International Research at Fraunhofer ITWM

Research to Combat Phantom Traffic Jams: When a Single Vehicle Is Enough to Calm Traffic

Press Release / September 14, 2026

Why does a traffic jam suddenly form on a highway even though there is neither an accident nor roadwork? Prof. Dr. Benjamin Seibold from Temple University in Philadelphia is exploring this question here at Fraunhofer ITWM. As a visiting researcher, he is working with a team from the division »Mathematics for Vehicles, Systems and Production« on studies using the RODOS^® driving simulator. The focus is on how driver assistance systems and automated vehicles – as well as human drivers – can stabilize traffic flow and prevent so-called phantom traffic jams.

Previous analyses have already shown that, under certain conditions, even a single intelligently controlled vehicle can be enough to reduce traffic congestion and improve traffic flow for many other vehicles. The group is now investigating how human driving behavior, vehicle automation, and traffic flow interact.

In doing so, they combine mathematical models, computer simulations, real-world traffic experiments, and simulator studies to address a question that affects us almost every day: How can we reach our destinations more smoothly, safely, and sustainably? We interviewed Prof. Dr. Benjamin Seibold on this topic:

Benjamin, You’ve Been Studying Traffic Flow for Many Years. What Fascinates You About This Topic?

Traffic is a fascinating example of a complex system. Even small decisions made by individual drivers can have a major impact on many other road users. I’m particularly interested in how such effects can be described mathematically and how we can use this knowledge to make traffic safer and more efficient.

Many people are familiar with this phenomenon: Traffic suddenly comes to a standstill, even though there’s no accident and no roadwork. What’s behind these phantom traffic jams? What’s actually happening there?

Phantom traffic jams occur when traffic flow is unstable – that is, when small changes in speed build up. If one vehicle slows down slightly, the vehicles behind it brake more sharply, eventually causing traffic waves that propagate through the traffic and travel backward along the road. This happens even though there is no visible obstacle.

What exactly is a phantom traffic jam, and why does it happen in the first place? In a TED-Ed video, Benjamin Seibold explains the phenomenon in a clear and accessible way for a broad audience.

In a previous presentation at Fraunhofer ITWM, you compared phantom traffic jams to detonation waves. What makes this analogy so fascinating?

At first glance, traffic and physics seem to have little in common. In fact, however, both systems exhibit similar patterns in the propagation of waves. Such analogies help us apply new mathematical tools to traffic problems and better understand complex processes.

© Fraunhofer ITWM
Interview with Prof. Dr. Benjamin Seibold and Esther Packullat at the driving simulator.

Now let's go into more detail: What are you currently working on during your stay at Fraunhofer ITWM?

Together with researchers at Fraunhofer ITWM, I am investigating how human driving behavior affects traffic flow and, in particular, how human drivers interact with potential driver-assistance systems that can mitigate and prevent traffic congestion. We are particularly interested in situations where traffic jams develop from normal traffic flow.

What role does the RODOS^® driving simulator play in this?

RODOS^® makes it possible to study traffic situations under controlled conditions. Test subjects move through a realistic traffic environment and interact with simulated vehicles. This allows us to collect data that would be very difficult or impossible to obtain in real-world traffic. In my previous experiments with real vehicles, each scenario occurs exactly once. RODOS^®, on the other hand, allows us to obtain reliable statistics on human driving behavior.

As part of the study, participants drive on a virtual circular track. Why this particular scenario?

A circular route is particularly well-suited for observing the formation of phantom traffic jams. There are no intersections, traffic lights, or other disruptive factors. This allows us to focus entirely on the interactions between vehicles and better understand how traffic congestion develops.

In a well-known study, you showed that in such a situation, even a single controlled vehicle can be enough to stabilize the flow of traffic consisting of about 20 vehicles. That sounds almost unbelievable. How does that work?

Many people think of traffic as the sum of individual vehicles. In reality, traffic flow behaves like a complex dynamic system in which phenomena can arise that are unintended by individual drivers – such as phantom traffic jams. However, if a vehicle drives in a particularly smooth and anticipatory manner, it can dampen traffic waves before they spread through the traffic. In our experiments, we were able to show that even a single vehicle controlled in this way is sufficient to stabilize 20 other vehicles. We virtually reproduced this ring experiment in the RODOS^® driving simulator.

