City buildings are waking up: Slovak AI gives unused spaces a second chance

Cities are living organisms that are constantly changing. However, many of us pass by silent testimonies of the past in our neighborhoods every day – empty schools, unused offices or dilapidated public buildings. We often ask ourselves: “Why is it closed?” “Couldn’t there be a daycare center, kindergarten or cultural center here instead?”

Finding the right function for such a facility is not just a matter of idea. It is a complex urban planning puzzle. This is exactly what the Slovak team from the Creative Industry Košice (CIKE) organization has set out to solve, developing the NextUseAI tool as part of the SAM-SUD (Smart Asset Management – ​​Sustainable Urban Development) project funded by the European Union. Their goal is to create an intelligent system that tells cities what is suitable for a given location in terms of the organization of functions in the city, what cities prefer within their strategies and, above all, what their residents in a given location really need.

The Challenge: Millions of Micro-Localities in a Digital World

Planning a city for people means understanding space. The concept of the “15-minute city” says that everything important should be accessible by foot. But in order for artificial intelligence (AI) to be able to advise the city, it must first process a huge amount of data about every street, sidewalk and existing service.

The Slovak team worked with hundreds of gigabytes of data, which included map data from OpenStreetMap, digital relief maps and databases with thousands of civic amenities. The challenge was to transform this data into complex mathematical matrices of walking distances.

“In the initial phase of the project, we needed to quickly identify the weak points of our process in rapid iterations and quickly get to the results, understand them and re-adjust the input parameters. Already at this stage, we needed to process a significant amount of data in which the neural network could find significant patterns. This would have taken weeks for a regular computer. Without extreme performance, it would not have been possible to achieve the first meaningful results in such a short time,” says Róbert Pollák, head of the NextUseAI research team.

Solution: The power of the MeluXina supercomputer

The breakthrough came thanks to access to the European supercomputer MeluXina in Luxembourg, specifically to its part designed for artificial intelligence (AI Factory). Slovak experts were provided with powerful graphics accelerators that can process thousands of operations at once.

In this environment, the team built and tested advanced neural networks. They “learned” to recognize relationships between buildings, services and their surroundings in various cities. “The supercomputer allowed us to experiment with different settings and quickly correct errors, which would not be possible under normal conditions. Thanks to this, we were able to generate a method of creating spatial recommendations for the two largest Slovak cities – Bratislava and Košice – in a short time, in such a way that the models can learn from the spatial arrangement of one or more cities and make recommendations for another city.” adds Timotej Kendereš, CIKE data analyst responsible for working with the MeluXina supercomputer.

Results: Data at the service of people

The result is not just a dry table of numbers. The AI ​​model aims to suggest specific functions for unused spaces in the city to urban planners and strategists in order to improve civic amenities and pedestrian accessibility in neighborhoods. NextUseAI will then evaluate the spatial recommendations in the context of the needs of residents and city strategies, along with an explanatory report.

Although the results are currently in the experimental phase and serve to calibrate the entire system, they have shown an important thing: artificial intelligence can see connections that humans may miss. For example, the system can identify “dead spots” in a city where a specific service is missing and suggest its placement in an unused building nearby.

a used building nearby.

Impact and potential for the future

Next Use AI does not end in the laboratory. The ambition is to create a practical tool that will help city management make decisions based on data, not intuition.

For residents in the future, this means:

More efficient local government: Public money will be invested in buildings with a clear purpose and benefit.

Less time in the car: Services will be where people actually live.

A more beautiful environment: Abandoned buildings will get a new chance and will not fall into disrepair.

The use of European supercomputers thus brings a technological leap to Slovak urban planning. It shows that AI doesn’t have to be just an abstract concept, but a useful neighbor that will help us build cities where we can live better and healthier.