Featured Useless Knowledge

Canned Oxygen Canned Oxygen Could be the Next Bottle Water

The emergence of another possible billion dollar business, based on bottling and selling a natural commodity.

Chaos and your everyday Traffic Jam

Posted by Failed Success on 04/10/06 at 06:08 AM

traffic jamCan the careless actions of a single person leave behind no physical trace, aside from a troublesome traffic jam?

The story is all too familiar. You’re on your way somewhere, when suddenly you find yourself stuck in traffic. It’s not a typical “rush hour” time of day and normally traffic does not get backed up like this. You figure there must be an accident or some type of serious incident up ahead just out of sight.

You slowly creep your way forward, bumper to bumper, continually looking for the flashing lights of ambulances and police cars; the telltale signs of a roadway incident. Then, all of a sudden, traffic begins to move normally again. There is no sign of an accident, incident, or any other cause of the slowdown in traffic. What happened?

People all over the world encounter situations very much like this one everyday. Usually, these situations result in a string of various profanities and harsh accusations aimed at fellow drivers’ mothers and driving capabilities.

Traffic JamWhat causes these mysterious traffic jams that continually appear throughout the day for no reason whatsoever? Is it simply the fact that most people just don’t have a clue how to drive? That’s very possible, and in reality there are so many variables involved in something like a traffic jam. But is it possible that the entire traffic jam could be both the continuing and end result of a chain reaction set in motion by a single driver who was in too much of a hurry?

Let’s examine one possible scenario:

It is a clear, sunny day and the roads contain no obvious hazards that would cause problems with traffic. Traffic on this particular highway is pretty thick, but it is flowing smoothly and steadily. One of the drivers, let’s say a man in a red car, decides that people in his lane are moving much too slowly for his taste. He quickly changes lanes in an attempt to get to a quicker moving lane. He fails to properly check his mirrors and cuts off another driver in the lane beside him. This forces that driver to apply his brakes to avoid getting clipped by the red car.

There is no collision, and the man in the red car continues on. The driver who was forced to use their brakes has now slowed a bit, causing the driver behind them to apply their brakes as well. In turn, the person behind them must hit their brakes as well. Human reaction time being what it is, each subsequent time that someone has to hit their brakes, it becomes increasingly more sudden; eventually leading drivers farther down the road having to slam on their brakes and come to a complete stop to avoid rear ending the person in front of them.

As a result of this standstill in that lane, drivers begin to change lanes to move to one of the other lanes that are moving properly. As this occurs, the other parallel lanes begin to slow as people are merging over. The same process repeats itself in the other lanes until the entire highway has slowed to a crawl and eventually a complete stop.

If the traffic on the highway remains dense and constant, this chain reaction could continue to travel back for miles and could last for an indefinite amount of time. The man in the red car who was in such a rush, probably to hurry home and read the latest article on FailedSuccess.com, may already be home now. He may have sat down on his chair, opened a cold can of beer, and began watching television; while hundreds of drivers are caught in the traffic jam that his actions created.

Butterfly EffectThis scenario was once referred to as the “ghost of a traffic jam” by Hannah Holmes, a columnist for Discovery Online, in a lighthearted article on the subject. An occurrence such as this is considered by many to be a plausible theory and a probable cause of many of the traffic jams we experience everyday.

The scenario is seated in a theory of Chaos and Non Linear Dynamics. It is similar to the principles of the “domino effect” and the “butterfly effect”. The “butterfly effect” leads to a conclusion that if a butterfly flaps its wings (a seagull was first used as an example for this condition and was changed to a butterfly to make it more poetic), that small disturbance in the chaotic motion of the atmosphere could create a chain reaction; amplifying the effect to that of a large atmospheric motion capable of altering the weather in another part of the world. The “butterfly effect” illustrates the impossibility of making predictions for complex systems.

This sensitive dependence on initial conditions is the essence of the Chaos Theory. Our scenario could also be loosely described in terms of a “domino effect”, but the “domino effect” relies on a linear series of identical events. The “butterfly effect”, however, seems to fit our scenario better as it insists that the effect amplifies the condition upon each iteration. In our example of the traffic jam, the effect was amplified each time as drivers were forced into an increasingly stronger, more abrupt action; resulting in a slowdown and eventual standstill.

Butterfly Effect ModelAll of this could also lead to additional amplified variables. Car accidents could have been caused by all of this action and reaction, leading to more traffic and events as emergency vehicles are rerouted to the scene of the accident. Some drivers may have taken a different exit than normal to avoid the traffic, leading to increased traffic and incidents on side streets; spreading the condition beyond the confines of the original highway. This would begin impacting traffic and people’s lives all over the city as more actions and reactions result from the original traffic jam. Each reaction led to another reaction creating an end result that was impossible to predict.

It’s possible that both theories have some place in describing a situation such as ours. It’s also entirely possible that neither can accurately describe the sequence of events. Many of the events may have occurred without the presence of our “man in the red car”. It’s also possible that if traffic hadn’t been following so closely behind the people forced to brake suddenly, the resulting traffic jam would never have happened. As much as we try to calculate and theorize “chaos”, it is still “chaos”. The long term effects and resulting behavior will become impossible to forecast.

So the next time you find yourself stuck in bumper to bumper traffic, it’s very possible that the jackass that caused it is already at home watching the latest episode of American Idol; while you sit in traffic with nothing to do but ponder the finer points of non linear dynamics.

