[1M1-OS-20a] 社会現象とAIと可視化(1/3)

[1M1-OS-20a-01] COVID-19 Infection Diffusion Analysis by Simulation Considering Human Behavior Change

〇Naoki Wakabayashi¹, Takumi Sugiura², Fumito Ihara², Ryoya Watanabe², Daiki Kishimoto², Takahiro Higasa², Satoshi Kurihara¹ (1. Faculty of Science and Technology, Keio University, 2. Graduate School of Science and Technology, Keio University)

Keywords:simulation, COVID-19, Multi Agent

Since COVID-19 was reported in 2019, the disease has spread to most of the countries and regions in the world. Hence, it is important to predict the number of infected and severely ill people and to estimate the effects of policies to control the spread of infection. Mathematical models are commonly used to make such predictions and estimates. However, statistical methods such as mathematical models have difficulty in predicting infection and estimating the effects of preventive measures that consider changes in human behavior. In this study, we developed a multi-agent simulation system in which agents move according to a power law in an environment, and infection spreads in a small-world manner through contact between agents. The simulation was designed so that the effects of preventive measures could be estimated, and it could reproduce the spread of infection in the past. We simulated prediction of the spread of COVID-19, assuming human flow and infectiousness of latest variant.