This course offers an introduction to urban digital twins, focusing on their theoretical foundations and practical implementations, with a specific emphasis on the EPFL campus. As cities become increasingly complex, the need for sophisticated simulation and AI-based modeling technologies is more critical than ever. Urban digital twins serve as integrative platforms that combine real-time data, historical context, and predictive modeling to support urban planning, infrastructure management, and environmental stewardship. AI-based approaches are now essential at every stage of the design, development, and operation of such models.

Throughout the course, students engage with the principles of creating digital replicas of urban environments that are dynamic, interactive, and capable of simulating real-world conditions. Topics include recent advances in data acquisition techniques, data management strategies, and the application of AI for design and predictive analysis.

The centerpiece of the course is a group project in which students progressively build a digital twin of the EPFL campus. This project is supported by case studies based on real data, covering domains such as energy, food, and mobility.

The first part of the course introduces the theoretical foundations and core methods of urban digital twins. Students explore the conceptual frameworks underlying these technologies and their role in modeling complex urban systems. Topics include data infrastructures such as GIS, point clouds, and 3D modeling, as well as sensor integration and the fundamentals of simulation and AI in urban contexts.

The second part focuses on practical applications and case studies. Specific sessions address the challenges of modeling infrastructure systems such as energy, mobility, waste, food and more. Ethical and governance issues related to the use of urban data and AI are also explored. This segment is enriched by a series of invited lectures, offering diverse perspectives from practitioners working on operational urban twins in various cities. These sessions provide students with exposure to advanced AI tools, real-world case studies, and reflections on policy, governance, and digital ethics.

The third part of the course is dedicated to project development and synthesis. Students refine their digital twin models through iterative feedback sessions, leading to a final presentation.

urban digital twins, real-time data, predictive modeling, AI-based simulation, geospatial analysis, infrastructure systems, scenario planning, data governance.

While prior knowledge of information technology, GIS, 3D modeling and AI is beneficial, it is not mandatory. The course is designed to accommodate students from various technical backgrounds, providing foundational training as needed.

• Explain the foundational concepts of urban digital twins and understand their role in contemporary urbanism
• Manage the planning and execution of a urban digital twin model, incorporating historical data and future projections
• Analyze the challenges and opportunities associated with implementing digital twins in urban settings.

• Plan and carry out activities in a way which makes optimal use of available time and other resources.
• Make an oral presentation.
• Respect the rules of the institution in which you are working.
• Use a work methodology appropriate to the task.
• Communicate effectively with professionals from other disciplines.
• Negotiate effectively within the group.
• Set objectives and design an action plan to reach those objectives.
• Assess progress against the plan, and adapt the plan as appropriate.

Lectures, invited lectures and group project

Active participation and group project

Midterm presentation (30%)

Final presentation (70%)

• https://go.epfl.ch/URB-410https://go.epfl.ch/URB-410