The traffic jam mystery finally solved

By | December 20, 2007, 10:01am PST

Summary: When you’re caught in a traffic jam, you probably don’t know why. Is there an accident ahead of you? In most of the cases, the answer is no. And all the electronic devices installed in your car can’t help you. You’re stuck for a while — until the traffic gradually improves. Now, European mathematicians have solved the traffic jam mystery. The mathematical model they’ve developed shows that traffic jams are mostly caused by a single driver who brakes too much when faced to an unexpected event. Of course, the driver behind him also slows down, and so does the next one, until the road is totally blocked. But read more…

When you’re caught in a traffic jam, you probably don’t know why. Is there an accident ahead of you? In most of the cases, the answer is no. And all the electronic devices installed in your car can’t help you. You’re stuck for a while — until the traffic gradually improves. Now, European mathematicians have solved the traffic jam mystery. The mathematical model they’ve developed shows that traffic jams are mostly caused by a single driver who brakes too much when faced to an unexpected event. Of course, the driver behind him also slows down, and so does the next one, until the road is totally blocked. But read more…

This mathematical model has been developed by Dr Gábor Orosz of the Dynamical Systems & Control research institute of the School of Engineering, Computing and Mathematics at the University of Exeter, UK. Orosz, who also maintains a personal homepage, worked on this project with Gábor Stépán of the Department of Applied Mechanics at the Budapest University of Technology and Economics (BUTE) and who leads the research group on ‘Dynamics of Machines and Vehicles.’

Here is a description of what did the two mathematicians. “The team developed a mathematical model to show the impact of unexpected events such as a lorry pulling out of its lane on a dual carriageway. Their model revealed that slowing down below a critical speed when reacting to such an event, a driver would force the car behind to slow down further and the next car back to reduce its speed further still. The result of this is that several miles back, cars would finally grind to a halt, with drivers oblivious to the reason for their delay. The model predicts that this is a very typical scenario on a busy highway (above 15 vehicles per km). The jam moves backwards through the traffic creating a so-called ‘backward travelling wave’, which drivers may encounter many miles upstream, several minutes after it was triggered.”

And here is how the team thinks it’s possible to remedy to this situation. “‘When you tap your brake, the traffic may come to a full stand-still several miles behind you. It really matters how hard you brake — a slight braking from a driver who has identified a problem early will allow the traffic flow to remain smooth. Heavier braking, usually caused by a driver reacting late to a problem, can affect traffic flow for many miles,’ said Orosz. The research team now plans to develop a model for cars equipped with new electronic devices, which could cut down on over-braking as a result of slow reactions.”

For more information, this research work has been published in the Proceedings of the Royal Society A under the title “Subcritical Hopf bifurcations in a car-following model with reaction-time delay” (Volume 462, Number 2073, Pages 2643-2670, September 8, 2006). Here is a link to the abstract which starts like this. “A nonlinear car-following model of highway traffic is considered, which includes the reaction-time delay of drivers. Linear stability analysis shows that the uniform flow equilibrium of the system loses its stability via Hopf bifurcations and thus oscillations can appear. The stability and amplitudes of the oscillations are determined with the help of normal-form calculations of the Hopf bifurcation that also handles the essential translational symmetry of the system.”

Like certainly most of you, I’ve never heard about the Hopf bifurcations before today. But I found lots of details on this Wikipedia page.

You also might want to read the full paper from Orosz and Stépán (PDF format, 28 pages, 338 KB). But if I except the introduction and the conclusion, this is not something I’d recommend to anyone without a math degree.

Here is an excerpt from the conclusions of this paper. “Stop-and-go traffic jams can develop for large enough perturbations even when the desired uniform flow is linearly stable. These perturbations can be caused, for example, by a slower vehicle (such us a lorry) joining the inner lane flow for a short-time interval via changing lanes. It is essential to limit these unwanted events, for example, by introducing temporary regulations provided by overhead gantries. Still, if a backward travelling wave shows up without stoppings, it either dies out by itself or gets worse ending up as a persistent stop-and-go travelling wave. In order to dissolve this undesired situation, an appropriate control can be applied using temporary speed limits given by overhead gantries that can lead the traffic back ‘inside’ the unstable travelling wave and then to reach the desired uniform flow.” And be warned again: read the paper only if you’re good in maths!