Further Reading:

Learn more about the Butterfly Effect
Learn more about Chaos Theory
The Butterfly Effect on the Expressway by Kevin S. Forsyth
The Fluid Dynamics of Traffic Jams and Solutions to Solving Traffic

Discuss this Story

Display Name:

Submit the word you see below:

Previous Comments

  1. I live in the us, and there are a few simple rules people should follow, to wit: If you are going to drive the speed limit or slower, stay out of the left (or Passing) lane. If you are in the left lane, and someone driving faster than you comes up from behind, change lanes at the first safe opportunity and let them pass. This will help cut down on the weaving in and out of lanes, and really, is just common sense.

    Posted by bigbadjohn  on  09/23/09  at  10:29 AM
  2. Traffic jams are caused by:

    1. Distracted drivers(cellphone, grooming, rubbernecking, etc.),
    2. People riding their brakes,
    3. Drivers making unnecessary staggered movements,
    4. Drivers impeding traffic, by excessively slow driving,
    5. Highly indecisive drivers,
    6. Drivers intentionally holding up traffic,
    7. Drivers buying vehicles without the ability to drive within the
    potential of their vehicles,
    8. Drivers without locus of control, and
    9. In various situations, drivers in excessive waiting patterns.

    The aformentioned drivers, are key players and villians, involved in traffic being tied up.  Drivers need to learn, not only, how to operate their vehicles, also needed, is knowledge of how to drive their vehicles, with maximum efficiency and well executed driving methods.  In parking lots, people have a tendency to wait extenuous periods, for a parking space, driving the wrong way in parking decks, and drivers cause gridlock using onboard entertainment systems.

    Posted by Khalid  on  07/26/09  at  02:14 AM
  3. ppp;]p[;p;pp[plpl gfhffgfghggh jhhjhgjhgg hgffgfggfdhgfh gfhghhgfgfc dic ghgfkj

    Posted by wtf  on  02/13/09  at  11:14 AM
  4. sollution is to tax people when they brake,,,,,,,,,,let say ten pence,every time,,,willem 0044 7907046482,bulbgarden@googlemail.com,

    Posted by willem  on  01/18/09  at  09:10 AM
  5. Lots of the files from the Megaupload system at the global web are available for downloading at http://megauploadfiles.com/ without registration!

    Posted by freddy  on  01/13/09  at  07:10 AM
  6. To eliminate traffic jams, there are some actions that must be taken:
    1. Prohibit car circulation, unless very special exceptions.
    2. Educate people to walk short distances and ride bycicles for medium urban distances.
    3. Improve public transportation: buses, trains, undergrounds, etc.

    I assure you, after these actions are taken, traffic jams will just disappear!

    Posted by James  on  12/28/08  at  04:37 PM
  7. I DON’T GET ALL OF U ?????

    Posted by damn it!!!  on  11/05/08  at  10:21 PM
  8. Responding to comment #50.

    You are contradicting yourself.  If the optimum way to move traffic is to pack the cars as closely as possible, then it is very wasteful to have the passing lane practically empty.

    Posted by Ejay Hire  on  01/04/07  at  10:14 AM
  9. Slowing down and leaving large amounts of space in front of you EXACERBATES traffic jams, for the simple reason that traffic is caused by having too many cars on the same section of road at the same time.  Leaving 4 seconds of distance means you are simply taking away space from other cars and slowing everybody down behind you.

    If people would just keep out of the passing lanes unless they are actually in the act of passing, that would probably help more than anything else.

    Posted by Matt Hearn  on  01/03/07  at  11:54 AM
  10. The solution to traffic jams and traffic gridlock.
    Drive there and back without stopping?
    Look up http://www.ubtsc.com.au to find the solution.
    There are a couple of other solutions as well.

    Posted by jozef goj  on  12/30/06  at  10:52 PM
  11. #43 is the only one who gets it: cars decelerate more rapidly than they can accelerate.

    Any incident (crash, merge, lane change, etc.) will cause cars to slow down.  At high density this will lead to a situation where more cars are slowing down at a faster rate than cars are accelerating.  Thus they stack up behind each other in a queue.  The queue is waiting to reach the point of acceleration; this position gradually moves back in the line like a wave.

    Some variables are at play in every situation:
    1) Rate of slowdown
    2) Rate of acceleration
    3) Density of traffic
    4) Wavelength (length of queue)

    You can plot a curve in any traffic jam using those variables that would show a “saturation point”, where density is such that the rate of slowdown is greater than the rate of acceleration.

    As long as deceleration > acceleration, the queue (wavelength) will increase. 

    The problem is that increasing any variable will also increase the other variables.  It is the tendency of the system to grind to a halt.

    The comparison of the highway to a conveyor belt is inaccurate given that cars behave more like particles than a connected medium like a belt.

    A better analogy would be to compare cars to electrons traveling down a wire.  Both types of particles move independently, tend to be pushed along by other particles and have “force fields” that create natural distances between particles.

    Another way to say it is that roads = networks, cars = packets and intersections = routers.  There are striking similarities between both systems.

    The only answer guaranteed to work?  Do like Internet providers and create highways with 400 lanes!  Not practical of course, but there are techniques to be shared: some cities use reversible lanes in areas during rush hours, thus dynamically reallocating network resources to where it’s needed most.

    Too bad you can’t drop cars off the network like you can packets (oops, got dropped! start your commute home all over again...)