© Fraunhofer ITWM
Benjamin Seibold and Tim Nicolai at the driving simulator.

You've since applied the findings from the ring experiment to real highways. What were you able to demonstrate?

In project »CIRCLES« (Congestion Impacts Reduction via CAV-in-the-loop Lagrangian Energy Smoothing), we investigated whether the findings from theory, computer simulations, and controlled experiments could also be applied under real-world conditions. Our findings showed that automated speed control can, in principle, also work in real traffic. Such field trials are important for bridging the gap between research and practice.

Many people think that autonomous vehicles would automatically solve phantom traffic jams. Is that really the case?

Not necessarily. Our research shows that automation alone does not guarantee a stable flow of traffic. What matters most is how the vehicles are controlled and how they interact with one another. Even modern driver-assistance systems can exacerbate instabilities under certain conditions. That is why it is important to understand the underlying mechanisms in detail.

But how can just a few automated vehicles really make a difference?

Researchers have been studying this question for several years. Theoretical studies, computer simulations, and real-world experiments suggest that less than five percent of well-coordinated vehicles are sufficient to dampen traffic waves and make traffic flow significantly more smoothly. Furthermore, these control vehicles do not need to be fully autonomous; on the contrary, our research shows that traffic flows can be smoothed using existing technology.

What are you hoping to gain from the experiments at RODOS^®?

We would like to better understand the conditions under which traffic flow can be stabilized. We are particularly interested in the role that specific driver-assistance systems can play and how they interact with human drivers. This last aspect is especially crucial because modern traffic is a complex, dynamical system in which people interact with mechanical systems and automation in real time.

You've also been involved in large-scale field experiments. What makes these experiments so special?

They allow us to test insights from theory and computer simulation under real-world conditions. Researchers from various disciplines collaborate on that work, ranging from mathematics and engineering to industry and government agencies.

You earned your Ph.D. at the University of Kaiserslautern and are now conducting research in Philadelphia. How does it feel to be back in Kaiserslautern?

Kaiserslautern has had a profound influence on my academic career, particularly in terms of linking mathematics to industrial challenges – an area in which Kaiserslautern has always been a true pioneer. I began my studies in mathematics in Kaiserslautern in 1997, just as the Fraunhofer ITWM was becoming a Fraunhofer Institute, and I earned my Ph.D. in collaboration with the Fraunhofer ITWM.

These experiences were essential to my subsequent career at the Massachusetts Institute of Technology (MIT) and then at Temple University. It’s wonderful to return for a research project and actively collaborate with new colleagues who are working on similar questions. Such international collaborations are always immensely valuable to science. The fact that such a great project has come about in Kaiserslautern is, as we say in the U.S., »the icing on the cake«.

© Fraunhofer ITWM
Team at the driving simulator: René Reinhard, Sebastian Emmerich, Tim Nicolai, Vicky Holfeld, Benjamin Seibold, Jonathan Schneider

More Information About Research on Phantom Traffic Jams

A Single Vehicle Can Improve Traffic Condi-tions for Many People

In the so-called »Arizona Ring Road Experiment«, Benjamin Seibold’s research team demonstrated for the first time using real vehicles that even a single controlled vehicle can be enough to influence about 20 human-driven vehicles. The vehicle mitigated stop-and-go waves and ensured a more even flow of traffic.

The results were remarkable: Compared to traffic conditions with severe traffic jams, fuel consumption and CO₂ emissions dropped by about 40 percent. At the same time, the number of hard braking maneuvers decreased by a factor of 100, and nitrogen oxide (NOx) emissions fell by about 70 percent.

Video about the Arizona Ring Road Experiment

The Experiment That Made Phantom Traffic Jams Visible

This research is based on the famous »Sugiyama Ring Experiment« from 2008. It demonstrated conclusively for the first time that traffic jams can form even without accidents, construction zones, or lane changes. The collective driving behavior of road users alone is sufficient to trigger so-called phantom traffic jams.

Video about the Sugiyama Experiment

Do Autonomous Vehicles Solve the Problem Automatically?

