For instance, students in China must use software installed on their smartphones, which generates different codes for users’ health status by analysing information such as body temperature. One US university is testing its wastewater system to detect infection. Video content analytics software powered by deep learning and artificial intelligence is also a possible tool for detecting congested areas on campus.
In Singapore, a group of researchers at the National University of Singapore (NUS) and the Singapore Management University (SMU) launched the Singapore Spacer project in April 2020. This software system was first used on the NUS campus, and it works by capturing the Wi-Fi signals from mobile devices across campus, and mapping when and where people congregate. The system will next be deployed on the SMU and Nanyang Technological University (NTU) campuses.
By providing insights into where people are congregating, universities can learn very useful information about traffic chokepoints and the efficacy of social distancing policies. Essentially, it provides feedback that allows administrators to improve these safety measures.
This project was initiated by SMU Associate Professor of Information Systems Rajesh Balan and NUS Yong Loo Lin School of Medicine Professor Michael Chee, who wanted to mitigate the harsh social and economic impact of lockdowns through the use of data, and assembled a multi-disciplinary team to do so. Prof Balan drew on his previous research to supply the technology for this project.
With private companies including Aviation Virtual and ESRI providing mapping support, the system’s core features were ready in two weeks and implemented in phases over a month.
The Spacer system does not capture data in real time, Prof Balan explains. It has a two-hour latency, and that works because it is not meant to be an enforcement tool. “It is meant to show the impact of policy decisions, such as whether staggered timings for the usage of the canteen are resulting in less human traffic.”
It also does not compromise anyone’s privacy, he stresses. “The data collected is anonymised and aggregated. We cannot identify any specific person.”
Given the nature of the system, it works well for campuses, as the majority of students switch to the campus Wifi when they are on the grounds. Convention centres, office buildings, malls, and hospitals are some examples of spaces where a Wifi-based data collection approach will likely still be effective. To expand beyond these types of contexts, however, will require seeking permission to collect cellular data. And of course, the system would only work in populations where mobile devices are widely used.
Prof Balan has been awarded a grant under the National Research Foundation (NRF) Central Gap Fund, and plans to use the grant money to improve current prototypes with a view towards commercialising the underlying technology.
If used widely, the system is a potentially effective complement to the Trace Together contact-tracing platform, in terms of allowing for the general population to move about relatively freely while still maintaining crowd control measures that are informed by data.
With memories of complete lockdowns still fresh in the minds of many, there’s no doubt that most people would agree with Prof Balan’s assessment: “The alternative is extremely brutal. Nobody wants that.”