One week prior to the Singaporean government’s introduction of a slew of stay-at-home measures, SparkBeyond’s platform had already predicted which points-of-interest require distancing.
The actionable insights outlined which schools and community clubs to close; if banning religious gatherings is necessary; and determine when to ramp up sanitizing efforts in MRT stations. As the insights adapt to changing dynamics, this further increases the efficiency in resource deployment.
The insights on the top table (above) and policies on the bottom are color-coded so that they match accordingly. This shows how SparkBeyond can support policy-makers to formulate policies. On closer examination, some of the insights may have a large radii, for example the ones related to cinema and educational institutions, and it may seem too drastic to shut these places down. However, one can argue that if these places are not closed down, there will be a lot of surrounding footfall (students going to shopping malls, coffee shops, nearby bus stops). Thus, shutting them down makes sense. When we enforce blanket lockdowns, it safeguards places of interests and the surrounding areas.
In generating a dynamic heatmap of Singapore, SparkBeyond augmented local data with a wealth of external sources. This allowed us to forecast the correlation between community activities, social clubs, and religious gatherings, and the number of infected patients. The map below highlights infection clusters of communal and religious gatherings.
Based on the heatmap above, Singaporean authorities can deploy resources more efficiently to places with greater need. Note that all the outputs on this and the previous slides are based on analysis using limited public data. Moving forward, SparkBeyond is able to generate much deeper actionable insights if more granular and accurate data from the government is provided.