"Bad" drivers reduce traffic jams

[Free4Life11]

Very interesting!

Jerks actually reduce the risk of traffic jams

Tuesday, July 28, 2009

The next time someone cuts you off on your morning commute, don't be so quick to call the driver a jerk; you may have a reason to say thanks. According to the latest physics research, rule-breakers (drivers passing you on the wrong side or changing lanes too close to the intersection) actually help smooth the flow of traffic for the rest of us.

"The interesting finding is that if most of the people are law-abiding, and you have a certain amount of people who are breaking the rule, then you are actually getting the minimum chance of a [traffic] jam," said Petter Minnhagen, a physicist at Sweden's Umea University and an author of the paper published in the journal Physical Review E.

Physicists at the school uncovered this phenomenon while constructing a computer model of how a crowd of people move across a confined space, such as a pedestrian-only street. They divided the space into squares, like a chessboard, and randomly placed pedestrians in some of the squares. Like real people, the model pedestrians had a certain small probability of momentarily pausing, as if they had run into a friend or had bent down to tie a shoelace.

To make things more interesting, the researchers then tossed a few mavericks into the mix, who didn't follow the rules the other pedestrians used. The physicists ran the simulation over and over, each time boosting the percentage of rule-breakers. At first pedestrian deadlocks worsened. But as more and more rule-breakers joined the fray, something entirely unexpected occurred: traffic flowed best when only about 60 percent of pedestrians were obeying the rules.

Simple interactions of individual cars, people, or molecules add up to large patterns in a system. The high concentration of pedestrians in a small area increases the chances of a jam, but rule-breakers made the crowds spread out.

Morris Flynn, a University of Alberta professor who uses computational methods to study car traffic, agrees with the explanation. Because rule-breakers "carve out their own path," Flynn said, they dilute large concentrations of rule-abiders moving in the same way. He pointed out an example familiar to anyone who has driven on a two-lane road: breaking the speed limit to pass a slow vehicle prevents a long chain of cars from forming.

However, there is one rule you shouldn't break, according to a new analysis of how high-volume traffic flows along a highway. Cecile Appert-Rolland, a physicist at the University of Paris-Sud, looked at the tailing distances between cars traveling on a busy two-lane expressway in the suburbs of Paris. Most people have heard of the "three-second rule" for following distances; after the car ahead of you passes a point on the road, count to three. If you pass the same object before you get to three, you're following too closely. This time-based measure of the distance between cars is what Appert-Rolland terms the "time headway."

Her research showed that tailgating drivers were more likely than a non-tailgater to have a car in the lane next to them, so they weren't just speeding up in order to change lanes. She also found that these short time headways tended to extend across several vehicles, creating a platoon.

"We can identify at least seven or eight cars where they have time headways of half a second," she said. Considering that a driver's reaction time is about one second, these platoons are disastrous pileups waiting to happen. "If the first one brakes, the second one has to brake harder, the third one even harder, and the last wouldn't be able to brake hard enough."

So while unexpected behavior may help with congestion, always follow the three-second rule, because if you're tailgating, chances are you won't be the only one.

http://www.physicscentral.com/buzz/blog/index.cfm?postid=3414795237807494042

 

[sbell111]

Never be the fourth tailgater in a row. Got it.

 

[Eeyores Butterfly]

These are all computer simulations. While interesting, I have my doubts about how well they would work in the real world. The other night the bad drivers were definitely making the issue worse by going as far as they could and then forcing their way in (of course, this was due to an accident and a tree across the road). Not to mention that bad drivers are more likely to cause and be in accidents. I would rather have drivers follow the rules and risk getting in a jam, then not get home alive.

 

[sbell111]

... Not to mention that bad drivers are more likely to cause and be in accidents. ...

I wonder if this is actually true. Of course, we would first have to agree on what a 'bad driver' is.

 

[Eeyores Butterfly]

Cutting people off, excessive speeding, etc. sounds like what they are talking about. And statistics show that these types of behaviors are more likely to cause accidents. If you are tailgating and weaving in and out of traffic, it gives you far less reaction time should somebody need to hit the brakes. We know excessive speed is linked to higher accidents. Have you ever seen somebody who was cut off have to swerve to avoid the person? I have. In those cases they were lucky nobody was next to them or there would have been an accident. Most rear endings are caused by people following too closely to be safe.

 

[sbell111]

I have never seen someone who was 'excessively speeding' cut in front of someone and cause them to slam on their brakes. This is because someone who is excessively speeding is going faster than the person that he cut in front of.

I have seen a number of 'bad' behaviors cause accidents. However, I apparently don't define 'bad behaviors' the same as you do because I would have included such things as driving distracted, driving impared due to such things as alcohol or age, driving too slow, failure to merge smoothly, impeding traffic, failure to properly check a lane prior, etc.

Don't even get me started on those people who stop to merge early in an interstate merge lane rather than using the entire lane to match the speed of traffic.

BTW, the article didn't address 'bad drivers' it addressed 'rule breakers' and defined them as people who use behaviors such as passing on the right or changing lanes too close to the intersection. I think that most people could understand how these behaviors would actually be good for traffic flow.

If a slow person is in the left lane, passing on the right keeps everyone behind from getting backed up. Similarly, changing lanes as one approaches an intersection 'evens up' traffic, thereby causing smoother flow.

 

[AndyLL]

This study shows that sometimes smart people make stupid assumptions.

Their study was modeled in a square where the people could move in many different directions.

Therefor when a 'rule breaker' appeared it caused a ripple that spread in all directions... which could help break up groups.

Traffic only flows one direction. So the ripple usually only spreads backwards. That slows traffic and if the ripple is big enough there is a point in the wave where traffic completely stops.

The one example they used to prove it works for traffic ( speeding to pass a car) was invalid.

1. The rule breaker was not breaking up the pattern... they removed themselves from it.
2. The traffic is only improved if EVERY car breaks the rule and passes.
3. The same effect can be achived without breaking the rules in many ways.
4. The rule breaker is just as likely to cause disruptions in the patterns in the lane next to them or the oncoming lane.