Speakers
Description
As global urbanization accelerates and carbon neutrality targets are pursued, urban metro transit has become a key element of sustainable transportation due to its efficiency, low emissions, and affordability. Meanwhile, the quality of first-and-last-mile connections plays a crucial role in metro accessibility(Zuo et al., 2020). In China, the rapid expansion of dockless shared bicycle services has increasingly facilitated first-and-last-mile connections (Shen et al., 2022). However, challenges remain, including long connection distances, high detour rates, and inadequate cycling infrastructure, which limit metro service capacity and travel efficiency. Current methods for processing shared bike trip data face limitations, including low accuracy in extracting connecting trips from the overall dataset, and insufficient rigor in reconstructing cycling routes from origin-destination (OD) data, which fail to align with the actual road conditions(Wei et al., 2023). Meanwhile, existing studies predominantly focus on the isolated connection feature analyses, the impact of the built environment on trips(Tong et al., 2023), or preferences for connecting transport modes (van Kuijk et al., 2022), lacking a comprehensive framework to identify connection features and directly guide improvements in the first-and-last-mile connection environment. This gap significantly hinders the advancements in metro transit accessibility and its equity.
The research aims to develop a big-data-driven connecting trip identification algorithm and analyze the built environment features in bike-sharing catchment areas through a multi-dimensional approach, thereby establishing a systematic evaluation framework for the bicycle-metro integration system and providing quantitative guidance for optimizing first-and-last-mile connection environments. To achieve these objectives, the study utilizes dockless shared bicycle data collected from all metro stations in Shanghai’s Jing’an District, an administrative area featuring diverse urban functions and station types, during a typical workday. An innovative identification approach is proposed, integrating a stepping method, sliding window detection, and DBSCAN clustering to more accurately identify connecting trips and delineate bike-sharing catchment areas for each station. In addition, the Baidu Maps cycling route planning API is employed to simulate the cycling routes under real-world urban conditions. Based on these identified trips and routes, multi-dimensional analyses are conducted at both the station catchment area and connecting road segment levels. At the bike-sharing catchment area level, key indicators, such as the dominance of cycling connections, road detour rates, and tidal balance indices, are measured to characterize the overall features of every individual station. At the connecting road segment level, thirteen assessment indicators are established across four dimensions, including infrastructure, road connectivity, function and vitality, and environment. Both the Analytic Hierarchy Process (AHP) and the entropy weight method are applied to assign weights, thereby evaluating the bikeability of each road segment. The usage frequency of road segments, calculated from identified routes, serves as another feature.
Regarding the results, the classification of each station’s bike-sharing catchment area and connecting road segments is achieved based on these features through hierarchical clustering and a two-dimensional evaluation matrix. Drawing upon these clusters, differentiated optimization strategies are proposed for the connection environment, with specific case examples illustrating how the strategies are tailored. In conclusion, this research introduces an operational diagnostic and optimization framework for first-and-last-mile connection environments of bicycle-metro integration systems. It highlights how big data and emerging technologies can drive the transformation of urban transportation, aligning with global goals of carbon neutrality and sustainable development. The findings provide actionable insights for policymakers and urban managers to implement targeted interventions, fostering an efficient, eco-friendly, and inclusive urban transport system.
References
SHEN, S., LV, C. X., ZHU, H., SUN, L. J. & WANG, R. C. (2022) Potentials and Prospects of Bicycle Sharing System in Smart Cities: A Review. Ieee Sensors Journal, 22, pp.7519-7533.
TONG, Z. M., ZHU, Y., ZHANG, Z. Y., AN, R., LIU, Y. L. & ZHENG, M. (2023) Unravel the spatio-temporal patterns and their nonlinear relationship with correlates of dockless shared bikes near metro stations. Geo-Spatial Information Science, 26, pp.577-598.
VAN KUIJK, R. J., CORREIA, G., VAN OORT, N. & VAN AREM, B. (2022) Preferences for first and last mile shared mobility between stops and activity locations: A case study of local public transport users in Utrecht, the Netherlands. Transportation Research Part a-Policy and Practice, 166, pp.285-306.
WEI, J. M., WANG, Y., LIU, Z. & CHEN, Y. Y. (2023) Correlation between the built environment and dockless bike-sharing trips connecting to urban metro stations. Journal of Transport and Land Use, 16, pp.131-161.
ZUO, T., WEI, H., CHEN, N. & ZHANG, C. (2020) First-and-last mile solution via bicycling to improving transit accessibility and advancing transportation equity. Cities, 99, 14.
| Keywords | Transportation Big Data; First-and-Last-Mile Connectivity; Shared Bike; Bicycle-Metro Integration |
|---|---|
| Best Congress Paper Award | Yes |
