This is an excellent analysis --- I can't wait to start experimenting with traffic manipulation myself!
I read the FAQ, but do not have time to read all the other guestbook entries, so my points might have been raised by others already:
* What would it take to get this information (including simulations) into the curriculum of every driver's ed course in the country, as well as every driver's re-ed course (that you have to take to atone for speeding tickets)? Obviously you will not be able to erase all selfish and short-sighted driving behavior, but since individual drivers can have a disproportionate positive effect on traffic, what percentage of drivers would need to understand and apply these principles to eliminate all behavior-induced traffic nationwide?
* Here in San Diego, the majority of on-ramps for sections of freeway that are subject to rush-hour traffic jams control the rate and merging of entering cars with alternating stoplights that allow one or two cars per lane at a time. I wonder what are the impacts to your theories from this input?
Reuben Settergren <firstname.lastname@example.org>
San Diego, CA USA - Wednesday, December 18, 2002 at 12:54:14 (PST)
while i'm thinking about it.... how many times have people made it somewhere faster by staying in the right lane because everyone is jammed up in the left? why is it everyone wants to be in the left lane? I live in an area where there are a lot of tunnels that go under rivers. anytime the traffic gets thick, there is a wave starting just after the tunnel starts going uphill. I think it's from people's delay in realizing that they need to give more throttle. If this is the problem(i'm pretty sure it is) wouldn't you think that the mon-fri twice a day drivers would figure this out after days, weeks, months, years of doing the same thing??????? why is our learning curve for driving so flat?? so many people whine and fuss about how much they hate traffic, yet so few people make an effort to change or improve their driving habbits.
Va Beach, va USA - Tuesday, December 17, 2002 at 17:48:59 (PST)
What would happen if everyone, immediately after passing a car in the left lane, returned to the left lane? I admit I usually drive too fast; however, sometimes I feel like I'm trying to make a point that some people are comfortable driving well above 55. I have no problem with people driving slowly.... i just wish we could merge to the right when driving slowly. Why isn't this taught in driver's ed.? Left lane= fast lane....right lane= slow lane. I often see people enter a highway and for now reason merge over 2 lanes to the left when no one was in front of them in the right lane. I tell my parents that if someone passes them on their right side then they're doing something wrong. They argue that they don't want to drive fast and they want to relax. I understand that completey. However, I think that they shouldn't be in the way of other people that want to drive faster.
Virginia Beach, VA USA - Tuesday, December 17, 2002 at 17:12:22 (PST)
I live at about 14 miles from my work. In light traffic commute takes me 20 minute. If the traffic is high but without stop-and-go waves, it is about 30 minutes. When waves form, I drive between 40 and 60 minutes, rarely a bit longer. I enjoy a very loose workday schedule - when I need I can shift my workday by up to 2 hours, so I have experienced area traffic pretty much at any time. My observations completely agree with the non-linear theory of how traffic jam is created. I am positive that on the drive back a huge contributor to the everyday traffic jam is the 2 mile long HOV lane from the 4-lane capital beltway to the 4-lane I-270 North. The HOV lane makes regular traffic use 2 narrower than normal lanes. The HOV lane is allowed to all traffic until 3:30PM; there are NEVER backups until 3:40 or so. By 3:45 a slow area starts building up; by 4PM it grows usually a mile long, and extends to 4-5 miles until 6:30, when HOV restriction is lifted. Then most often the clogged area dissolves in 20 minutes. Ironically, the solid area - that forms in my opinion because of the HOV restriction - slows down HOV vehicles a lot more before they are able to get on the HOV lane than they gain after they reach it.
Rockville, MD USA - Tuesday, December 17, 2002 at 07:44:20 (PST)
One point to add the the comment about tanks from email@example.com...
> I remember a guy describing the frustration of driving the last
> tank in an Army convoy. Although the algorithm was simple (maintain
> a certain distance from the tankk in front of you) and the
> boundary conditions benign (the lead tank maintains a near-constant
> speed), he says the driver at the tail of the convoy alternates
> between panic braking and frantic acceleration.
Bizarrely, this phenomenon occurs in retail supply chains where they call it "the whiplash effect". Small changes in demand at the store get propogated back through the chain (warehouse, distribution center, factory) until the factory sees huge swings in orders from one week to the next. It's something to do with the time order information takes to move back up the chain - faster information exchange reduces the effect. Presumably there's a mechanism whereby initially small errors (in this case, the lead tank doesn't drive at exactly the "official" speed) get magnified because the tank behind takes time to notice and react to them. And the tank behind that one, and so on.
Chris Evenden <firstname.lastname@example.org>
New York, NY USA - Monday, December 16, 2002 at 13:40:11 (PST)
Your content is very much like the theory I have been working on for a few years. I have been gathering insight on these topics for a while, and have a domain registered where I wanted to publish the findings/recommendations (urbandriver.org). We should put our resources together and make our findings more well known! I drive 75+ miles a day in Atlanta, possibly the worst commute in the US, but I can see that you have very similary experiences. I look forward to hearing from you. TJ Leverette Traffic Philosopher Founder, Urbandriver.org ~ 'Make Traffic Better'
TJ Leverette <urbandriver>
Douglasville, GA USA - Sunday, December 15, 2002 at 19:18:09 (PST)
Great thoughts, and I'm sure they work in theory, but I do have one issue. As you suggest we back off the traffic in front of us to avoid futher jamming, thereby creating a space between the front of us and the back of the jam, how do we keep everybody driving around us from speeding up and filling in the space we're trying to create?
Paul Kelly <email@example.com>
St. Louis, MO USA - Friday, December 13, 2002 at 13:46:10 (PST)
Read the FREQUENTLY ASKED QUESTIONS - billb]
I was intially looking for a web page on animating ocean waves and swells when I came across your page. I love it. Thank you for taking the time to demonstrate what is anti intutive is sometimes the best thing for a society at large. I have often used driving as a metaphor for life styles---way too fast, quick frequent stops, versus paced and better flow. Thanks again.
Free range hobbits <firstname.lastname@example.org>
Las Vegas, NV USA - Thursday, December 05, 2002 at 23:27:30 (PST)
You can make use of traffic waves even when traffic isnt heavy. I often drive on I-95 between Washington, D.C. and Richmond, Va., a trip of about 105 miles. When traffic is only moderate (say, 40 to 60 percent of highway capacity) on this three-lane interstate, Ive noticed that cars travel in clusters that range from about a quarter-mile to one mile long. Gaps between clusters are typically one-quarter to three-quarters of a mile long. From inside a cluster, traffic appears to be relatively heavy with a lot of jockeying. But if I set my cruise control so Im running 3 to 5 mph slower than the rest and let traffic go by me, Im soon in a gap where I can easily position myself so there isnt a car within a quarter-mile of me in either direction. Then I bump the cruise speed up so Im traveling the same speed as the clusters ahead and behind. In a gap its far less stressful than in a cluster, and surely safer. Try it.
Washington, DC USA - Sunday, December 01, 2002 at 08:16:05 (PST)
As you might already know in the Netherlands your solution to merging lane traffic jams is made practise. Reading your article I thought if it could also relate to sequential (work) activities in doing a project, by reducing delay period during the execution of work. What do you think? I read that you dont consider yourself an expert, but its nice to brainstorm about it. Regards, Ed van der Tak The Netherlands
Ed van der Tak <email@example.com>
Rotterdam, Netherlands - Saturday, November 30, 2002 at 08:01:19 (PST)
Great Research! It makes heaps of sense. Haven't yet tried out your techniques, will do so soon. Well done for enhancing such a subject, that everybody always curses about. You posess the rare art of turning a problem into a research, thus (partially) evaporating the problem. I'm forwarding the URL of your site to my friends. Don't remove your pages from the web...! Thanks, Michiel Stapel
Michiel Stapel <firstname.lastname@example.org>
Utrecht, Netherlands - Tuesday, November 19, 2002 at 13:48:47 (PST)
SMOOTHING THE TRAFFIC-WAVES MIGHT INCREASE SPEED. I played with M. Trieber's java simulation a bit. The ring-road traffic simulation gives interesting results. With maximum congestion (80 cars per KM in a lane), the average speed depends on the presence of traffic waves. Without any traffic waves, the average speed is a little above 42KPH. With many small traffic waves (11 waves encountered for each KM of progress), average speed drops by 20%, down to 36KPH. With a great huge waves (3 waves encountered per KM of progress), avreage speed drops 30%, down to 30KPH. It seems that the "fast moving" sections of the waves don't make up for the time lost while waiting in the "stopped" sections.
Bill Beaty <>
Seattle, WA USA - Monday, November 11, 2002 at 22:44:13 (PST)
NOTE: The Seattle Times article mistakenly implies that I can cure ALL traffic jams everywhere. I never said anything like that! (read the traffic waves website and see.) I guess it's my fault for not requiring that I see the article before publication.      In the same way that aspirin "cures" pain, these driving techniques are a "cure" for traffic jams. Aspirin certainly has an effect, but that doesn't mean we can use it to stop pain during major surgery! If the exits are overloaded and clog up the highway, no fancy driving technique can do anything besides improve your attitude. However, whenever a highway is merely congested and nearing capacity, the actions of one or a few drivers can have a large effect. They can trigger a major phase-change in the traffic patterns (in other words, either triggering traffic jams or busting them up.)
Search around during your commute and you'll find places where traffic jams are caused by the "bank-teller lineup" effect; where drivers form a very long, very close-packed row and won't let anyone in. Or find a jam where there seems to be no cause at all. In those situations your individual actions can have a major impact. I've found several in Seattle:
1. Busting the exit-lane jam on northbound I-5 near Yesler (at the left hand express lane exit that backs up all the way past the Rainier Brewery.) The jam occurs when people won't let anyone merge into the exit lane (yet drivers do get mad at this and forcibly merge anyway.) Simply allowing lots of drivers to merge ahead of you occasionally wipes out that jam. I've done this a few times, try it yourself!
