Effects of urban mobility structure change on disease diffusion patterns in Singapore during the COVID-19 pandemic

Keywords: Urban mobility networks, community detection, complex network, disease diffusion, urban dynamic
Abstract Type: Virtual Paper Abstract

Abstract

The pandemic has changed mobility patterns and lifestyles in cities. During the outbreaks, urban mobility is restricted formally (lockdown policies) or informally (individual choices). This dynamic affects not only urban mobility patterns but also the structure of urban spatial interactions---the network structure of disease spreading. Therefore, understanding the spatial interaction structure and its dynamics is crucial for disease control. In this study, we use Singapore public transport ridership data to capture the urban mobility structure and explore how it has changed in the past three years. We focus on two concepts: (1) the expansion diffusion caused by the spatial effects (geographical proximity and distances) and (2) the relocation diffusion caused by the network effects (population movement through the public transportation network). We detect network communities to identify the invisible movement boundaries that emerge from the actual flow data. Then, we analyze the intra- and inter-communities flow structure that captures the possible disease-spreading routes for the expansion and relocation diffusions, respectively. The intra-communities flow indicates the internal spatial distribution of high flow density and high flow connections with various communities, vital for the understanding of local expansion diffusion. The inter-communities flow shows the relationships between different regions of the city, signifying the chance of movement of disease cases from place to place. Understanding the intra- and inter-communities network structure can provide a more holistic picture of the disease-spreading process. This framework can also be used for disease management strategies simulation and future mobility-related urban planning.