    Posted by Lives in Country; Rides Horse  on  12/30/06  at  12:45 PM
  12. traffic jams are caused by the physics of too many people on the road.  these events that cause them are basically nucleation of a solid phase from a fluid phase.  the fluid phase was unstable at such a high density and a small event triggered its nucleation into a solid one.

    some of the people commenting that are getting way too emotional about this are quite funny. 

    if you’re in one of those situations where the highway is packed but flowing you should just consider yourself lucky.

    also, avoid traveling when everyone else is.

    furthermore, the traffic jams around merges other poorly designed pieces of roadway also arise from the underlying physics of the situation.

    psychology does certainly play a role but this is not what’s causing the large crowd to move more slowly- that should be the physics as well.

    Posted by alex  on  12/29/06  at  03:16 PM
  13. I like the different speed limits for different lanes idea. Other ideas that might be worth a look:

    No Cell Phone Use in the Fast Lane.

    Special permit for the fast lane.

    Greater distances in signs for offramps or highway splits (which lane to be in for which highway up to 2 miles (4KM) before the split. (less last minute merging).

    Special, seperate lane (1/2 lane) for motorcycles. I’ve seen too many Auto VS Motorcycle accidents where lane changes are involved.

    Increase the enforcement of turn signals and encourage a ‘thank you’ blink of the hazard lights or a turn signal ‘answer’ of a flash of the headlights that it is OK to change lanes.

    Increase the use of lanes which can be used in either direction with traffic flow to add more lanes durring heavy commute times (usually a couple of the middle lanes). This allows more traffic flow in areas where parts of the highway are congested in a single direction in the morning, and the opposite direction in the evening.

    Take a closer look at merging traffic which causes congestion and examine alternate merge configurations which may work better.

    Meter lights on freeway onramps to stagger vehicles entering the freeway (allowing easier merging with vehicles already on the road).

    Heavy enforcement of people using merging lanes for cutting around heavy traffic and merging back in.

    Better public transportation systems

    stacked roadways to add more lanes without having to buy more land.

    Allow people to have dash-cams and record traffic offenders for traffic enforcement by the police for a cut of the fines generated.

    Posted by Hang up and Drive  on  12/28/06  at  03:51 PM
  14. I think bottle neck is one of the cause of traffic jam.
    Also people who are curious about accidents; they even turn their heads after they passed, with no car in front of them.  The most ridiculous issue for other drivers.

    Posted by Mark  on  12/28/06  at  03:33 PM
  15. Traffic jams are caused by slow moving people when they block all lanes while moving at the same slow speed instead of sharing the same lane. The solution to prevent traffic jams is - Different Speed Limits for each lane on the highway to prevent the above situation. Another alternate solution is - Establishing Minimum Speed Limit for just the fastest lane. People who spend less time on road are the least likely participants in a traffic jam, unless they are there because of slow people in the front.

    Posted by VM  on  12/27/06  at  11:39 PM
  16. Many people have commented that raising speed limits will prevent traffic jams. Unfortunately, it doesn’t do that. It makes them worse.
    For example, if a car has stopped in the middle of the road, if the traffic is going slowly (30 mi/hr) then let’s say that the density of traffic is such that one car reaches the blockage per second. Let’s also assume that people at the front of the blockage take one second to notice that the car in front has started, so that one car moves off from the front of the jam each second. If the speed limit is 30, the blockage stays one car long. Each second, a car is added to the end, and one moves off from the front.
    If the speed limit is 45 mi/hr, then 1.5 cars will arrive each second, while 1 will move off, so the traffic jam lengthens by 1 car every 2 seconds. Before long it extends for miles down the road. And if the speed limit is 60 mi/hr, the traffic jam lengthens by 1 car per second, and so grows twice as fast.
    Now, the disadvantage (longer jams) of a higher speed limit is offset by the advantage (faster travel times). But since jams cause stress, anger, and accidents, it may be better policy to lower speed limits on congested roads than to raise them

    (All of this only holds if the limit is followed, of course.)

    Posted by Anonymous  on  12/27/06  at  10:50 PM
  17. You who think maintaining a nice distance between cars and flowing around problem areas must live in the boondocks.  When all lanes are at capacity, it truly takes but one dolt to foul the system up.  Whether its the guy doing 45 in the far left lane, or the numb-nut who brakes hard to catch the exit he failed to plan for, when all lanes are at capacity, it really can take but one dolt to screw it up. Of course sometimes it is even more subtle - failure of people to maintain their speed as going up hills, or around bends, also has a snowball effect lasting far behind the irregularity in the road.  This happens constantly.

    Changing lanes and maintaining distances do not and cannot compensate for irregularities in speed differentials when lanes are full.

    Posted by El Jefe  on  12/27/06  at  10:44 PM
  18. Almost 3 years ago I bought a building lot about 300 yards from work.  It cost a good 60% premium at the time.  I fill up my truck every 40 days or so and in the summer ride a motorcycle.

    I am not sure when the last time I was on a freeway.

    Posted by Traffic?  on  12/27/06  at  10:09 PM
  19. I’ve seen this effect before, it need not be caused by the man in the red car. More of a soliton wave of traffic jamminess. Essentially more people arrive in the slowed down region than leave the area where the speeds have returned to normal. This creates a region of slowness that more or less sits in the same location, without any visible cause. Eventually as traffic numbers go down it will dissipate

    Posted by Cenobyte  on  12/27/06  at  08:50 PM
  20. The most annoying thing i see everyday driving home is that once people get into slow-go traffic, some people always seem to try and get ahead by cutting in and weaving. Or perhaps they want to get off on the left off-ramp but ride in on the faster moving right (driving on the left in AUS) lane and jump back at the last minute. Even to assume that these people arent going to physically slow down traffic if other people dont brake when they get cut off, people will always brake when they get cut off like this because you can never assume any other driver knows what they are doing, your car is the only one you actually have control over. If you just assume everyone else isnt a moron, thats how accidents happen. Its a rather wonderful negotiation process really, and most of the time it just magically works without incidents or accidents. Even with a few dickheads thinking they’ll get there quicker than everyone else by jumping 4-5 cars ahead in a traffic jam.

