Friday, 29 October 2010

Unexplained Traffic Jam: Immaculate Congestion

On a Sunday sometime ago while reading the newspaper, I stumbled upon the "New Words of the Week". This section offers readers a small compendium of neologisms with a brief description of meaning. The first entry on Sunday was "immaculate congestion", obviously a pun on the phrase "immaculate conception"; "congestion" referring to traffic jams and "immaculate" meaning for no apparent reason. In summary, the author of the item referred to a phenomenon that I myself have seen: I'm in my car; I'm driving normally then for apparent reason, traffic becomes congested. I must slow down as we may now be bumper to bumper but then just as suddenly, traffic speeds up and I'm back to driving normally. However, I have not seen an accident; I have seen nothing which would explain the congestion.

I directed my attention to the Google search engine looking for the right keywords to find information about traffic jams that occur for no apparent reason. Curiously enough, there are quite a few people who have already talked about that phenomenon. Ha! As the saying goes: "There is nothing new under the sun". In fact, a group of Japanese researchers have created an experiment where they managed to duplicate the phenomenon. I refer to the original study paper published in the New Journal of Physics, Volume 10, number 3, dated March 2008.

•    The test involved a circular track with a circumference of 230 m and 22 vehicles.
•    The test was filmed: a panoramic view and a 360 degree overhead shot.
•    Cars were tuned to cruise at 30 km/h.
•    Drivers were instructed to follow the vehicle ahead in safety while trying to maintain their cruising speed.

At first things proceeded normally but after some time fluctuations in the cars began to appear: slight changes in speed, changes in distance between vehicles. These became more pronounced until a point was reached where cars were so bunched up; some vehicles were obliged to come to a full stop. The scientists observed that this "stop-and-go wave" propagated in the opposite direction to the movement of the vehicles. The video of this experiment shows how well the action of a driver can have a cascading effect in traffic and how the effect can last for some time.

The test empirically showed that a traffic jam is not always caused by a bottleneck: no accident, no obstacle or no lane closures due to construction. The average vehicle density exceeds some critical value and as a consequence the "free flow" state is unstable. A minor fluctuation can be caused by somebody hitting their brakes and that fluctuation propagates and a jam can ensue. Obviously a bottleneck such as a lane closure can increase vehicle density and thus increase the likelihood of a fluctuation causing a problem.

By reading this study, I wondered what I could do myself to avoid a traffic jam, that is, avoid being part of one or avoid causing one. Not surprisingly, at least for me, it is advisable to simply leave enough space between my car and the one in front of me, not to follow too closely. When there is not enough space between cars, braking a car produces a chain reaction where each driver has to brake behind the first even harder to avoid a collision. That is the cause of the compression movement. With enough space between vehicles, there is less braking, less compression and the action of a driver has less consequence for all drivers who are behind him.

Of course, the hard part of leaving enough space in front of me between me and the next car is people cutting in. When the traffic gets thicker, people get more impatient and some just weave in and out of the various lanes trying to get ahead. I back away from the guy in front of me to leave some space and sure enough, somebody from the lane beside me cuts in. It is sometimes virtually impossible to leave a safe distance.

There is nothing new under the sun. By chance I came across this column in the Sunday newspaper and I discover all this information on a phenomenon that I have been observing - participating in? - for a long time. Curious.


New Journal of Physics - March 2008
Traffic jams without bottlenecks—experimental evidence for the physical mechanism of the formation of a jam

The original Japanese study

The 2 films of the experiment: panorama and overhead 360
These videos are not streamed so when you click, you have to wait for the complete video to download to your computer before you see it. Both are about 3 MB. Be patient.

YouTube: Shockwave traffic jam experiment
This is a streamed version of the panoramic view of the test. I couldn't find a streamed version of the 360 overhead view.


Site Map - William Quincy BelleFollow me on Twitter

No comments: