Synthesized individual attributes considering city blocks for real-scale agent-simulation

2021 Fiscal Year Final Research Report

• Project/Area Number: 19K23229
• Research Category: Grant-in-Aid for Research Activity Start-up
• Allocation Type: Multi-year Fund
• Review Section: 0107:Economics, business administration, and related fields
• Research Institution: Aoyama Gakuin University
• Principal Investigator: HARADA Takuya 青山学院大学, 理工学部, 助教 (70847201)
• Project Period (FY): 2019-08-30 – 2022-03-31
• Keywords: 合成人口 / リアルスケール社会シミュレーション / データベース / 深層学習 / 離散最適化

Outline of Final Research Achievements

In this research project, we synthesize an artificial society that considers city blocks. For this purpose, we synthesized individual attributes across Japan consistent with the city block-level statistics.
We also assigned the synthesized households to buildings within the city block. The experimental results show that we could estimate the use of houses and apartment buildings with 95.2% and 72.9% accuracy, respectively. In addition, a comparison of the proposed method and the conventional method for Fushimi Ward, Kyoto City, Kyoto Prefecture showed that the proposed method succeeded in creating personal attributes with 5% improved accuracy.

Free Research Field

社会システム

Academic Significance and Societal Importance of the Research Achievements

本研究課題により生成された個人属性データベースは他の研究者への提供を予定している．個人属性データの提供により，他の研究者が感染症や経済政策など，様々な社会問題をシミュレーションする際に，シミュレーション内に生成する個々人の属性の設定に時間をかける必要がなく，シミュレーション設計や分析サイクルを早めることができる．これにより，急を要する社会問題への対策を迅速に分析することができる．