A Simulation of Covid-19 Transmission until the Steady State Using Graph Algorithm

Abstract

This research computes the average days to reach the steady state of COVID-19 infection by repeated simulation of human interactions using a bi-directed graph. It examines the effect of one infected person on a community comprising groups of people who interacts daily with each other such as school, commuter train, family, etc. Randomization is used to determine group structure and a bi-directed graph models the network of frequently interacting people. Once a person is infected, the incubation period is 5 days, and this person possibly infects other people for the next 6 days. A randomization distribution determines whether each person directly linked to an already infected person will be infected, or not. The effective reproduction number of the actual data is used in the study. Finally, this study examines how many days are required to reach the steady state where a new infected person is not observed any more. The study highlights number of days at the infections peak and how many days are required to reach the steady state. Future research would consider multiple starting points, variants, ages, genders, ethnicities, seasons, and regions. Furthermore, comparison with results of AI machine learning will be examined.