Smarter Road Maintenance for City of Ghent
At Growing pAI, we specialize in helping governments and organizations turn complex data into actionable intelligence. One of our most exciting collaborations to date is with the City of Ghent, where we’re using artificial intelligence to improve the timing of road maintenance — saving money, reducing disruption, and extending the lifespan of city infrastructure.
From ambition to action: a strategic AI roadmap
It all started with a simple, honest question from the City of Ghent: “We believe AI can help us, but we’re not sure how.”
Rather than jumping straight into implementation, we began with what we believe is always the right first step — an AI opportunity workshop. Together with city officials and stakeholders, we explored dozens of possible applications of artificial intelligence across departments and services.
From that workshop, we carefully selected three high-potential ideas to develop further. One of them stood out: the use of predictive AI to optimize the timing of road resurfacing.
The challenge: prevent costly road repairs before they escalate
Maintaining Ghent’s 1,403 kilometers of roads is a continuous and resource-intensive task. The challenge isn’t just knowing where repairs are needed — it’s knowing when to intervene in a way that’s both cost-effective and minimally disruptive.
If maintenance is done too late, the damage often extends beyond the top asphalt layer into the underlying structure — requiring a full and expensive reconstruction. But if done just in time, resurfacing the top layer alone can restore the road and avoid major costs. Our goal was to predict that optimal maintenance window using data and AI.
The true value of this project isn’t just in identifying where roads will degrade — it’s knowing when it’s still economically viable to act. By forecasting the point at which the surface layer begins to fail (but before the underlying structure is compromised), the city can plan smaller, preventive interventions — extending road life and drastically reducing the need for disruptive overhauls.
This tool allows us to schedule minor interventions earlier and more efficiently. In doing so, we can avoid costly reconstructions and minimize disruption.
Joris Vandenbroucke
Schepen, Gent
Building the Solution: Predictive Modeling Using 20 Years of Data
The City of Ghent has maintained detailed records on road quality over the past 20 years. We enriched that data with external sources, including:
- Public transportation data from De Lijn
- Road traffic intensity measurements
- Environmental and contextual data from the city itself
Using this rich dataset, we tested various AI modeling techniques — including random forests, gradient boosting machines, lineair models and neural networks — to predict how road conditions evolve over time. Our main focus was to forecast the road quality three years into the future, identifying the moment just before structural damage occurs.
After rigorous evaluation, we selected a the best model, to capture non-linear relationships between variables such as material type, traffic load, weather exposure, and previous maintenance history.
What’s Next?
The project is currently in its pilot phase, overseen by Ghent’s digital services department, District09. While the first models focus on car roads, the framework is built to scale — and can eventually be applied to bike paths, footpaths, and beyond.
We’re proud to be working quietly behind the scenes to turn long-held intuition into data-driven decisions — and we believe this is just one example of how public data, AI, and real collaboration can lead to smarter, more sustainable cities.
AI continues to evolve, and we see exciting opportunities to refine hybrid approaches—combining the best of deep learning and traditional modeling. By strategically integrating AI where it truly adds value, businesses and policymakers can make smarter, data-driven decisions.
At Growing pAI, we specialize in building AI solutions that make a real impact. Want to explore how AI can transform your environmental initiatives? Let’s talk: via mail axel@growingpai.com or phone +32 475 54 2216.