    I reckon anyway if everyone just cruised slowly without having to brake (and i mean EVERYONE, not just most of us) then the whole thing would ease up eventually. I agree with whoever said this before but if YOU are the only one doing this, everyone will just see the advantage and jump in front. (Anyway i think you can only do that if you own a volvo ;))

    Posted by Arg!  on  12/27/06  at  08:26 PM
  21. I can’t agree more then what 34 and 35 say.

    1. Keep a safe/GOOD distance.

    This makes for smooth bottle neck issues...everyone can drive fast, squize in and carry on.

    2a) Fast lane is for fast cars stay out if your not moving at least a little fast then everyone else.
    2b) If the car infront of you is doing 140, DON’T Ride their ASS. if you wish to go faster, find a highway that permits it. Weaving at High speeds is stupid. Once they raise the speed limits so that more can join you do not do this.

    3. Let people on the Highway .

    4. Signal.

    5. getting on the high...then Get going! (people merging on the highway at 80km/h should have the linsence suspended till they can prove they know how to drive again. (if the Car is not good enough to accelerate fasten enough, DON’T GET ON THE HIGHWAY.

    I am all for raising the speed limit on Major highways. raise the speed to 120-130.

    stiffen the fines more...for too slow or fast.

    Posted by cdpage  on  12/27/06  at  07:29 PM
  22. There’s only one reason why we have traffic jams… and its the same thing that causes pollution, the destruction of our environment, the depletion of natural resources, and the extinction of plant, fish, and animal species world-wide… you guessed it.  This planet is bearing more humans than it can sustain.

    No humans… problem solved.  That was easy, wasn’t it?

    Posted by shock1970  on  12/27/06  at  07:27 PM
  23. We should just make owning a car illegal, and driving one punishable by 10 years in prison.  That would fix a lot of problems.

    Posted by Danny Yee  on  12/27/06  at  07:14 PM
  24. The main traffic problems are 1) drivers that drive side by side and thus prevent people from passing in any lane, 2)fixed arbitrary speed limits and the misguided belief of safty they give people, and 3) drivers that do not understand how to proactivly prevent traffic jams (like when merging, or in slow conditions—hint- if you are driving in bad snow, its easier for all if you keep a constant speed - even 3mph is better than start and stop- your goal should be to NOT stop your car) and 4) vehicles/drivers that have a different idea of about acclerations - when getting on the express way STEP ON IT! When comming up on someone that is going to slow - PASS THEM. and for god sake pass the F&*%&^$%! semi!!

    Posted by kevn  on  12/27/06  at  05:30 PM
  25. The suggestion that everyone move slightly slower than the car in front of them is just poor thinking: think it out before the next time you drive, please.

    And driving just at or below the speed limit is rude to the 99% of other drivers who are keeping up with the flow of traffic: please keep that stuff to the left lane only and when there’s only one lane in your direction, pull off and LET ME BY! Don’t be rude.

    The best thing to do is to avoid tailgating so you can slow down by letting off of the gas rather than lighting your brake lights, which will cause two or three of the ninnies behind you to folow suit.
    In my observations, over 75% (aprx.) of the folks who hit their brakes without seeing a stop sign or a red light ahead of them do so for no good reason; they are using the brakes to control their fear and not their car.

    Posted by Mark Somerville  on  12/27/06  at  05:29 PM
  26. If you feel like starting your comment with something like “This article is rubbish.” go over and check in the mirror because you have the words “I am stoopid” tatooed on your forehead.

    Why? Because this article, while quite well written and quite interesting, is basically old news, as the theory behind it is quite widely accepted.  Just because you don’t understand the theory does not mean it is wrong.

    Most people know that you should keep your distance, but people are not perfect (far from it), and they will always drive too close, react too slowly, and react too much.

    If you want to criticise the article, make sure you understand chaos theory and complex systems first.

    Posted by me  on  12/27/06  at  05:28 PM
  27. I picked up an early book on this subject:

    The Theory of Road Traffic Flow by Winifred Ashton
    John Wiley and Methuen c. 1966.

    This has summaries of various theories, and includes descriptions of these effects.

    There’s supposed to be a maximum capacity, according to the simplest theory.  Add more than this and the speed drops.

    Posted by Bill Hickok  on  12/27/06  at  05:10 PM
  28. Posted by thesameguy on 12/27 at 01:19 PM:
    “You cannot rely on people moving at slower speeds or leaving more room to alleviate congestion.  People moving at slower speeds will eventually aggravate traffic behind them…
    IMHO, best bet is to dramatically increase speed limits to get people on and off the freeways as quickly as possible.  Well, thatís how I avoid traffic, anyway.  Itís amazing how quickly the commute goes twenty minutes before quitting time at 90mphÖ”

    Posted by Mike Burke on 12/27 at 01:31 PM:
    “OK, all you folks who think that simply maintaining a proper distance and adjusting speed nullifies this hypotheses need to think again.”

    Bravo! Finally some who understand! Those saying to slow down and increase following distance don’t quite get it…

    It comes down to how much time your car + your following distance takes up on the road. Reduce that time and capacity increases. Drive slower and/or increase following distance and your time usage increases, reducing capacity. I’m amazed how few people understand this.

