The SAP Labs India supported City-Scale Epidemic Simulator can help local authorities improve preparation for health, logistics, demand and supply situations
As India steadily transitions to more-relaxed lockdown measures, having an extensive understanding of how the Covid-19 virus might spread and estimating its future course is key to driving policy decisions for logistics, health, demand and supply circumstances. Towards this, SAP Labs India has built a non-pharmaceutical intervention—the ‘City-Scale Epidemic Simulator’—an agent base simulation model to estimate the number of possible infections and impact of non-pharmaceutical interventions on Covid-19 for India after the lockdown. The tool has been delivered as open source for the community and can be adapted or help in possibly avoiding a second wave of the pandemic or any outbreak.
To tackle the Covid-19 pandemic and to identify its impact, various scenarios can be simulated and results achieved in a short span of time. By using technologies such as Artificial Intelligence (AI) and Machine Learning (ML) it can factor in the numbers, to the populations and to the areas starting at the ward level and then going across the city.
The initiative was led by Rahul Lodhe, in collaboration with Indian Institute of Science, Bangalore (IISc) and Tata Institute of Fundamental Research, Mumbai (TIFR).
Sindhu Gangadharan, SVP & MD, SAP Labs India, says, “Ever since the Covid-19 outbreak, we at SAP Labs India had been working on solutions leveraging AI and ML to bring about realistic solutions to manage this pandemic situation and forecast the number of causalities. We collaborated with IISc and TIFR to support the research with SAP’s technology expertise to develop the tool within a few weeks.”
The simulator provides analysis from ward-level scenarios to city, district and state levels. The model estimates the change by calculating backwards from deaths observed over time to estimate transmission that occurred several weeks prior, allowing for the time lag between infection and death.
The simulation suggests how the disease may evolve once restrictions are lifted. It is different from any other forecasting mechanism as it is more detailed—the sample is differentiated by age,occupancy area and the mode of interaction in each ward or city. For example, based on parameters drawn from different parts of a city, the simulator would also be able to predict the number of hospital beds or ICUs required, besides drawing up a plan on tackling a surge in Covid-19 active cases with the city’s existing health infrastructure.
The tool will empower local authorities to improve preparation for health, logistics, demand and supply situation. It is targeted to help epidemiologists, war room analysts, healthcare sector and state government bodies.