We apply advanced computational techniques to model and optimize complex, multi-modal transport systems. From the AI perspective, our approach is multidisciplinary and we routinely apply multiple AI techniques -- including machine learning, automated planning or operations research -- to tackle the smart mobility challenges at hand.
- Route planning and optimization where we have worked on a variety of route planning problems, including intermodal route planning, routing for bicycles and more recently route and charging planning for electrical vehicles. Our specific focus is on multi-criteria route planning algorithms which are capable of finding the Pareto set of optimal solutions with regards to multiple optimization criteria (such as time or price).
- Vehicle allocation in large-scale mobility on-demand systems where we explore vehicle routing algorithms for large-scale ride pooling problems. Our focus is on exploring how traditional algorithms for dial-a-ride-problem can be adapted to better handle the specific structure of real-time, on-demand ride pooling problems.
- Dynamic pricing for mobility services where we explore how dynamic pricing can be used to improve the allocation of limited mobility resources, such as EV charging slots or seats on public transport services. Our focus is on providing market-based allocation techniques that can automatically adapt to the current and predicted supply and demand, and achieve consistently better utilization of the respective mobility service compared to fixed-price alternatives.
We are very open-minded about exploring other problems. Whenever the problem at hand involves modeling or optimizing the movement of people, vehicles or goods on transport networks, we are likely able to help. If you have such a problem, then drop us a message and we will be glad to discuss!