    When I reach the end of a backup, I briskly accelerate and keep up with the car ahead (or pass if they fail to accelerate quickly). While doing so, I usually notice that a few others behind me are not accelerating quickly and are allowing 10 or 20 seconds following distance to appear. Those folks accelerating slowly are why the backup persists. And if you were to tell those folks that, few would understand…

    Want to increase capacity of a freeway and reduce conjestion? Raise the speed limit to the 85% percentile (80 to 90 MPH on most city freeways) and have big signs everywhere that say TO REDUCE CONJESTION MAINTAIN SPEEDLIMIT. And other signs that say TO REDUCE CONJESTION AVOID EXCESSIVE FOLLOWING DISTANCES. And onramp signs that say SLOW DRIVERS WILL BE TICKETED.

    Posted by Brad  on  12/27/06  at  04:46 PM
  29. /rant


    my point is that maybe we should really start focusing on more important things like .... not using cars independantly controlled during distances where we all are going the same place? system that manages that intermediary period between custom-for-my-location-only driving (like to ur house from the main street) and gets us from point a to b with no HUMAN ERROR involved

    Posted by joe brockhaus  on  12/27/06  at  04:32 PM
  30. your website does not render/format correctly in opera.

    Posted by sunknown  on  12/27/06  at  04:05 PM
  31. don’t dally in the pass lane. problem solved.

    Posted by nap_tehgorf  on  12/27/06  at  03:04 PM
  32. The proximate cause of the traffic jam is NOT the guy in the red car, it’s the person in front of him moving slowly that he must go around.

    Something I’ve noticed is especially bad out here in Hawaii is complete lack of lane etiquette.  People in the left most and next lane going as slow as 5mph below the speed limit (55mph), despite resonable conditions.  Now the reasonable driver traveling at 55mph must cross 3 lanes of traffic to get around the slow moving (and rude) drivers. 

    The slow drivers in the left lane create a 15mph speed differential, which is why people have to hit their brakes to go around them, causing even more slow down.

    Lack of turn indicator use also causes cars responding to lane changers to hit the brakes, whereas if they had warning of the impending lane change they could simply let off the gas in anticipation of the lane change and not have to slow traffic very much. 

    The third factor is the increase of trucks and SUVs on the road.  These vehicles have long stopping distances and slow acceleration, and tend to be very brake happy.  They also have very high and wide profiles, making it more difficult for drivers behind them to see traffic conditions and anticipate and avoid upcoming hazards and slow-downs, meaning the cars behind the trucks end up using their brakes a lot more.

    Posted by Guy  on  12/27/06  at  02:59 PM
  33. Interesting article; just a bit too short, I was enjoying reading it, then it ended!

    The first comment shows how much difficulty some people have in understanding the point, by that person’s thinking that every car in a jam moves perfectly together at the same speed. Comment 10 shows that those drivers who attempt to maintain a constant slow speed, instead of moving forward to close the gap, are as much a cause of the jam as the “man in the red car”. Such misguided actions, unclear perceptions and the public’s difficulties comprehending the complexity of the system show that we can talk about it forever, but it won’t sink in and therefore won’t ever be fixable. Unfortunately.

    Posted by [CF]Gareth  on  12/27/06  at  02:53 PM
  34. Bigger buffers.  It also helps to keep an eye on the car in front of the car in front of you- then you know if you really need to brake.

    Posted by Edified  on  12/27/06  at  02:45 PM
  35. Please fix the text encoding in the story. It’s filled
    with broken Wordisms, rendering it unreadable on
    many other-than-Microsoft browsers:: í for example

    Posted by Mister Pointer Outer  on  12/27/06  at  02:42 PM
  36. This article doesn’t take into account psychology. while taking classes to become a firefighter we learned about building evacuations and that when you put walking people closer together (front to back - like in traffic) the natural tendancy is to slow down. Apply this to a highway where there is light traffic and you will notice a relative low density of cars. Now look at a high traffic time and there is a high density of cars per same span of road. The cars are closer together and therefore psychology (as well as reaction time) makes people drive slower.

    One or two people may be able to run down a hallway - but you can’t get a crowd to run down a hallway if they are close together - they naturally slow. So we can observe a marathon starting to see this in effect.

    Another thing to consider is that not every cause of a backup is there when you get to the point of impact. Where I am in California, a typical accident will block traffic for about 20-30 minutes and then get cleared to the shoulder or off the freeway. If you are far back in the traffic, and the accident is gone by the time you get there - you won’t see a thing. The Highway patrol also gets reports of objects in the roadway or stalled cars on an inside shoulder that need to be moved off the road and will run a traffic break - essentially stopping the traffic for a minute or two to clear the hazard. When they are done, the hazard and the CHP are gone by the time most drivers get to the point of the stoppage.

    Let’s not forget merging traffic. When more cars come onto the roadway the current traffic must make room for them. Add a lot of cars and you have to add a lot of room - or slow down.

    It’s also helpful to think of a freeway as more of a conveor belt and less of a line. With a line - there is assumed to be someone at the front. But there is no ‘front’ on a freeway. Just a contunious flow of cars 24/7/365.

    Posted by name  on  12/27/06  at  02:34 PM
  37. To prove the point of the article, do this: Sit in the trees along an interstate and watch the unimpeded traffic pass by for a while. Then, go stand near the shoulder at the side of the road and watch the traffic jam begin. You haven’t done anything to the traffic. You just stand there.

    Posted by MADurstewitz  on  12/27/06  at  02:22 PM
  38. Actually this effect and the mathematics behind it has been known for at least a decade if not decades. CalTrans has huge amounts of data and video showing how an accident’s effects linger long after the wreck was cleared. Typically you’ll see a dense clump of slower moving cars that moves slowly backwards along the freeway.

