Can everyone get to work on time when the unexpected happens on the roads?

Meet Rafał Kucharski, transport systems expert at Cracow University of Technology and TU Delft



“Traffic events, such as accidents, road closures and demand peaks happen daily on road networks. Using PTV software solutions, we managed to recreate and predict the rerouting of drivers after unexpected events. The model was successfully implemented and can be integrated in large-scale networks”

My daily struggle
My work is a bit like being a driver in a big city. Whenever I understand some part of the transport system and the traveler behavior, there is a new problem emerging due to new technology or data. And it seems I am never closer to fully understand the full picture. But that is also what motivates me.
The driver’s dilemma
Traffic events typically lead to problems such as oversaturation, congestion and delays. Drivers then realize the situation is atypical, and decide to change their path, to partially avoid the negative consequences of the event. These events change both travel times and traffic flows.
Forecasting how these events affect traffic is of huge importance. That’s because proper monitoring (to assess impacts) and management (to take actions) rely on predictions. But predictions are challenging. For example, it is especially difficult to predict how drivers are first informed of an expected event and can then react by changing their route.
PTV steps in
In our project, we aimed to improve prediction of traffic flows after unexpected events. We worked closely with PTV software suites, such as Visum and traffic assignment model PTV OPTIMA.
We managed to create an algorithm that reproduces the rerouting of drivers and the traffic evolution after unexpected events. Yes, even how drivers get informed about events and the probability of them changing path! The algorithm was successfully implemented and integrated with DTA applicable for large-scale networks.
So what’s next?
We are testing the results on the big networks, trying to understand how traffic evolves in cases of unexpected events. We also plan an optimization scheme for electronic traffic signs (VMS) to minimize delays due to traffic events. The algorithm will be applicable in practice for traffic managers using PTV Software.
From my experience, the PTV software is gradually expanding and meeting the objectives of academia and professionals. It is increasingly intuitive – not only at the GUI side, but also from its API – which is key for advanced users.
More about Rafal:
I am currently doing a PostDoc at TU Delft in the Netherlands, while on a sabbatical from Kraków University of Technology. I am part of a team modeling the new universe of MaaS, led by prof. Oded Cats. I worked with PTV at various levels of my career. I am especially interested in network science – not only traffic networks, but also airlines, mobility patterns, social networks and trade connections. In my free time I’m a full-time dad to my always curious young explorers.
Get to know the PTV expert:


Anett Ehlert, Senior Technical Product Manager
“As an experienced modeler, Rafal has contributed tremendously to the Visum community. So, it’s only natural he took on himself the challenge of solving the daily struggle of many drivers facing congestion and delays. We hope Rafal will continue to explore new areas and offer his advice for other Visum users. If you want to learn more about how cities can deal with road closures, accidents and other unexpected events using PTV software, watch this webinar by PTV expert Luca Paone”.

Discover other heroes championing efficiency:

Leave a Reply