Quantitative Evaluation of TOD Performance Based on Multi-Source Data: A Case Study of Shanghai

Abstract

Transit-oriented development (TOD) has been widely adopted as a primary urban planning strategy to better integrate transit and land use; further, the pedestrian-oriented perspective has been receiving increasing attention. However, most studies so far have only focused on few features and fail to capture comprehensive perceptions in the transportation (T), pedestrian-oriented accessibility (O), and urban development (D) dimensions. New emerging urban datasets provide a more refined and systematic approach to quantify the characteristics of metro station areas. This study offers a more efficient and convenient process and comprehensive approach to measure TOD performance. With a combination of traditional data collected by an official department, high-resolution open data, and innovative technology, large-scale analyses of 347 metro stations in Shanghai were conducted. Fifteen indicators for T, O, and D were chosen to categorize TOD performance into five clusters. Radar charts, boxplots, and colored maps were used to display numerous quantitative factors for each type. Combining the results with the Shanghai Comprehensive Plan (2017–2035) showed that the majority of Cluster 4 is located at the center of the Five New Towns. The correlation analysis between ridership and TOD performance showed that the transportation dimension indicator has a strong correlation with daily ridership, followed by the O and D indicators. Moreover, ridership per capita was found to be affected by resident density, employment density, O value, and D value, whereas no significant correlation was found between ridership per capita and T value. Population plays a pivotal role in metro passenger traffic, indicating ridership per capita had a high, strong correlation with resident density, with R = 0.658 for weekdays and R = 0.654 for weekends. This study reinterpreted the node-place method and 5Ds framework, resulting in a renewal method with new datasets and analysis tools. It contributes to providing pedestrian-oriented TOD planning methodology for urban planners and policymakers by combining T, O, and D dimensions and visualizing the results with current urban planning.