Sources: University of Exeter, December 19, 2007; and various websites

You’ll find related stories by following the links below.

Kick off your day with ZDNet's daily e-mail newsletter. It's the freshest tech news and opinion, served hot. Get it.

Topics

Disclosure

Roland Piquepaille

http://blogs.zdnet.com/emergingtech/?page_id=566

Biography

Roland Piquepaille

Roland Piquepaille passed away in early January 2009. He lived in Paris, France, and spent most of his career in software, mainly for high performance computing and visualization companies, working for example for Cray Research and Silicon Graphics. He left the corporate world in 2001 after 33 years immersed into it. In 2002, he started a blog about technology trends and how they will affect our lives.

Related Discussions on TechRepublic

Did you know you can take part in these discussions with your ZDNet membership?
19
Comments

Join the conversation!

Just In

RE: The traffic jam mystery finally solved
meimeili 23rd Sep
Thanks for sharing happy imitation rolex watches
0 Votes
+ -
In Our Area
jallen@... 20th Dec 2007
In our area these slow downs from drivers braking to hard are directly related to the very short entrance and exit ramps on our interstates. Some of them are only 150 feet in length. It would really help if they continued the entrance ramp on down the road to the next exit ramp (usually about 1 mile) so that incoming and exiting vehicals had more time to merge into and out of the traffic flow. But what do I know, I'm not a DOT Traffic Engineer.
0 Votes
+ -
I suspect...
rapson 20th Dec 2007
...short entrance/exit ramps are hold-overs from the days when there were far fewer cars on the roads. It's definitely past time to re-evaluate road design for modern traffic loads.

OT, my pet peeve is: developing commercial and residential properties before improving the roads feeding them. Is there a theoretical model for that?

Carl Rapson
0 Votes
+ -
Thanks for sharing happy imitation rolex watches
0 Votes
+ -
I wonder if the same math can be used to prove that speeding helps traffic flow. happy

Seriously though, it can't be hard to prove mathematically that if all traffic were traveling the same speed and the same distance apart, the same effect will happen. That is, in order to maintain the same distance between cars, the traffic following a merging car must slow down. How many cars would it take to merge onto a highway before traffic stopped?
0 Votes
+ -
Another part of the problem
Linux User 147560 20th Dec 2007
is merging, or not merging correctly. I see people getting on the freeway at 30-40MPH... traffic is flowing at 70+MPH... anyone see a problem here?

Or when people are entering the freeway at speed @ssholes already on the freeway will speed up so as to prevent the merging traffic to slow or stop!

That's part of the problem. Another source of trouble are those with rubber necks. AKA rubber neckers. These morons cause a ton of grief because they will literally stop to look at something. And this is of course in agreement with the model provided.

As always Roland, excellent blog! devil
0 Votes
+ -
15 vehicle per kilometer?
kd5auq 20th Dec 2007
Try 150 vehicles per mile as in Houston! All the other
posts before this one are valid. I'm kinda glad I don't
live in L.A. (it's a wonderful city, but I couldn't cope
with the traffic).
0 Votes
+ -
Tailgating
Pony99CA 20th Dec 2007
I would think tailgating exacerbates this effect. Without a safe distance between you and the car ahead, if it brakes even a little, you have to hit your brakes harder. With a safe distance, you might only need a tap.

I was taught that you should never brake on a freeway unless you have to, and tailgating means that you have to brake more often.
0 Votes
+ -
re: Tailgating
techmail@... 5th Aug 2008
You are absolutely correct! Leaving proper distance between you and the car ahead means that you can effect a small speed change by simply taking your foot off the gas.

The "2 second" rule (staying 2 seconds behind the vehicle ahead of you) works at all speeds and provides the needed cushion for you to take evasive action in an emergency and it will smooth traffic flow.

However, it appears to be impossible for some people to see beyond the bumper of the vehicle ahead of them, even when their tailgating leaves them stopped because the car ahead is turning across traffic (left in US, right in UK/AU) and they don't have space to go around the stopped vehicle.

John
0 Votes
+ -
The Slinky Theory
slopoke 20th Dec 2007
This jibes well with what a friend of mine years ago termed the "slinky theory" of traffic flow. It states that once a stopage has occured the stopage itself moves as a wave function much like the bunching up of a slinky when you hold both ends and bounce it from one had to the other.

