Improving the Resilience of Socio-technical Urban Critical Infrastructures with Digital Twins: Challenges, Concepts, and Modeling

Abstract

The increasing number of crises, including natural disasters and military conflicts, underscores the importance of resilient critical infrastructures (CIs), especially for urban areas. However, current approaches for CI modeling, monitoring, and resilience assessment are lacking a holistic view of cities as complex, interconnected, and socio-technical systems. This paper explores the application of the Digital Twin (DT) concept as a promising tool to assess and improve the resilience of urban CIs in light of various hazards. DTs are virtual real-time representations of a physical system that can be used to perform real-time analysis, simulate what-if scenarios, and provide decision support, during crises and normal operations. To this end, we identify and discuss key challenges for the development of Urban Digital Twins (UDTs), including data management, technical and social modeling of CIs, integrated CI co-simulations, model validation, and resilience assessment. To address the complex nature of urban areas as systems-of-systems, we present overarching modeling concepts by considering CI interdependencies and socio-technical interactions, resulting in the concept of the Socio-technical Digital Twin (STDT). Beside incorporating agent-based modeling, we discuss the issue of demand synchronization and propose the concepts of model selection and model transfer to facilitate the modeling process for UDTs. Furthermore, a multi-layered modeling framework for interdependent urban CIs is presented, where the proposed concepts are integrated and an overview and discussion of the technical and social modeling of CIs is provided, with a particular focus on the power, water, and transportation domain. Finally, we deal with the quantitative resilience assessment for interconnected CIs and discuss ways of integrating these methodologies in DTs. Our approach frames CIs as socio-technical systems and integrates the human perspective into the modeling process and resilience assessment. The presented modeling framework can be used to simulate various scenarios for analyzing their consequences in advance and measuring resilience in a holistic context. Moreover, the proposed concepts and modeling approaches can support future developments towards UDTs for crisis management.