Time-delayed Geodemographic Scaling Algorithm for Fast COVID-19 Simulation

COVID-19 has brought a series of challenges to human society and safety. Here a fine-grained COVID-19 Agent-based model (ABM) is proposed to investigate the possible outcomes of different policy interventions and their influence on subpopulations. A simulation platform is built to mimic the individual social contacts in different activities as the primary virus transmission way. Also, the simulation platform utilises a novel scale extension algorithm to enhance the simulation size while keeping the granularity, allowing the simulation size to range from a community with 30,000 residents to a megacity with 20 million residents. Finally, the model takes the recent Shenzhen outbreak as a case study to validate the simulation platform. The simulation results are consistent with the reported data, which lays the foundations to support different simulation scenarios to answer what-if questions efficiently.