Benjamin Seibold is passionate about traffic.
That’s not to say the Temple professor enjoys the stop-and-go on his way to work. Rather, he makes it his work to combine his other passion, mathematics, with research to alleviate traffic.
In addition to teaching a graduate-level course on mathematical modeling, Seibold, professor of mathematics at the College of Science and Technology, has worked with academics and researchers around the world to find ways of comparing even the most congested traffic jams with natural phenomena.
“In the past three years, we’ve gained a lot of new results. But the general premise of what we’re trying to solve and find is still the same,” Seibold said.
Seibold’s research deals with “phantom traffic jams,” or traffic jams that occur for no specific situational reasons, such as an accident or a lane closure. When the amount of traffic on a road is sufficiently dense, small changes in vehicles’ accelerations can have a rippling effect and cause a phantom traffic jam. These jams relate to a certain phenomenon the team called “jamitons,” a term derived from the soliton, a type of wave that behaves similarly to a jamiton.
“We often call them ‘stop-and-go waves’ of traffic, where you drive, and suddenly you have to brake because the person in front of you brakes, and then as a consequence you force the person behind you to brake, and this then triggers a wave that goes backwards on the road, a wave of braking vehicles,” Seibold said.
By treating the flow of traffic like that of liquids and gases, one can relate it to mathematics and physics, Seibold said.
“The equations that describe gases when they are reacto
The initial research began when Seibold was at the Massachusetts Institute of Technology in collaboration with four professors who are now spread across the world. They discovered the connection between traffic waves and detonation waves, which is the type of wave produced by combusting gas, and published their findings.
In the past three years, Seibold has been researching ways to use mathematical models to predict travel times, something many global positioning systems fail to do accurately, he said.
“The problem with a typical GPS is that when you enter your intended destination, it takes the current traffic situation that is observed, and says how long it will take you to get there,” Seibold said. “If you want to know your travel time, you also need to know the evolution of the traffic for the next hour.”
Some mathematical models have been implemented that can predict future traffic, but these models are much simpler than what is needed. One such model, the “Mobile Millennium Project,” cannot predict phantom traffic jams or jamitons. Seibold is collaborating with Daniel Work, a former traffic engineer in San Francisco who is now a professor at the University of Illinois at Urbana-Champaign, to create more sophisticated models that incorporate phantom traffic jams and jamitons.
“The art of traffic modeling is to formulate models that are simple but still capture and reproduce phenomena that are observed in reality,” Seibold said.
“If someone could come up with a solution that could reduce this level of congestion by even 1 percent, this would have a tremendous economic impact,” Seibold said.
According to a recent study by the Texas Transportation Institute, the annual cost of congestion on the U.S. economy has risen to $120 billion.
“But these are just numbers,” Seibold warned. “The message is this: It’s a big deal.”
Joe Brandt can be reached at tuf15598@temple.edu.
Be the first to comment