Ig Nobel Prize for TU/e research into crowd movement

September 10, 2021

Recognising pedestrian patterns might become essential in predicting and managing the active flow of pedestrian crowds in the future.

Professor Federico Toschi (right) and university researcher Alessandro Corbetta with the Ig Nobel Prize.

Professor Federico Toschi and university researcher Alessandro Corbetta have won a so-called Ig Nobel prize with their analysis of the pedestrian movements of 5 million passengers in Eindhoven's train station. The scientists discovered that on average people keep a minimum distance of 75 centimetres from each other to avoid collisions.

The alternative Nobel Prizes were created for 'achievements that first make people laugh, and then make them think'. The prize, which has been awarded annually since 1991, has a cult status among scientists. In the past, the Dutch-British physicist Andre Geim even managed to win an Ig Nobel Prize before receiving a real Nobel Prize in Physics ten years later.

Most scientists are very honoured with the prize, as are Toschi and Corbetta. "We are very proud of this prize. This recognition is nice," says Corbetta about his research, the results of which were also published in the journal Physical Review E in 2018. The prize was presented by Nobel Prize winner Professor Martin Chalfie.

Avoiding contact 

For the research, Toschi and Corbetta installed four sensors under the platforms of Eindhoven railway station. For six months, they observed 5 million pedestrians within a measuring area of 27 square metres and discovered that, on average, people keep a minimum distance of 75 centimetres from each other. Pedestrians were found to be subconsciously constantly avoiding collisions with oncoming people by changing their footpath metres in advance if a collision was about to occur.

About 18,000 pedestrians were found to be facing each other in pairs. In other words; a potential collision hazard. Corbetta: "About 80 of these pedestrians actually collided with each other. The remaining pedestrians adjusted their paths until they were at least 140 centimetres away from each other and thus avoided a collision."

Statistical model

With this 'big data', the researchers developed a statistical model that can predict pedestrian movements very accurately. This way, you know in advance how many pedestrians will run, walk, dodge, turn around or collide in a defined area, such as a corridor or tunnel. Corbetta: "While building our model, we found two 'social forces': a long-range force based on sight, and a short-range force to prevent actual touch. These forces cause people to adjust their current walking path to avoid collisions."

Since its publication in 2018, both scientists have continued to develop their model and collect large scale data. Corbetta and Toschi's ultimate goal is to understand pedestrian movements.

"I dream of eventually understanding the dynamics of dense crowds," Toschi explains. Indeed, statistically speaking, there seem to be universal characteristics of pedestrian movements in a crowd, independent of the measurement location. Toschi: "For example, we see that about 1 person per 1000 people turned around and left the tunnel on the same side. Even if this person was alone and independent of motivation."

Managing crowds 

Recognising these patterns may be essential in the future in predicting and eventually managing the natural movement patterns of pedestrians. Toschi: "This way, we can design safer and more efficient places where many pedestrians come together, based on the natural walking behaviour of people. Think of museums, but also festivals. How do we prevent congestion? How can we spread the crowds as much as possible over an area? We try to manipulate the social system, as it were, but with good intentions."

Toschi: "For example, we have tried in the past to see if we can send people in a certain direction by means of light. In that respect, our research fits perfectly in the Ig Nobel picture; at first it is research that people might be surprised by, but soon they see that we can set important changes in motion with it."

Frans Raaijmakers
(Science Information Officer)