Geospatial Analytics of Urban Bus Network Evolution Based on Multi-Source Spatiotemporal Data Fusion: A Case Study of Beijing, China

Xiao Li, Shaohua Wang, Liang Zhou, Yeran Sun, Jiayi Zheng, Chang Liu, Junyuan Zhou, Cheng Su, Dachuan XuISPRS Int. J. Geo Inf 2025 · DOI

TL;DR

This study provides a comprehensive spatiotemporal analysis of Beijing's bus network evolution from 2006 to 2024, revealing how its expansion, driven by urban development and administrative planning, has impacted accessibility and regional balance. By fusing multi-source data and employing advanced geospatial techniques, the research offers critical insights for optimizing urban transit systems and promoting sustainable development.

What Problem Does This Paper Address?

Urban bus networks are vital for city dwellers, yet understanding their dynamic evolution and how they adapt to changing urban landscapes is complex. While previous research has often focused on static network structures or specific aspects like service quality, there's a need for a more holistic, spatiotemporal analysis that captures the intricate interplay between network growth, urban planning, and population needs. This paper addresses this gap by investigating the evolutionary patterns, coverage efficiency, and regional balance of Beijing's bus network over nearly two decades, aiming to provide a scientific basis for sustainable urban development and transportation resource optimization.

Methods

The study employs a multi-faceted approach combining geospatial analysis and complex network theory. Key methods include:

  • Spatiotemporal Analysis: Analyzing data from multiple time points (2006, 2009, 2017, 2024) across different spatial scales (ring roads, administrative divisions) to track changes over time.
  • Coverage Rate (CR) Calculation: Using road networks and administrative divisions to segment urban areas into parcels and assess the proportion of parcels served by bus stops, reflecting service range and accessibility.
  • Population Conveniently Served by Buses (PCSB) Spatial Statistics: Creating 500-m buffer zones around bus stops and overlaying them with population density data to quantify the population benefiting from bus services in different regions.
  • Betweenness Centrality (BC) Analysis: Applying complex network theory to identify critical nodes (bus stops/routes) that lie on the shortest paths between other nodes, indicating their importance and influence within the network.
  • Multi-Distance Spatial Clustering Analysis (MDSCA): Utilizing Ripley's K(d) function to quantify the spatial distribution characteristics of bus stops and identify clustering patterns at various scales.

Datasets

The research utilizes a variety of datasets:

  • Bus Network Data: Collected for four time points (2006, 2009, 2017, 2024). Data for 2006 and 2009 were compiled from the National Science and Technology Ministry's 863 Program project and the Beijing Municipal Bureau of Statistics. Data for 2017 and 2024 were obtained from Amap, a leading digital map provider. This includes route names, number of stops, and geographic coordinates.
  • Population Data: 2020 population data at a 100m x 100m resolution, sourced from the Figshare platform. This dataset was generated using ensemble learning models and calibrated with data from China's Seventh National Population Census.
  • Road Network Data: 2024 road network data from OpenStreetMap (OSM), providing information on road names, types, and numbers.
  • Administrative Division Data: Vector data for Beijing's administrative divisions.

Key Insights and Ideas

  • Centralization and Gradual Expansion: Beijing's bus network has historically been highly concentrated within the central urban areas (especially within the Second Ring Road). However, it has shown a consistent trend of outward expansion, particularly in areas like Tongzhou and Yanqing, leading to a more balanced distribution over time.
  • Influence of Urban Planning and Geography: The diffusion of the bus network is significantly shaped by administrative boundaries and the ring road system. Geographic constraints, such as mountainous terrain in the northwest, have led to slower development and lower coverage in those regions compared to flatter, more developed areas.
  • Emergence of New Hubs: While traditional central districts remain important, the study highlights Tongzhou's emergence as a new transportation hub, reflecting its development as a sub-center.
  • PCSB as a Key Indicator: The PCSB analysis reveals significant improvements in public transport accessibility, especially in peripheral areas, though a gap persists between these areas and the core urban districts. This underscores the impact of infrastructure development on residents' daily lives.
  • Clustering and Directional Imbalance: Bus stops exhibit strong spatial clustering, with density growth being uneven and directional, favoring the southern and southeastern directions due to rapid urbanization.

Key Results

  • Network Growth: The number of bus stops and routes has increased year by year, leading to a more complex network structure.
  • Coverage Improvement: While central areas maintain high coverage, peripheral regions have seen significant improvements in PCSB, with some districts experiencing over 50% change rates. For example, Shunyi District's PCSB proportion increased from 19.54% in 2006 to 74.54% in 2024.
  • BC Shifts: Betweenness centrality has tended towards regional balance, with an increasing number of central hubs, and Tongzhou becoming a notable new hub.
  • Directional Density: Bus stop density has increased significantly in the southern direction (from 0.95 in 2006 to 2.59 in 2024), while western and northern directions show slower expansion.

Limitations and Future Work

The authors acknowledge several limitations:

  • Limited Time Period: The study period (2006-2024) is relatively short for fully interpreting long-term evolution and lacks suggestions for optimization based on a broader timeframe.
  • Incomplete Factor Analysis: The study did not analyze multiple influencing factors in depth to fully interpret the evolution characteristics of the bus network.

Future research directions suggested include:

  • Longer Time Frame Analysis: Extending the study period to explore influencing factors and mechanisms in greater depth.
  • Spatial Optimization: Combining intelligent transportation and big data technology to further optimize bus route and stop layouts for improved efficiency and service quality.
  • Broader Application: Applying the developed evaluation system to other cities to explore their network evolution and assess spatial layout rationality.
  • Integration of Other Factors: Incorporating more comprehensive analyses of population distribution, urban planning, and resource allocation to contribute to sustainable urban transportation planning and management.

Summary updated

BibTeX
@article{li2025geospatial,
  author = {Xiao Li and Shaohua Wang and Liang Zhou and Yeran Sun and Jiayi Zheng and Chang Liu and Junyuan Zhou and Cheng Su and Dachuan Xu},
  title = {{Geospatial Analytics of Urban Bus Network Evolution Based on Multi-Source Spatiotemporal Data Fusion: A Case Study of Beijing, China}},
  journal = {ISPRS Int. J. Geo Inf.},
  year = {2025},
  doi = {10.3390/ijgi14030112},
}
Figure