2018 Fiscal Year Final Research Report
Statistical validation methods of evacuation guidance system as an example of social simulation assessments
Project/Area Number |
16K01291
|
Research Category |
Grant-in-Aid for Scientific Research (C)
|
Allocation Type | Multi-year Fund |
Section | 一般 |
Research Field |
Social systems engineering/Safety system
|
Research Institution | Meijo University |
Principal Investigator |
|
Project Period (FY) |
2016-04-01 – 2019-03-31
|
Keywords | 群衆シミュレーション / 社会シミュレーション / 妥当性評価 / 減災 / リスクマネジメント / 避難シミュレーション |
Outline of Final Research Achievements |
In the context of major disasters, development of prevention plans for addressing unforeseen situations and incidents is necessary. Agent-based simulation is employed for studying prevention plans that incorporate advanced information technology. Although simulations can be conducted on the basis of the behavior of an agent (civilian) and information exchange among agents, certain issues are inherent in social systems; inadequate data for verifying simulation results, difficulty to conduct experiments for validating the simulations. In this study, by considering evacuation guidance with new equipment as an example of a social system, three evaluation criteria are proposed for simulation results, namely consistency, awareness, and accountability. Additionally, in this study, simulation results are examined through statistical methods, macro level indices such as escape time and rate, and micro level agents’ behaviors.
|
Free Research Field |
人工知能
|
Academic Significance and Societal Importance of the Research Achievements |
災害時に人々を安全な場所に迅速に誘導する必要性については言うまでもない.9月11日の世界貿易センタービルや東日本大震災に,被災者がとった避難行動に関する調査が公開されている.その様な過去のデータや実験室レベル・シミュレーションの結果を基に,避難誘導の仕方によって避難行動が変化する報告があるがその効果の評価方法に統一した方法はない. 実証データの少ない社会シミュレーション結果の妥当性,有効性を,過去の知見と統計処理の両面から評価する事は社会シミュレーションの適用範囲が広がる中で重要になっている.又,減災の点からも,状況に合わせた効果的な避難誘導方法を示す事は有益である.
|