Zhong, Chen; (2014) Urban Transformation Towards Polycentricity: Detecting Functional Urban Changes in Singapore from Transportation Data. Doctoral thesis (Ph.D), ETH Zürich.

Preview

Text eth-47214-02.pdf - Submitted Version Download (30MB) | Preview

Abstract

This research seeks a deeper understanding of urban dynamics. The main idea is to integrate urban planning knowledge with methods from geographic science, resulting in a systematic methodology for urban studies. Specifically, advanced spatial analysis methods are highlighted and applied in a study to detect polycentric urban transformation using transportation data. This research originates from observing a gap between available urban data and the information that could be extracted from such data. Information for a better understanding and management of urban change is in high demand, especially in this age of urban transformation. However, the large urban mobility data that is available and contains such information is in- sufficiently used due to a lack of analysis methods. To help fill this gap, this research proposes integrated spatial analysis methods capable of measuring the changing spatial structure of urban stocks and flows based on multiple years of transportation data. Particular interest is given to the phenomenon of polycentric urban transformation, which is an ongoing urban process in Singapore as well as many other cities. The conducted research starts from a review of state-of-the-art studies on Polycentricity. The main argument of this research is that Polycentricity is a matter of how people utilize urban space in reality. In other words, beyond physical urban settings, Polycentricity is an emerging spatial structure of urban stocks and flows in socioeconomic urban space. By assessing original plans with reference to measured spatial structures from urban mobility data, we can help to evaluate urban function- ality and planning strategies and uncover urban problems. To achieve such a measurement and assessment, this research presents a generic framework explaining how different levels of data services function in urban studies and planning. This is a general framework and is not limited to the issue of polycentricity. The core elements of this framework include a geospatial pipeline, integrated spatial analysis methods, and a set of visual analytics tools. To validate the generic framework and implement the theoretical methodology into practice, a case study of Singapore is conducted. Based on the refined definition of Polycentricity, functional changes in Singapore are emphasized and detected from travel survey data and smart card data from multiple years. The latter data is a newly available large dataset generated by an automatic fare collection system. In particular, statistical analysis is performed to extract travel behaviours at the individual level; urban centrality is measured from aggregated urban activity patterns by a spatial convolution to identify the spatial structure of urban stocks; and a spa- tial network model is built as an example of analogy models to identify the spatial origination of urban flows. In these analyses, sets of urban indices of Polycentricity, such as density, entropy, and centrality, are defined, and their measures are bound to the proposed spatial analysis methods. By applying these measures to data from different years, the path of the functional changes in Singapore can be traced. By referring to a descriptive analysis of physical development in Singapore, the driving forces, impacts, successes, and anomalies of polycentric urban transformation can be identified. In sum, this work presents a quantitative approach to urban analysis that explicitly identifies ongoing urban transformation. Specifically, the impact of infrastructure development on people's lives and, in return, how cities are reshaped by individuals’ needs are examined using information extracted from mobility data. The urban studies in this dissertation represent a way to incorporate human behaviour into urban and transport design plans, thus leading to more livable cities. In a broader sense, it presents a systematic framework that facilitates geospatial techniques for impact assessment using big urban data in urban studies and planning.

Download activity - last month

Export as CSVJSONXML

Download activity - last 12 months

Export as CSVXMLJSON

Downloads by country - last 12 months

Export as XMLJSONCSV

Archive Staff Only

View Item