    This effect only happens when the density is pretty high (i.e. during commute times). There is a critical density above which traffic flow becomes very sensitive to any disturbances which is why you sometimes see traffic abruptly change from freeflowing to stop and go. I’ve been in enough freeway jams that I can sense when the traffic density is approaching the critical point, but there’s not a whole lot you can do about it.

    I’ve also tried being a ‘fixer’ by traveling at the average speed of traffic and leaving lots of space ahead of me to absorb any slowdowns. Unfortunately this requires a certain bit of concentration and only benefits the people behind me. And cars behind me will become convinced that I’m slowing them down even though I’m traveling at the same average speed as the cars in front of me. Something about seeing space open up makes them feel like they’re being cheated, and they’ll pull stunts to get into that space.

    Posted by Maetenloch  on  12/27/06  at  01:50 PM
  39. I haven’t driven to work in like 7 years.  Too many nuts calling each other idiots in slow traffic.

    Posted by I sleep on the train (or bus)  on  12/27/06  at  01:34 PM
  40. I have seen this often and I think it’s simply a matter of too many drivers for the road infrastructure to handle.  I think it’s true that bad drivers aggravate the problem, mostly the people who believe they have the right to drive in the left lane slower than other traffic.  The left lane should be a passing lane and the slow drivers should try to stay to the right and change left to pass if able, but it doesn’t explain everything. 

    Traffic seems to flow similar to supersonic fluid dynamics.  If there is a disturbance, there is a shock wave effect where the “fluid” or traffic compresses.  If the density of the traffic doesn’t exceed the maximum possible density of the vehicle’s size versus the space available, the traffic can keep flowing and it will “spring back” quickly.  If it does exceed this, then a simple jam occurrs much like too many logs in a river.  The only solution I can see for this is more carpooling, smaller cars, or more and more lanes.  Unfortunately everyone seems intent on driving a huge top-heavy SUV.  I have a feeling when fuel prices climb a bit more, this trend might start to go away.

    Posted by Chris  on  12/27/06  at  01:32 PM
  41. traffic isn’t like weather in so much as we know everything that is involved in the system. we have rules to accomodate that which we leave up to distributed nodes(people & cars) in the system. perhaps its time we re-think our paradigm and decide that we should stop making dumb assumptions about fellow peers, and reacting in a worst-case-scenario type of mindset. (compare this to food products, FDA approved grade D meat is safe, but to classy restaraunts serve grade D meat? )

    point is, the quality of ur measurement and design of the system in which you are measuring will obviously has an affect on the average deviation from the ideal

    when u apply failsafe logic like this, and stop assuming that its OK to NOT WATCH THE ROAD and PAY ATTENTION, then it’s practically impossible for moving traffic to jam.

    Posted by joe brockhaus  on  12/27/06  at  01:31 PM
  42. b) if the car next to him wouldn’t have been so jolted by the ‘suprise’ of a car moving into his lane, and instead been expecting the ‘red car’ to move into his lane, or not SLAM his breaks and allow the car to move into his lane without causing panic on the road behind him. (odds are the red car was gaining on the car in font of him, maybe tailing .. maybe he was getting to that point bc the moron that he ‘cut off’ was hanging excessively behind the car in front of him who was going the same/faster speed that him, and there was no way for the ‘red car’ to move behind him to pass the slower car in front of the ‘red car’ (because the non-aware driver that slammed his break when someone changed lanes was just ‘cruising along’, following the car & brake lights in front of him)

    maybe it is absurb for me to think, but when i’m driving, i’m watching ALL THE CARS I CAN SEE (and assuming that all those cars are doing the same). i watch 1) how they are moving side to side in their lane(so i can tell when they will change lanes), 2) their relative speed to the cars in front of them (so i can slow down slightly to allow them confidence to move into my lane, flick my lights, etc)and 3) how they react to the cars around them (to make the best judgement on the above 2.

    if everyone is doing just this, there is a reeeeealllllly low probability that anything wrong could happen as a result of calculable HUMAN ERROR (as opposed to not-so-calculable things like deer jumping into the road or an axel falling off a semi-truck)

    Posted by joe brockhaus  on  12/27/06  at  01:26 PM
  43. 50 years ago traffic was initially caused by too many cars or accidents on the road, now that we are able to quantify what causes problems, why are we trying to satiate a quest for the impossible (’updating’ flawed logic to appear correct) when we can change the environment in which the problem arises, while still maintaining the high-level logic that we all care about at the root of things (get to point b from point a using cars and roads—this isnt right place to argue a revolutionary method of transportation nothing like we are accustomed to that CANNOT FAIL, so i will save that)

    the jam is caused by every subsequent driver behind the ‘cut off’ driver to fail to be aware of the road & cars ahead, and clearing an assured distance (btw assured clear distance should be variable: if i’m fully capable and trusting in my automobile that distance is very short, but i’m also able to change this distance at a much faster rate than someone with a base distance of much further away, due to reliance of the vehicle or reliance of individual nerves). (if they have to stop so fast, then they obviously can see the cars in front, so blind turns don’t apply here)

    a) the ‘red car’ could easily have been the car that that the red car ‘cut off’, only difference is the red car moved into a(n assumed) FASTER(more left) lane to avoid the car in front of him. TFA fails to include the pertinent details that caused the car to the left of the red car to brake so suddenly. the ‘cut off’ car should have been fully aware of the willingness of the car next to him to get over, and if not, should not position HIS car such that he is in the blind spot of the red car. in this scenario the ‘cut off’ car should have sped up to allow the red car to move into the faster lane, behind him. since there is an assured clear distance behind and in front of this car, there should be no problem. MOREOVER, if u are traveling on the highway at such a speed and see a car about to

