopting a synergistic perspective on rail transit networks and station-area functional morphology, provides scientific evidence for optimizing rail transit planning and urban spatial layouts. Based on clustering and prediction results, we propose resilience-based transit system strategies aimed at addressing the diverse mobility needs of different demographic groups while enhancing the adaptability of the transportation network. Our findings not only offer practical guidance for Shanghai’s metro system planning but also call for a global reevaluation of urban transit development models in a fairer and more sustainable manner, providing insights into addressing challenges of inequality in infrastructure services.

References

YANG, L., CHAU, K. W., SZETO, W. Y., CUI, X. & WANG, X. 2020. Accessibility to transit, by transit, and property prices: Spatially varying relationships. Transportation Research Part D: Transport and Environment, 85, 102387.
XIAO, L., LO, S., LIU, J., ZHOU, J. & LI, Q. 2021. Nonlinear and synergistic effects of TOD on urban vibrancy: Applying local explanations for gradient boosting decision tree. Sustainable Cities and Society, 72, 103063.
SHEN, J. & WU, F. 2020. Paving the way to growth: transit-oriented development as a financing instrument for Shanghai’s post-suburbanization. Urban Geography, 41, 1010-1032.
FENG, H., CHEN, Y., WU, J., ZHAO, Z., WANG, Y. & WANG, Z. 2023. Urban Rail Transit Station Type Identification Based on “Passenger Flow—Land Use—Job-Housing”. Sustainability, 15, 15103.
ZHANG, C., LIANG, Y., TIAN, T. & PENG, P. 2024. Sustainable Transportation: Exploring the Node Importance Evolution of Rail Transit Networks during Peak Hours. Sustainability, 16, 6726.