He also went on to point out that traffic congestion tends to have standing wave characteristics with stopage events at intervals and free flowing traffic in between. The stopages themselves move in the reverse direction to trafic as described in the article.

This was all done with "seat of the pants" observation and no fancy math involved.
0 Votes
+ -
This was a mystery?
dave.leigh@... 20th Dec 2007
Then again, there have been huge sums wasted on such things as "why do dogs have sex?" and "why does hot tomato sauce burn the roof of your mouth?" This "mystery" is in the same class of patently obvious non-mysteries, IMO.
0 Votes
+ -
kind of
mdemuth 21st Dec 2007
You can have a vague idea that gravity attracts things, or you can attempt to read Newton and Einstein.
I know for a fact that water forms drops, why should there be a study to find out why? It isn't like there is any mystery there. It just does it.
0 Votes
+ -
Even if the phenomena is obvious from just looking or watching at it, a mathematical model might help in automating the prediction or in helping avoid ocurrences of traffic jams (or whatever)
0 Votes
+ -
Computers Need Models
MichP 5th Aug 2008
To solve the problem, the overhead gantries that display temporary speed limits must be automatic, not run by some guy sitting in a booth. You can't tell the computer "just look at it."
0 Votes
+ -
All this money, work, spinning, calculating and BS to rediscover that traffic jams happen because some car(s) drive slower than the average? Or there are too many cars for the road?

What unmitigated nonsense!! Spin all you want, just NOT at my expense to justify the employment of a statistical bean counter playing with his toys! So socialist!!

Why not a billion dollar grant to statistical beanies to model the curve variations of tying shoe laces in order to understand the quantum variabilities of spacial constrains! That way, we could model the economics of properly designed shoe laces in order to maximize the efficiency and minimize the carbon footprint of oversized shoe laces as they interface with possibly undersized and harmful global warming impacting shoe eyelets!!

This is pure "wealth redistribution" from the producers to the "we're smarter than you" elitists of the worlds Marxist! Period!
0 Votes
+ -
Tailgating is the answer. Leaving a big gap means you hit the car in front much harder than if you are sitting tight. Think of a train. Few smashes, no braking required, and no backward waves. Problem solved happy
0 Votes
+ -
Tailgating??
jimonline@... Updated - 5th Aug 2008
You must be one of the idiots that try to rear-end me every time I get in my car (I happen to be one of the few drivers that follow the speed limit in the right lane). Tailgating is proven to be extremely dangerous since human response time is NOT fast enough to prevent accidents.

Equating driving to a train? The train cars are physically connected. When one stops, they all stop at the same time. The ONLY way to accomplish what you are suggesting would be to have your vehicle computer brake for you. So, if I need to slow down for merging traffic or an obstacle, your vehicle will automatically brake for you.

Otherwise, I can always use your insurance money...
0 Votes
+ -
Have you no humor? You reveal yourself.
0 Votes
+ -
RE: The traffic jam mystery finally solved
D14BL0 Updated - 6th Aug 2008
I can't believe this. I just blogged about traffic jams last night, and how pathetically people react to them, only to find out that people actually RESEARCHED something like this? It took me about two minutes sitting in the middle of a traffic jam to figure it out. No math required.

I swear, there is NO hope for society.
0 Votes
+ -
Length of a piece of string finally solved!!!
graham.lv Updated - 6th Aug 2008
"The length of a piece of string is as long as when you come to the end of it.." _(Theory proven Gra-e-mo)

---
I personally like the sticker on Chinese wheelbarrows that reads: "Not meant for highway use."

Join the conversation!

Formatting +
BB Codes - Note: HTML is not supported in forums
  • [b] Bold [/b]
  • [i] Italic [/i]
  • [u] Underline [/u]
  • [s] Strikethrough [/s]
  • [q] "Quote" [/q]
  • [ol][*] 1. Ordered List [/ol]
  • [ul][*] · Unordered List [/ul]
  • [pre] Preformat [/pre]
  • [quote] "Blockquote" [/quote]
ie8 fix

The best of ZDNet, delivered

ZDNet Newsletters

Get the best of ZDNet delivered straight to your inbox

Facebook Activity

White Papers, Webcasts, & Resources
ie8 fix