    Posted by joe brockhaus  on  12/27/06  at  01:24 PM
  44. this is rubbish.

    first of all you’re making a claim about a problem of absolute NO IMPORTANCE to the actual problem that causes said problem.

    secondly i’ll suggest u think about this. what reason does the car have to brake when the red car cut him off? moreover what does cutting someone off mean? far too often i’m in the car with someone who gasps, ‘omg that person just cut me off! what an ***hole!’. while the car may be close in front of him, it is going the same speed, or faster than the car getting ‘cut off’. how do u get ‘cut off’ by someone going the same, or faster speed than yourself? you weren’t planning on stopping before the car got in front of you. as far as you could tell at that moment, the cars in front of you were moving at the same pace. the car is obviously willing to go faster than you (he got in front of you for a reason), and if he’s not then we’re breaking rules that we’re assuming are constant, or you, the cut off driver, are not paying attention to the road and the cars around you.(moreover if you are moving faster than that car and are a) on the left u are passing, b) you are breaking the law. as such (passing) you are the one taking the risk of driving faster than the car to ur left, who has the right-of-way)

    here, a set of circumstances is defined for a system that is primarily set up to fail, suggesting that had one element of the system not acted in a certain way (btw inside the given confines of said system), the system would not have failed (that in itself is a laffable claim - defending a system that you’re starting off by degrading). yet comparisons are made to the same system that would somehow work without flaw, had the ‘red car’ not switched lanes.

    current traffic designs and models are based on failed human tactics to manage traffic that donít scale with the same results. Which leads to the common false assumption that more of what one person is doing (that disrupts the failed system) has to be wrong

    Posted by joe brockhaus  on  12/27/06  at  01:23 PM
  45. The biggest problem I have seen is with drivers slowing down to observe some incident which should not deter them.

    EG1: A stranded motorist is safely in the breakdown lane not obstructing the regular flow of traffic. This will cause some of the more oblivious and ignorant drivers to slow down, albeit sometimes subtly but often drastically. Why do they need to observe.

    I make it a point NOT to look at accidents on the other side of a divided highway, that have no effect on me, and I tend to leave a larger space in front of me than most drivers but I have still come close to rear-ending someone due to his very abrupt, sudden, and drastic braking in the LEFTMOST lane just so that the driver in front of me can satisfy his curiosity by observing some construction or accident in the opposing lane which has no bearing on our existence (save for the accident he almost caused).

    Posted by Jamie  on  12/27/06  at  01:12 PM
  46. Set up web cams, yes, and then PULL the DRIVERS’ LICENSES of the punk that caused the jam. Send him to Guantanamo. Televise what happens next as a warning to other drivers.

    Either they’ll stop doing it, or we’ll remove enough drivers off the road that the problem will just go away.

    Posted by Down with O.P.M.  on  12/27/06  at  12:56 PM
  47. 1. this Does happen. but not to this extent. it doesn’t get worse as the slinky effect moves down the line. it only gets worse as congestion increases. (for instance if it backed up past to major on ramps)

    2. this is not all because of one idiot. its be cause of MANY IDIOTS!

    a. The red car is obvious
    b. the Car(s) directly behind them breaking when the don’t need to...or traveling too closely.
    c. The ministry of transportation, (they don’t future proof traffic needs, they put on ramps and off ramps too close, and traffers from collectors to express lanes to close to on ramps.
    d. the City for not providing a better public transit system that would get more drivers off the road.

    EVERYONE can improve traffic jams, just don’t drive so damn close to the car in front of you...if they Break, let off the gas, there is no need to break… your car is movie More then 100 KM/H(60MP/H) if you let off the Gas, IT WILL SLOW DOWN!!!! and you should have enough space still to break if need be.

    Posted by cdpage  on  12/27/06  at  12:51 PM
  48. OK, all you folks who think that simply maintaining a proper distance and adjusting speed nullifies this hypotheses need to think again. The technical reasons are a little too complex for a discussion in this area, but I’d like to recommend a couple of books: “The Goal” by Eliahu Goldratt is a very easy read and describes why constraints cause irrecoverable queueing errors. “System Dynamics” or “Industrial Dynamics” by Jay Forrester show similar models, but may require a good knowledge of functions and relations to follow the logic.

    The only thing that would make the “common sense” idea of contracting and expanding work is a synchronization or timing device tied to the precise dynamic constraint in the system. (It may help to think of the timing signal in a modem or computer bus.) The constraint moves, and affects all dependent traffic behind it. Curves, for instance, are a slowdown in any traffic pattern, but many people ignore the fact that changing lanes creates a curve in the flow which affects traffic behind in more than one lane.

    In Houston, we have a Mayor who thinks that clearing stalled cars will smooth out traffic significantly, but he ignores the fact that that the traffic at the end point must diffuse and park as fast or faster than than the incoming traffic can fill the grid. By following the traffic cams, I have observed the “front of the queue” coming face-to-face with “nowhere to go”. It looks like a butcher trying to pack sausage into smaller and smaller casings.

