Modelling urban vibrancy with mobile phone and OpenStreetMap data

Exploring non-linear built environment effects on urban vibrancy under COVID-19: The case of Hong Kong

The coronavirus disease (COVID-19) pandemic has enormously changed the way people perceive and use urban spaces, exacerbating some pre-existing issues including urban vibrancy decline. This study aims to explore built environment effects on urban vibrancy under COVID-19, which will help recalibrate planning models and design principles. Based on multi-source geo-tagged big data of Hong Kong, this study reveals variations in urban vibrancy and employs machine learning modeling and interpretation methods to examine built environment effects on urban vibrancy before, during, and after the outbreak of COVID-19, with review volume of restaurants & food retailers as the indicator for urban vibrancy and built environment depicted from five dimensions (i.e., building form, street accessibility, public transport accessibility, functional density, and functional mixture). We found that (1) urban vibrancy concussively decreased during the outbreak and slowly recovered afterwards; (2) built environment's capability to stimulate urban vibrancy was weakened during the outbreak and restored afterwards; (3) the relationships between built environment and urban vibrancy were non-linear and moderated by the pandemic. This research enriches our understandings of the role of the pandemic in influencing urban vibrancy and its correlation with built environment, enlightening decision makers with nuanced criteria for pandemic-adaptive urban planning and design.

Do Vibrant Places Promote Active Living? Analyzing Local Vibrancy, Running Activity, and Real Estate Prices in Beijing

Although extensive research has investigated urban vibrancy as a critical indicator for spatial planning, urban design, and economic development, the unclear relationship between local vibrancy and active living needs to be clarified and requires more in-depth analysis. This study localizes urban vibrancy at both hyper-local and neighborhood scales by integrating high-resolution, large-scale, and heterogeneous urban datasets and analyzing interactions among variables representing vibrancy's environmental, economic, and social aspects. We utilize publicly available urban open data, Points of Interest requested from API, and leisure running trajectories acquired through data mining to investigate the spatial distribution of various vibrancy indicators and how they interact with physical activity at the local scale. Based on these variables, we then construct linear regression models and Geographically Weighted Regression (GWR) models to test and estimate how local vibrancy and physical activity relate to residential real estate characteristics. The results reveal the strong impact of urban form on local vibrancy but not physical activeness. At the neighborhood level, all vibrancy factors are statistically significant to local residential real estate prices but with different interactions based on location. Our study highlights the importance of accounting for locality and different physical, environmental, social, and economic factors when analyzing and interpreting urban vibrancy at a granular scale within a city.

How Did the Built Environment Affect Urban Vibrancy? A Big Data Approach to Post-Disaster Revitalization Assessment

Quantitative assessment of urban vibrancy is crucial to understanding urban development and promoting sustainability, especially for rapidly developing areas and regions that have experienced post-disaster reconstruction. Taking Dujiangyan City, the hardest-hit area of the earthquake, as an example, this paper quantifies the urban economic, social, and cultural vibrancy after reconstruction by the use of multi-source data, and conducts a geographic visualization analysis. The purpose is to establish an evaluation framework for the relationship between the urban built environment elements and vibrancy in different dimensions, to evaluate the benefits of post-disaster restoration and reconstruction. The results show that the urban vibrancy reflected by classified big data can not be completely matched due to the difference in the data generation and collection process. The Criteria Importance Though Inter-criteria Correlation and entropy (CRITIC-entropy) method is used to construct a comprehensive model is a better representation of the urban vibrancy spatial characteristics. On a global scale, comprehensive vibrancy demonstrates high continuity and a bi-center structure. In the old town, the distribution of various urban vibrancies show diffusion characteristics, while those in the new district demonstrated a high degree of aggregation, and the comprehensive vibrancy is less sensitive to land-use mixture and more sensitive to residential land.

Recent advances in urban system science: Models and data

Cities are characterized by the presence of a dense population with a high potential for interactions between individuals of diverse backgrounds. They appear in parallel to the Neolithic revolution a few millennia ago. The advantages brought in terms of agglomeration for economy, innovation, social and cultural advancements have kept them as a major landmark in recent human history. There are many different aspects to study in urban systems from a scientific point of view, one can concentrate in demography and population evolution, mobility, economic output, land use and urban planning, home accessibility and real estate market, energy and water consumption, waste processing, health, education, integration of minorities, just to name a few. In the last decade, the introduction of communication and information technologies have enormously facilitated the collection of datasets on these and other questions, making possible a more quantitative approach to city science. All these topics have been addressed in many works in the literature, and we do not intend to offer here a systematic review. Instead, we will only provide a brief taste of some of these above-mentioned aspects, which could serve as an introduction to the collection ‘Cities as Complex Systems’. Such a non-systematic view will lead us to leave outside many relevant papers, and for this we must apologise.

Population estimation beyond counts-Inferring demographic characteristics

Mapping population distribution at a fine spatial scale is essential for urban studies and planning. Numerous studies, mainly supported by geospatial and statistical methods, have focused primarily on predicting population counts. However, estimating their socio-economic characteristics beyond population counts, such as average age, income, and gender ratio, remains unattended. We enhance traditional population estimation by predicting not only the number of residents in an area, but also their demographic characteristics: average age and the proportion of seniors. By implementing and comparing different machine learning techniques (Random Forest, Support Vector Machines, and Linear Regression) in administrative areas in Singapore, we investigate the use of point of interest (POI) and real estate data for this purpose. The developed regression model predicts the average age of residents in a neighbourhood with a mean error of about 1.5 years (the range of average resident age across Singaporean districts spans approx. 14 years). The results reveal that age patterns of residents can be predicted using real estate information rather than with amenities, which is in contrast to estimating population counts. Another contribution of our work in population estimation is the use of previously unexploited POI and real estate datasets for it, such as property transactions, year of construction, and flat types (number of rooms). Advancing the domain of population estimation, this study reveals the prospects of a small set of detailed and strong predictors that might have the potential of estimating other demographic characteristics such as income.

Rapid indicators of deprivation using grocery shopping data

Measuring socio-economic indicators is a crucial task for policy makers who need to develop and implement policies aimed at reducing inequalities and improving the quality of life. However, traditionally this is a time-consuming and expensive task, which therefore cannot be carried out with high temporal frequency. Here, we investigate whether secondary data generated from our grocery shopping habits can be used to generate rapid estimates of deprivation in the city of London in the UK. We show the existence of a relationship between our grocery shopping data and the deprivation of different areas in London, and how we can use grocery shopping data to generate quick estimates of deprivation, albeit with some limitations. Crucially, our estimates can be generated very rapidly with the data used in our analysis, thus opening up the opportunity of having early access to estimates of deprivation. Our findings provide further evidence that new data streams contain accurate information about our collective behaviour and the current state of our society.