Beyond Visualisation of Big Data: Towards Dynamic Data-driven City Planning in Singapore

Abstract

Cities are complex systems shaped by numerous simultaneous dynamic processes, which either work in harmony and keep the system healthy and thriving or they can converge into conflict, resulting in anomalies and disruptions. With the development in new information, communication and other technologies, we can now gather, use and analyse the dynamic urban data alongside the conventional static or semi-static data to better understand these complex processes and their relationships and city’s everyday functioning. With the creation of this vast amount of data, ‘what, when, and how’ are the key questions that have become synonymous with its use. Recent studies (e.g., Kandt and Batty, 2021; Liu and Biljecki, 2022; Zhang et al., 2022) have focused on how to analyse and visualise relevant data for creating an effective response to the issues in an urban environment. However, the roadmap of which data to use for analysis is still debatable. Moreover, while recent attempts have been proposing smart new means for big urban data visualisation, they also fell short in data interpretation and planning guidance.

In response, this paper outlines part of the larger study done in Singapore to systematically approach the vast expanse of urban big data and develop an alert system for the city officials and agencies on the underlying anomalies or outliers in the everyday city functioning. The project developed a comprehensive framework “Data Cube” that harnesses on dynamic economic, societal, environmental, health and attitudinal data available in Singapore, including people movement and behaviour, use of public transport, driving behaviour, park use, economic spending, eating out, healthcare centre visits, attitudes, etc. The types of information gathered are of different temporal basis, thus representing the velocity of changes in the pulse of city areas of different granularities. Moreover, we developed a “DataCube-CityScan” platform, which harnesses on GIS, AI, data mining and isolation forest analyses to identify specific trends and anomalies/outliers in real-time and alert city officials to respond, monitor changes, plan their actions and maximise their resources timely. While the platform depends on data availability, accuracy and timely updating, by interlinking real-time analysis of static, semi-static and dynamic data, trend and outlier visualisations and supporting spatial and non-spatial information, it shows great capacity of guiding planning authorities’ decision-making processes, strategy- and policy-making. Through examples of outlier detection and further interpretation, we will demonstrate the full capacities of this novel tool and its implications on city planning and management, beyond the traditional static indices, such as density and intensity indexes and land use, among other conventionally used planning indicators.

References Kandt, J. and Batty, M. (2021) “Smart cities, big data and urban policy: Towards urban analytics for the long run,” Cities, 109, p. 102992. Available at https://doi.org/10.1016/j.cities.2020.102992. Liu, P. and Biljecki, F. (2022) “A review of spatially-explicit GeoAI applications in Urban Geography,” International Journal of Applied Earth Observation and Geoinformation, 112, p. 102936. Available at https://doi.org/10.1016/j.jag.2022.102936. Zhang, D. et al. (2022) “Big data analytics, resource orchestration, and digital sustainability: A case study of smart city development,” Government Information Quarterly, 39(1), p. 101626. Available at: https://doi.org/10.1016/j.giq.2021.101626.

Keywords: dynamic urban data; big data visualisation; city pulse; city planning; Singapore