Use Cases
The use cases are submitted via a form that is designed to inform this MIM. If you want to share your use case, submit it here: https://forms.gle/h4VNLLRQSEG1wqgT9
Brussels, Belgium
LDT Purpose:
Anything, with a current focus on mobility
City-wide/application-specific:
Both
Chosen Category:
Experimental LDT: It allows to run what-if simulations and scenario management, therefore it requires more structured data. Resulting models will have to be configured and deployed once and will allow for sufficient verification of the validity of their outcomes. This type of LDT requires a limited involvement of data and domain experts.
List of capabilities:
Data Platform that integrates different data sources
Real-time data sources
Models / Simulations (ML, AI)
Visualisation: 3D
Interoperability Issue:
Legal aspects in contracts for data sharing agreement
Cartagena, Spain
LDT Purpose:
Mobility, Low Emission Zones
City-wide/application-specific:
Chosen Category:
Predictive LDT: It has capabilities to anticipate events such as traffic congestion or flooding and allows to analyse their impacts. It requires near-real-time data flows and vetted models, which may include AI. This type of LDT allows for the possibility of virtualizing disaster exercises using replay of historic data. It will require some expertise such as domain experts and data engineers to be available and needs a mature-data organisation.
List of capabilities:
Data Platform that integrates different data sources
Real-time data sources
Models / Simulations (ML, AI)
Citizen Participation
Visualisation: 3D
Visualisation: Interactive
Visualisation: Dashboards
Interoperability Issue:
The CHIMERE air quality model, used to simulate and predict the atmospheric dispersion of pollutants based on emissions, weather, and topography.
Espo, South Finland
LDT Purpose:
Additional tool for urban design and development
City-wide/application-specific:
For Built environment in Espoo
Chosen Category:
Experimental LDT: It allows to run what-if simulations and scenario management, therefore it requires more structured data. Resulting models will have to be configured and deployed once and will allow for sufficient verification of the validity of their outcomes. This type of LDT requires a limited involvement of data and domain experts.
List of capabilities:
Data Platform that integrates different data sources
Real-time data sources
Models / Simulations (ML, AI)
Citizen Participation
Visualisation: 3D
Visualisation: Interactive
Interoperability Issue:
Exploring interoperability within the city but also with external stakeholders
Experimental LDT: It allows to run what-if simulations and scenario management, therefore it requires more structured data. Resulting models will have to be configured and deployed once and will allow for sufficient verification of the validity of their outcomes. This type of LDT requires a limited involvement of data and domain experts.
List of capabilities:
Data Platform that integrates different data sources
Real-time data sources
Models / Simulations (ML, AI)
Visualisation: 3D
Interoperability Issue:
Legal aspects in contracts for data sharing agreement
Leuven, Belgium
LDT Purpose:
Different: 1) Mobility, 2) Nature Based Solutions implementation (climate adaptive measures), 3) Cultural (to be)
City-wide/application-specific:
Application-specifc
Chosen Category:
Awareness LDT: It consists of an explorative visualisation with some layers, it can include sensors and sensor data. This type of LDT has limited data management capabilities and, accordingly, does not require much user expertise to operate it. It can be used for situational awareness in terms of traffic, logistics, environmental, demographic, health, etc. It is typically the start of the digital twin roadmap.
List of capabilities:
Data Platform that integrates different data sources
Real-time data sources
Models / Simulations (ML, AI)
Citizen Participation
Visualisation: 3D
Visualisation: Dashboards
Interoperability Issue:
different datasets needed in different models are not interoperable
Madrid, Spain
While the use cases are in Madrid, the existing solution may cover any type of community
LDT Purpose:
Prediction of energy consumption in buildings and security monitoring in roads maintenance, prevention of computing networks attacks
City-wide/application-specific:
Both
Chosen Category:
Predictive LDT: It has capabilities to anticipate events such as traffic congestion or flooding and allows to analyse their impacts. It requires near-real-time data flows and vetted models, which may include AI. This type of LDT allows for the possibility of virtualizing disaster exercises using replay of historic data. It will require some expertise such as domain experts and data engineers to be available and needs a mature-data organisation.
List of capabilities:
Real-time data sources
Models / Simulations (ML, AI)
Visualisation: 3D
Interoperability Issue:
common criteria to assess the required level of data quality
Sofia, Bulgaria
LDT Purpose:
Support data-driven decision making through implementation of several scenarious related to urban planning, air pollution, climate change (e.g. urban heat island and heat risk), walkability, pedestrian wind comfort, etc.
City-wide/application-specific:
City-wide
Chosen Category:
Predictive LDT: It has capabilities to anticipate events such as traffic congestion or flooding and allows to analyse their impacts. It requires near-real-time data flows and vetted models, which may include AI. This type of LDT allows for the possibility of virtualizing disaster exercises using replay of historic data. It will require some expertise such as domain experts and data engineers to be available and needs a mature-data organisation.
List of capabilities:
Data Platform that integrates different data sources
Real-time data sources
Models / Simulations (ML, AI)
Visualisation: 3D
Interoperability Issue:
Lack of data models and data standards on the data providers’ side, which requires additional effort for data management. Data silos and missing availability of APIs for data exchange at the municipal level (legacy systems and siloed infrastructures in the municipality are often not designed to interoperate). Different standards are used by distinct communities in principle. For instance, the FIWARE community employs Smart Data Models, while the geospatial community relies on OGC standards like CityGML. Even small terminological differences (e.g., “roadwork” vs. “street maintenance”) can lead to integration failures or misinterpretation unless a common vocabulary or semantic mediator is applied. This disparity requires mapping these standards to ensure seamless data interoperability. AI models require high-quality, semantically rich, and consistent datasets. Integrating fragmented, incomplete, or inconsistent municipal data degrades AI performance and trustworthiness. Beyond technical aspects, challenges also arise in organizational and cultural domains, such as establishing common rules, governance frameworks, and fostering a culture of data sharing and innovation within public administrations. Deploying and maintaining interoperable systems — especially those involving AI, semantic models, or real-time data — requires skilled personnel, who are often missing in public administration.
Valencia
LDT Purpose:
The Local Digital Twin (LDT) of Valencia is focused on improving services for citizens, enhancing decision-making, fostering innovation, and creating business opportunities for SMEs and companies. While the ultimate aim is to cover all municipal services, the following key areas are prioritised:
DANA recovery, supporting the city’s efforts to recover from past extreme weather events.
Emergency Plan, enabling better prevention and response to disasters.
Sustainability, advancing Valencia’s Green Capital initiatives.
Mobility, improving transportation and accessibility.
Urban Planning, supporting smarter, more sustainable development.
City-wide/application-specific:
City-wide
Chosen Category:
Awareness LDT: Most services are in this stage.
Experimental LDT: Some alerts and filters are configurated in a what if configuration.
Predictive LDT: It has capabilities to anticipate events such as traffic congestion or flooding and allows to analyse their impacts. It requires near-real-time data flows and vetted models, which may include AI. This type of LDT allows for the possibility of virtualizing disaster exercises using replay of historic data. It will require some expertise such as domain experts and data engineers to be available and needs a mature-data organisation. - use case of parking vacances
List of capabilities:
Data Platform that integrates different data sources
Real-time data sources
Models / Simulations (ML, AI)
Visualisation: 3D
Visualisation: Dashboards
Interoperability Issue:
Silos between different services: ex. Mobility.
Existing software (legacy).
How to implement LDT toolbox with the existing software.
Last updated