Resilient Society

Society-Technology-Policy Nexus（複合相互依存性モデリング）

It is obvious that the final objective to assess the resilience of infrastructure systems or social systems against disaster is to protect our lives from the threats and to enhance the chance for our survival. However, in almost all of the related studies, human lives have not been explicitly taken into consideration. Instead, they have focused only on the robustness/vulnerability or recovery efficiency of the lifeline infrastructure systems. In order to discuss social system's resiliency comprehensively including human lives, it is necessary to build a framework for such human-centered resilience assessment. For that purpose, we are now developing a model to describe multiple interdependencies lying behind social systems. This model consists of three social sub-systems, that is lifeline infrastructure systems, service systems and business activities, and our daily lives. This model offers a framework to capture and describe interdependencies within and between these three subsystems.  

インフラや経済活動の災害に対する脆弱性やレジリエンスの評価は，本来，人々の生命や生活を守ることを最終的な目的としているはずである．しかしながら，既存の関連研究では人々の生命，生活までをスコープに入れた研究はほとんど行われていない．包括的に社会システムのレジリエンスを議論するためには，本来第一に考慮すべき人々の生活に関する多様な視点を積極的に考慮することが重要である．本研究ではこのような人間中心の視点を社会システムのモデリングに加え，”ライフライン”－”サービス・経済活動”－”生活”といった異なるシステムの間に内在する複合的な相互依存性を整理，記述するモデルの開発とそれに基づく人間中心のレジリエンス評価方法の開発を行っている． 

• *Kanno T., Suzuki T., Koike S.,Furuta K.,2019, Human-centered Modeling Framework of Multiple Interdependency in Urban Systems for Simulation of Post-Disaster Recovery Processes, Cognition, Technology & Work,21(2):301-316, https://doi.org/10.1007/s10111-018-0510-2

• 菅野，2022，社会シミュレーションに向けたヒューマンモデリングとコンテキストモデリング，シミュレーション, 41(3): 131—135.

Human-Centered Resilience Assessment（人間中心のレジリエンス評価）

In the implementation of the model, it is necessary to consider both structures and processes in the social systems. We implemented system structure by network model and process by agent-based model. Lifeline networks are implemented as simple networks with nodes and links where the reachability of a node  in the graph from any source nodes determines the availability of the lifeline at that node. Service and company agents, citizens, and restoration agents move along the road and transportation networks based on their purpose. Each citizen agent has preferences and rates some specific services over others, which enables to assess individual-life resilience against disaster. Resilience triangles of the social systems are drawn from the simulation under the optimized recovery plan obtained by genetic algorithms (GA). The results present a simple outcome that it is necessary to consider such multiple interdependencies to enhance the resilience of social systems. The result also suggests that it is possible to improve social systems resilience not only by reinforcing critical  infrastructures, but also redesigning the activities of service and company agents. Besides, we are applying a co-evolutional method to optimize civil, industry, and lifeline resilience at the same time considering the complicated multiple interdependency.

本研究では，前述の複合的相互依存性モデルを，多層ネットワークモデルとエージェントベースモデルを用いて計算機上に実装し，様々な条件下での社会システムの復旧過程のシミュレーションとその過程におけるサービス・経済活動や人々の多様な生活の復旧のシミュレーションを行っている．また，遺伝的アルゴリズム（GA）を用いてインフラ被害からの最適復旧計画を求めインフラ，サービス・経済活動，人々の生活を考慮したレジリエンス評価や共進化アルゴリズムを用いた複数セクター・ステークホルダの意思決定最適化を試みている．

• *Ogawa S., Kanno T., and Furuta K.,2019,Co-Evolution of Disaster Resilience in Complex Social Systems, Proc.2019 Summer Simulation Conf:Article No.66

• Wakayama K., Kanno T., Kawase Y., Takahashi H., and Furuta K., 2020, A Comparison of the Post-disaster Recovery of Water Supply Systems Through GA Optimization and Heuristics, Proc. European Safety Reliability Conf. and Probabilistic Safety Assessment and Management Conf.: ??-??