An epidemic simulation based on a multi-agent system distributed in a continuous space

2023 Fiscal Year Final Research Report

• Project/Area Number: 21K12064
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 61040:Soft computing-related
• Research Institution: Soka University
• Principal Investigator: Unemi Tatsuo 創価大学, 理工学部, 教授 (50151915)
• Project Period (FY): 2021-04-01 – 2024-03-31
• Keywords: 感染シミュレーション / マルチエージェントモデル / 並列分散処理 / シナリオ分析 / 新型コロナウイルス感染症

Outline of Final Research Achievements

A multi-agent simulator for infection dynamics assuming COVID-19 was developed and implemented to reproduce the time evolution of epidemic, as well as tools for statistical analysis and visualisation. It is also equipped with setups of possible scenarios in the subsequent situations. The model is characterised by the fact that it simulates the dynamics of the epidemic by moving one agent per population in a two-dimensional space based on a behavioural model, and by the aggregate results of individual characteristics, such as the excess frequency of contact and the degree to which population immunity is achieved. A reduction of the computation time was achieved by parallel processing to withstand a simulation of one million individuals, and at the same time, functions were implemented to facilitate the setting of various parameters.

Free Research Field

知能情報処理

Academic Significance and Societal Importance of the Research Achievements

2020年から23年初頭にかけて世界中で流行した新型コロナウィルス感染症 (COVID-19) は、過去に人類が経験した感染流行とは異なる特徴があり、対策の策定も困難を極めた。本研究で開発実装したシミュレータは、内閣府の対策室が組織したシミュレーションチームでも利用され、緊急事態宣言の発出規模と時期やワクチン接種戦略の決定にも重要な参考資料を提出してきた。シミュレータ開発の経験は、つぎに再び起きる可能性のある新たな感染症への速やかな対応のための構えの整備にすいても示唆を与えるものとなった。