Simulations of COVID-19 spread by spatial agent-based model and ordinary differential equations

Abstract

The COVID-19 outbreak is currently the biggest public health issue in the world. In this paper, the epidemic spread is modelled via two structurally different approaches, a system of first-order ordinary differential equations (ODEs) and spatial agent-based model (ABM). Specific intervention strategies are introduced and the effectiveness of the strategies can be assessed by comparing the results with/without these strategies. The simulation results are qualitatively affected by different parameter settings of the ODEs-based model; hence precision of input parameters characterising the spread is of great importance. The implementation of spatial ABM brings novel features to the epidemics modelling: new states being easily incorporated; the parameter illustrating the moving willingness of people; and sub-models for hospital beds to reflect demands of medical resources. Our results suggest that the flexible characteristics of ABM render it a useful addition to the tool set of epidemics simulation models so as to figure out new effective strategies.