Via Autoblog, we get The Science Behind Traffic Jams from


Category: Travel, Weekend

Please use the comments to demonstrate your own ignorance, unfamiliarity with empirical data and lack of respect for scientific knowledge. Be sure to create straw men and argue against things I have neither said nor implied. If you could repeat previously discredited memes or steer the conversation into irrelevant, off topic discussions, it would be appreciated. Lastly, kindly forgo all civility in your discourse . . . you are, after all, anonymous.

9 Responses to “How Traffic Jams Happen”

  1. jmacdon says:

    I hate to be pedantic, but those are not theories. A theory is something that has been proved as well as anything can be in science. Think the theory of gravity or the theory of evolution. Both theories have stood the test of time, and we have yet to generate any meaningful data that disprove either (certain boards of education in Kansas notwithstanding).

    A hypothesis (which the average lay person often confuses with a theory), is an idea that appears reasonable that one might want to test with various experiments in order to prove or disprove. Only after surviving multiple such experiments can a hypothesis become a theory. The above are all hypotheses.

  2. blu says:

    Nice graphic, but it doesn’t really say anything about the science behind traffic, just some theories. If you want to know the science, I recommend Tom Vanderbilt’s book “Traffic”. One interesting thing that the book talks about is the strategies for merging when two lanes merge into one because of road work or something like that. There are “early” mergers, those who merge as soon as the sign says that a merge is needed, and “late” mergers, those who go as fast as possible for as long as possible and only merge when the roads actually merge. It turns out that late merging is the fairest and best strategy for everyone, allowing more cars to go through faster. If everyone adopted the strategy, it automatically makes everyone end up waiting an equal amount of time. The funny thing is early mergers think that they are doing everyone a service. I have even seen people straddle the line, to prevent late merging. They think they are forcing everyone to be fair, but it actually is worse for everyone when they do that.

  3. blu says:

    They actually are more than just a hypothesises. There is data to back up each. What is really the case is that they are all mechanisms that play a role to varying degrees in different circumstances.

  4. I have to disagree with the guy moving over causing all the others to slow down behind him. You are ignoring economics. If a guy moves over to add 1 to a lane he is also creating 1 empty space in the lane he has just left. Thus the lane he vacates gets the benefit of speeding up. Invariably, there will soon be someone from the guy’s new lane who will need to take his spot in the old lane thus rebalancing the equation. The chances of this increase directly proportionate to how fast his old lane speeds up. :)

    Where the problem lies is when everyone has to merge into one or two lanes and thus are all heading in that same general direction. I agree that the key issue is space and the need for it to be created.

    I have been trying this little experiment on my way home from work every day and the people who are buying into it are getting a faster merge with me. When you have two lines of cars that are forming one line, at some point two cars will each have to stop for a half car length in order to create a full car space in one line.

    A much easier solution is for everyone to create one car length between themselves and the next car in line as the line approaches the merge (when the line stops, you stop with one car length between yourself and the car in front). That way, when it comes time for you to merge you only need to slow slightly (or the front car needs to accelerate slightly) in order to create the space for the two cars to merge quite easily.

    Try it next time you are in a forced merge. Create a space between you and the front car and maintain it. When it comes time to merge, your merge will speed up. Once people around you get the pick up in speed, they will start leaving space also. The only caveat is line jumpers, but if you are in a section with no line jumpers it works really well. With jumpers around I keep it to half a car length and close in as they pass by

    OK, I think I just did my traffic geeking for the year

  5. @blu,

    The one who let the early mergers in are actually ‘paying’ double what everyone else is. First at the early merge and again at the late merge thus making themselves twice as late as they normally would be (a whole three seconds but it is the principle right?)

    We have a section on one of our roads when people will cross a line illegally to merge early and then people will zoom into their old spot thus allowing their line to get many more spots than the line letting them in. It can get very frustrating at times

  6. KJMClark says:

    One problem with the graphic is that just about everyone is using unsafe following distances. We also have research that suggests that opportunistic lane changing (done safely) helps traffic to flow better around bottlenecks.

    I haven’t read “Traffic” yet, but I’m betting the late/early merge problems are really due to people driving like jerks and unsafe following distances. Just about every problem on the road boils down to those. Usually researchers decide there’s nothing they can do about those and “initial condition” them away. The better solution would be to stick speeding/following distance/illegal passing-on-the-right cameras up and write copious tickets.

  7. ddl24 says:

    “the wave effect” is true and i think there is a video somewhere that shows it visually. it’s done in Japan and it’s a group of cars driving in circle…. In terms of late or early mergers, shouldn’t really matter. the key is to make every driver doing it in a consistent fashion, either all merge early or late; but in real life, it’s impossible. Keep safe distance, ie, a nice buffer zone is the key. but that zone will eventually will be neutralized by opportunistic drivers. more like a prisoners dilemma. No way to solve it until all car are driven by computers…..

  8. Thatguy says:

    No mention of slow drivers in the fast lane as a problem?!?!?!
    Slower traffic keep right and you will solve a considerable amount of the traffic problem. Small improvements in the queue yield logarithmic improvements in wait time. If you don’t have people obstructing a shared resource stranding roadway by driving slowly in the left lane, then you will have considerable improvements in traffic flow. If you are in anything but the 2 most right lanes and NOT passing, then you are a menace and you are largely responsible for the traffic.

  9. [...] Science Behind Traffic Jams via The Big Picture via Chart [...]