Many people assume that fully autonomous transportation would automatically eliminate phantom traffic jams. However, Benjamin Seibold’s research shows that the reality is more complex.

In an experiment, seven identical vehicles equipped with adaptive cruise control drove in single file. Although the lead vehicle reduced its speed only slightly, the braking response intensified from vehicle to vehicle. The last vehicle even had to deactivate the driver-assistance system.

The takeaway: Automation alone is not enough. What matters is how vehicles are controlled and whether they actively dampen traffic waves rather than passing them on.

Video about the experiment

Press Release

International Research at Fraunhofer ITWM

Research to Combat Phantom Traffic Jams: When a Single Vehicle Is Enough to Calm Traffic

Benjamin, You’ve Been Studying Traffic Flow for Many Years. What Fascinates You About This Topic?

More Information

Many people are familiar with this phenomenon: Traffic suddenly comes to a standstill, even though there’s no accident and no roadwork. What’s behind these phantom traffic jams? What’s actually happening there?

Privacy warning

In a previous presentation at Fraunhofer ITWM, you compared phantom traffic jams to detonation waves. What makes this analogy so fascinating?

Now let's go into more detail: What are you currently working on during your stay at Fraunhofer ITWM?

What role does the RODOS^® driving simulator play in this?

As part of the study, participants drive on a virtual circular track. Why this particular scenario?

In a well-known study, you showed that in such a situation, even a single controlled vehicle can be enough to stabilize the flow of traffic consisting of about 20 vehicles. That sounds almost unbelievable. How does that work?

You've since applied the findings from the ring experiment to real highways. What were you able to demonstrate?

Many people think that autonomous vehicles would automatically solve phantom traffic jams. Is that really the case?

But how can just a few automated vehicles really make a difference?

What are you hoping to gain from the experiments at RODOS^®?

You've also been involved in large-scale field experiments. What makes these experiments so special?

You earned your Ph.D. at the University of Kaiserslautern and are now conducting research in Philadelphia. How does it feel to be back in Kaiserslautern?

More Information About Research on Phantom Traffic Jams

A Single Vehicle Can Improve Traffic Condi-tions for Many People

The Experiment That Made Phantom Traffic Jams Visible

Do Autonomous Vehicles Solve the Problem Automatically?

Contact Press / Media

Prof. Dr. Benjamin Seibold

Contact Press / Media

M.A. Esther Packullat

International Research at Fraunhofer ITWM

Research to Combat Phantom Traffic Jams: When a Single Vehicle Is Enough to Calm Traffic

Benjamin, You’ve Been Studying Traffic Flow for Many Years. What Fascinates You About This Topic?

Many people are familiar with this phenomenon: Traffic suddenly comes to a standstill, even though there’s no accident and no roadwork. What’s behind these phantom traffic jams? What’s actually happening there?

Privacy warning

In a previous presentation at Fraunhofer ITWM, you compared phantom traffic jams to detonation waves. What makes this analogy so fascinating?

Now let's go into more detail: What are you currently working on during your stay at Fraunhofer ITWM?

What role does the RODOS® driving simulator play in this?

As part of the study, participants drive on a virtual circular track. Why this particular scenario?

In a well-known study, you showed that in such a situation, even a single controlled vehicle can be enough to stabilize the flow of traffic consisting of about 20 vehicles. That sounds almost unbelievable. How does that work?

You've since applied the findings from the ring experiment to real highways. What were you able to demonstrate?

Many people think that autonomous vehicles would automatically solve phantom traffic jams. Is that really the case?

But how can just a few automated vehicles really make a difference?

What are you hoping to gain from the experiments at RODOS®?

You've also been involved in large-scale field experiments. What makes these experiments so special?

You earned your Ph.D. at the University of Kaiserslautern and are now conducting research in Philadelphia. How does it feel to be back in Kaiserslautern?

More Information About Research on Phantom Traffic Jams

A Single Vehicle Can Improve Traffic Condi-tions for Many People

The Experiment That Made Phantom Traffic Jams Visible

Do Autonomous Vehicles Solve the Problem Automatically?

What role does the RODOS^® driving simulator play in this?

What are you hoping to gain from the experiments at RODOS^®?