All insights

Engineering

Optimising City Traffic with Large-Scale Simulation, IoT and AI

After more than four years of research: how large-scale traffic simulations, combined with real-time IoT data and AI, can optimise traffic throughput across an entire city. The story behind the PhD that shaped how we think about AI in complex systems.

Engineering Nº 01

After more than four years of research, I am proud to present the results of my PhD thesis to the public.

The core question: how can we use large-scale traffic simulations, combined with real-time IoT data and the power of AI, to optimise traffic throughput across an entire city?

Simulating a full city is a brutal computational problem. Millions of interacting agents, real-time data streaming in from sensors, and decisions that have to be made fast enough to matter. The main focus of my research was exactly there: optimising the performance of distributed, multi-level simulations so that city-scale modelling becomes practically feasible.

Why does this matter beyond academia? Because cities are among the most complex systems we interact with daily, and traffic is where that complexity hurts most visibly: congestion, emissions, lost hours. If simulation plus live data plus AI can squeeze even a few percent more throughput from existing infrastructure, the payoff is enormous compared to pouring more asphalt.

The deeper lesson, which has shaped everything since: AI creates the most value when it is embedded in a system that deeply understands its domain, fed by real-time data, and engineered to perform at scale. That conviction is the foundation of everything we now build at Rollo.

We reply within 24 hours

Rather talk it through than read about it?

Tell us where you are with AI. We'll show you where it already creates value in your team.

Tavernierkaai 2, Antwerp. Or we come to you.