    Posted by Mike Burke  on  12/27/06  at  12:31 PM
  49. You cannot rely on people moving at slower speeds or leaving more room to alleviate congestion.  People moving at slower speeds will eventually aggravate traffic behind them, causing the guy in the red car to abruptly change lanes and go around.  Same can be said people leaving a large gap in front of them - either that gap will be filled by the guy in the red car trying to get ahead, or by merging traffic seeing an open spot to fill.  Even assuming nobody decides to fill the large gap, the person creating it must slow in response to the guy in the red car “cutting him off” in order to maintain a constant distance.  While the effect of him slowing may not cause an immediate congestion, eventually it will unless EVERYONE follows suit.

    I can’t imagine any scenario where you do not have these “elastic” traffic jams.  Slower speeds and greater following distance may reduce the violence of the expansion and contraction, but it won’t cure it so long as people maintain control of their accelerators, brakes, and steering wheels.

    IMHO, best bet is to dramatically increase speed limits to get people on and off the freeways as quickly as possible.  Well, that’s how I avoid traffic, anyway.  It’s amazing how quickly the commute goes twenty minutes before quitting time at 90mph… :P

    Posted by thesameguy  on  12/27/06  at  12:19 PM
  50. At work I’ve got a nice view of a major highway, and have seen this type of thing happen several times. Nicely moving highway, and then 3 minutes latter, everyone is doing 20mph.
    Seems it would make sense to set up some web cams so scientist could collect some data.

    Posted by Pan_theFrog  on  12/27/06  at  10:38 AM
  51. Similarly, a single person can “fix” a traffic jam by proportionally increasing the distance between themselves sand the car in front of them and maintaining a fractionally slower speed than the car in front of them.  The “fixer” and the cars that follow avoid the stop-start-stop-start cycle, but travel at a continuous (slow) speed.  The space that opens in front of the fixer is filled with merging traffic, which further lineates the traffic flow.

    Posted by Ejay Hire  on  12/27/06  at  10:21 AM
  52. I saw actual footage, as well as computer models, of this exact scenario in college (late 80’s).

    One car slowing down (the car that got cutt off) can create a traffic jam that lasts hours and stretches for miles. It takes several times longer to clear than it did to start.

    Posted by Tim  on  12/27/06  at  09:27 AM
  53. This is total tosh, becuase, not all cars drive bumber to tail, so therfore a few cars slow down and the remainder just continue on. If a car stopped in the middle of a lane cars all the other cars would go around the car in that lane sure traffic would slow in that section but traffic would keep moving. Now place a speed camera, or a patrol car at the side of the road and cars will probably break and drivers look down at their speedos, now if enough drivers do this somebody is going to get shunted up the bumper. Patrol cars parked on the hard shoulder and in laybys are dangerous.

    Posted by Anthony  on  12/27/06  at  09:07 AM
  54. Although the aggregate behavior of traffic jams resembles fluid dynamics, we must not forget the real reasons for traffic jams:
    1.  Those who fail to do the expected speed given the driving conditions. 
    2.  Those who fail to go around those who are going slower than the expected speed in a timely manner.

    It’s easy to blame the person who changes lanes as the real culprit, but the less people change lanes, the less fluid the aggregrate behavior of a given set of cars. 

    The slow lane changes and over-compensation on the brakes causes a given car to contribute to the “log jam” affect of a traffic jam, which naturally inhibits a more fluid flow of traffic.

    EG 1:  Slow car in left lane can only cause a traffic jam, if those behind them fail to move to the right and speed up to compensate for loss of space in the lane to which they move.  This includes waiting too long to move.

    Am I right or wrong?
    Assume that the drivers in NYC or Houston TX are more likely to change lanes and use their gas pedal for passing slower cars than say drivers in Indianapolis or Dayton.  Is the thruput on Highways with aggressive drivers greater or less than for those with less agressive drivers?

    How well a given set of cars move passed a given point depends heavily on number of lanes and average speed. 

    Why I don’t like agressiveness as a measurement, as much as I like proactive driving habits:  An aggressive driver reduces the distance between them an the next car.  A proactive driver is more likely to maximize the distance between them and the next car while NOT slowing down, unless it is absolutely necessary to do so.

    Posted by John Meagher  on  12/27/06  at  07:43 AM
  55. I hate that guy.

    Posted by lefty.crupps  on  12/27/06  at  07:40 AM
  56. The reverse of this article is easily observed.

    When in slow-and-go traffic, leave a large space in front of yourself.  Travel slightly below the average speed of traffic.  Ignore the lane-cutters and stick to your purpose.

    You can easily defeat the “wave of brakelights” effect.  I do it all the time.

    I also travel at or below the speed limit 99% of the time.

    Posted by William Lindley  on  12/27/06  at  07:28 AM
  57. The driver in the red car may be the proximate cause of the jam, but think about this: if even just a small handful of people behind him in traffic maintained a proper following distance, they would be able to avoid the ever-narrowing window of braking time, and there would be no jam.

    The moral of the story is: leave a three- to four-second gap between yourself and the vehicle in front of you, and traffic jams not cause by actual traffic hazards or accidents will disappear :)

    Posted by D W  on  12/27/06  at  05:51 AM
  58. Unfortunately, this hypothesis is so obviously wrong it’s laughable that it even got published.
    If traffic is going at 5mph where I am, then in front, somewhere, somebody must also be going at 5mph at the front of the queue, with NOBODY in front of them. Somewhere, there HAS to be a front to the queue. So some moron is blocking the traffic. The hypothesis given also presumes that every successive driver is somehow slowed down even more than the driver in front, AND that none of the drivers in front increase their speed again… Why would they do that? It can’t be the cause. Idiot.

    Posted by Stupid scientists  on  12/27/06  at  05:00 AM