A Prediction Method for Viral Disease Outbreak Using a Multi-Agent Simulation Including Capacity Limitation for Isolation Wards and Stay-at-Home Orders

Abstract

The possibility of rapid spread of highly infectious viruses necessitates prompt response. Being able to predict the effectiveness of particular measures against spread is therefore desirable. Although prediction of the transmission and extent of infection using multi-agent simulation is being carried out to this end, conventional simulation models do not incorporate the concept of isolation wards. Since isolation wards play roles in the quarantine and treatment of patients, they constitute an important parameter that affects fatality rate and transmission of infection. Prediction, therefore, that takes this point into consideration is advantageous. Another parameter that affects transmission is the extent by which people go outdoors. Particularly in Japan, a stay-at-home order was issued across wide areas around the country as a measure against COVID-19, which spread rapidly in Japan around April 2020. It is also important to be able to verify the effectives of these kinds of measures before they are implemented. Therefore, in this study, we developed a multi-agent simulation model for viral infection that incorporates the use of isolation wards and the implementation of stay-at-home orders. This simulator will enable the verification of the effectiveness of isolation wards and staying indoors in accordance with the infectiousness of the virus and the scale of the artificial society. This paper discusses several simulation patterns and their effectiveness.