Classifying Transit-Oriented Development Neighborhoods Based on Network Analysis

Abstract

The aim of this research is to propose a method for developing and exploring the typology of public transportation stations based on network analysis. Transit Oriented Development (TOD), a type of urban development that aims at increasing both the usage of public transportation and the walkability within the neighborhood around the station, considers the influences of urban form, built environment, traffic flow and movement patterns in order to integrate transportation, land use and environmental policies. This notion has later been extended into the “Network TOD”, which is the network approach on a broader geographical scale than TOD, as it presents the potential of both creating the livable and attractive neighborhoods around the stations and at the same time shaping the polycentric cities in order to mitigate urban sprawl. The strength of the network approach consists in offering a more holistic perspective of evaluating the node, i.e. the station, based on the role it plays in the whole network system. In the case of urban and transportation planning, the benefit of employing this approach is that it not only focuses on the quality of urban environment within the boundary of the TOD neighborhood, but also considers the relationship between TOD neighborhoods.

In the last few years, due to the development of information technology and the increased availability of data sources, data-driven urban morphology research has vastly advanced. By employing a machine learning approach, this thesis develops an automatic, parametric, scalable and reproducible method based on network analysis and aims to meet the following objectives: • Assessing the existing models in network analysis and dealing with limitations of their validity (such as the size effect and the placement effect) on the values of the indicators. The results can be helpful for determining the optimal size and the location of the catchment area in order to mitigate the size and placement effects on network analysis indicators. • Evaluating the connectivity and resilience of two types of networks, namely public transit network and street network, by exploring their geometric and topological properties. • Quantitatively classifying and evaluating the importance of stations in the railway public transport system based on 1) their structural importance in the transportation network; 2) topological characteristics of street network in the neighborhoods; and 3) accessibility, density, and diversity of the points of interest in the neighborhoods. • Determining a suitable location of an intervention that matches the challenges faced by different types of neighborhoods in improving the walkability in the eighborhood.

Finally, being motivated by the outbreak of COVID-19 in 2020, we hope that this research can help urban planners to make informed and data-driven decisions on the location of facilities and services, such as temporary test stations and mobile vaccination centers, during emergencies and crises (for example, a pandemic).