Sustainability and urban resilience are high on the agenda today. It is often said that we “only” need more data to better understand how we can make our infrastructures, processes, and decision-making pathways fit for the future. At the same time, without technologies such as digital twins (DT) and artificial intelligence (AI), this data remains largely unused. But what exactly is a DT in an urban context, what defines it, and how does it concretely help us make cities more sustainable and resilient, especially with regard to disaster and crisis planning?
However, before we can discuss how DTs contribute to urban sustainability, resilience, , and disaster management, we need to clarify what is meant by the term and what is not. A DT of a city is a dynamic digital representation of all or part of the urban system. It combines the physical environment, urban processes and systems, as well as data from IoT devices and citizen feedback. Application‑specific functions such as machine learning and data analytics are integrated, while reality and the digital model are continuously synchronized. This enables simulations, forecasts, optimizations, and greater transparency of planning decisions. Crucially, a city DT also includes a social dimension. While in industry the production process is at the centre, the city twin in the end considers people with their diverse behaviors and vulnerabilities.
Embedded in a DT, AI refers to methods that enable computers to recognize patterns in data, derive decisions, or make predictions without every step being explicitly programmed. In the DT of a city, AI primarily takes on three roles:
- It continuously evaluates incoming sensor data and other sources (e.g. traffic, weather, energy consumption),
- detects patterns and anomalies (e.g. overloads, risks), and
- calculates scenarios and forecasts (e.g. flood progression, heat islands, infrastructure bottlenecks).
In this way, the DT evolves from a static model into a learning, predictive system that supports planning and operational decisions in everyday life and in crises.
From tools to shared urban infrastructure
While DTs, networked data infrastructures, and AI-based analytics can be regarded the central tools, the decisive question is how we organize them so that they actually create more sustainability and resilience in the city. A key aspect is not to conceive the development and operation of DTs as isolated silo solutions of individual departments, but as shared infrastructure of city administration, municipal utilities, and other stakeholders. This results in several approaches:
- Resource and cost efficiency: When city administration, municipal utilities, and other stakeholders jointly build the infrastructure for DTs, duplicate structures are avoided. This saves significant financial and human resources.
- Faster competence building: Through cooperation, the participating organizations can learn from one another, share experiences and best practices, and thus develop the technical and data competencies required for planning, operating, and using a DT.
- Synergies and better operations: Shared components can be used multiple times, simulations can be shared, and scenarios can be aligned between stakeholders. New possibilities for comparison and analysis emerge.
- Combining physical, governance, and social layers: DTs link the physical city (infrastructure, environment, real-time data) with a governance layer (planning logics, AI models, scenarios) and a social layer (vulnerability, inequalities, citizen feedback). This combination makes it possible to plan measures that are not only technically efficient but also socially just and resilient in the long term.
To create more resilience and sustainability, the technological foundation, of course, must be given: DTs should be conceived from the outset as open, extendable infrastructure. This includes clearly defined and long-term stable interfaces, data compatibility based on established, ideally open, standards, as well as a modular architecture that does not block future extensions. In this way, twin platforms can be gradually expanded to include additional data sources, functions, and partners – even where no concrete collaboration is yet foreseeable today.
A new way for disaster management
To unfold their full impact, these technologies must not be understood only as tools for planning, efficiency, or sustainability. They are also a strategic lever for proactive disaster management and, thus, the question whether extreme events become manageable disruptions – or prolonged crises for citizens, infrastructure operators, and city administrations. DTs offer the following opportunities:
- Early risk detection through DTs: Virtual city models make visible where disasters are particularly likely to occur, before anything happens. They show where critical vulnerabilities lie and where it is sensible to invest in risk reduction.
- AI-supported forecasts and early warning systems: The analysis of heterogeneous data sources improves pattern recognition for risks and anomalies and enables better extreme weather forecasts.
- Optimized crisis response and resource planning: A dynamic situation picture supports emergency services in acting resource-efficiently and with short reaction times through real-time monitoring.
- Building resilient infrastructure: The DT can serve as a virtual test laboratory where planned measures can be tested under various or even extreme conditions. In this way, robust infrastructures are built and downtime as well as reconstruction costs are reduced.
- Linking sustainability and climate adaptation: Resilience measures can be planned in such a way that they simultaneously reduce emissions and support long-term climate adaptation. Optimizing resource and energy consumption while linking sustainability objectives with risk management helps to make urban development more sustainable and resilient.
- Better collaboration through shared data platforms: Shared, up-to-date data creates a common situation picture for administration, operators, and emergency services and accelerates coordinated action.
For DTs and AI to have exactly this impact in practice, they must be deliberately understood and built as the backbone of disaster management. In view of increasing extreme events, their use is not a technological nice-to-have, but a central prerequisite for cities to remain capable of action: before crises, during the event – and in the rapid return to a robust normal operation.
And even further
When DTs and AI become the backbone of disaster management, the real question is no longer whether we use them, but how far we are willing to think them. After all, risks, infrastructures, and mobility patterns do not end at administrative boundaries. In the future, the focus will have to shift from individual twins to networked systems. These will consist of several municipal or regional twins that communicate via consistent interfaces and enable aggregated analyses at regional, national and supra national level. By sharing infrastructures, platforms, and models across jurisdictions, the tools themselves become more sustainable: not every city needs to build and operate its own full stack, and even computing‑intensive components such as AI can be pooled and used more efficiently. More resilience and sustainability will emerge when data and models do not end at city limits: cross-cutting risks – such as heat islands, flood dynamics, supply interruptions, or commuter flows – can be viewed as interconnected patterns and managed more effectively. Comparability between municipalities, shared scenarios, and coordinated measures will thereby become technically possible in the first place. For spatial development, this means that investments in climate adaptation, critical infrastructure, or disaster preparedness do not have to be optimized in isolation, but can be aligned with one another, with the aim of making regional systems as a whole more robust, resource-efficient, and sustainably oriented in the long term.