Bill Beaty <>
Seattle, WA USA - Sunday, October 06, 2002 at 21:44:52 (PDT)
Ingenious! I have thought about this traffic phenomenon as well, but not as in-depth as you.
We have a famous and dreaded section on an artery road in our city where
four 45mph lanes are converging into three 45mph lanes, and the jams line
up for miles and miles before it every morning. My thought is that it
would be wise to speed up that section to 4x45/3=60mph to keep an even
flow, but that may be wishful thinking since there will be overhead from
the people changing lanes.
Swedes generally drive very courteously, but as you have proved it only
take one aggressive move to make ripples down the flow.
The solution is as always educating the masses. Teach people that it is
for the benefit of themselves to allow people to drive into the lane in
front of them. That they only make matters worse by speeding up to the car
in front and then slamming the brakes.
Loved the article in the Times and the site itself, which gives much clearer information. I've always driven like this, and I think my main motivation is that I'm too lazy to always be slamming on the brakes. Also, I hate to waste gas.
I drive on 405 every day and I used to get angry at the drivers who swarm
around me from both sides to get into the space ahead, while ending up
driving the same speed they were going before. Recently I noticed that the
cars were behaving exactly the same way as blood cells, in those films of
how blood cells go through the veins. Although it is a little scary that
most people drive using the intellect of a blood cell, seeing it this way
did erase my frustration. "Be my guest, amoeba-brain," I think.
I used to live in Seattle and noticed the traffic waves when going up to North Seattle, except slightly differently. I noticed that there were places that an individual car could speed up relative to other cars if they were close to trucks. I saw the gaps that appeared and disappeared near them, and it was very similar to a vacuum effect. When a slowdown occurs you want to be behind a truck because they will continue moving forward since they have accumulated space in front. When things speed up, you want to get merge in front of a truck since it doesn't move as fast to accelerate. I would contend the "trucker etiquette" observation (in the faq) comes from the fact that it's ridiculously hard to accelerate and decelerate something as massive as a semi and it just happens to work well for traffic. I think it's a great idea to keep these spaces in front and behind us and perhaps if this could be pushed as some kind of constant norm driving would improve, but until people do it enough that those who deviate will get some negative feedback this will keep on happening. A constant campaign would be useful..
Other useful things to include (outside of pure design of freeway issues):
Anyhow, I think what you say makes a lot of sense and it's got a lot of people in Seattle (and worldwide) thinking about it. :) Hope more people read this and something happens about traffic in Seattle!
Ya know, I've done this very thing (opening up blank spaces ahead of me) for years, only weakly imagining that it was accomplishing anything. Thanks for supplying the theory behind my behavior.
And yes, it DOES feel better to drive slowly than to
Cool site. I've to mostly the same conclusions as you have driving on I-93 north of Boston (and I-95, I-495, etc). I thought of the state trooper thing - a row of them driving slow, but I think this would make things worse. People already slow to ~10mph below the speed limit when they see a cop. I've seen them do this when they need to clear something out of the road real quick and just need a gap - guess what, the traffic piles up behind them.
I think the first thing that needs to be done is driver education. Slow
drivers drive on the right and other basic stuff like this. Second,
closer to your trooper row idea is what they do in Germany (I've seen this
work - amazing) is have big signs (the kind with lights that can change)
that display the current recommended speed limit. Of course this only
works when people pay attention, which is why step one must occur first -
and would most likely take many years. Unfortunately there's nothing we
can do about the people who say "The speed limit is 65, so I'm entitled to
go 65 in the left lane" even though people are going 70-85 in the other
lanes. Though education may help with some of them.
Great stuff here! I drive about 500-600 miles a week and have begun to employ your techniques. They work. I suggest the additional in order to improve our traffic woes:
- Privatize driver's ed and let people pay dearly for intensive on-the-road ed and take it until they pass a stringent exam. This works in Europe and by golly, people are more serious about driving there, even despite being more aggressive. Paying between $50 and $100 per hour of ed would likely get people to buck up and become responsible drivers.
- Implement the ideas on this site in every driver's ed curriculum, so people develop the strategies here as habits.
- Implement a "2 Tickets, 2 weeks no license" policy. Do you think people would take their driving more seriously if they were keenly aware of the consequences of poor, ingnorant and at times, unlawful driving? I for one can't imagine not being legally able to drive for two weeks. It would ruin me. When I see people broaching the law, they do so not because they don't know. They do so know full well that they are breaking the law. Of course, some scofflaws would still drive even during suspensions, but I think this approach would have a strong psychological effect.
- Start rolling blocks as suggested in the site, while eradicating speed traps that line our highways. You want to see a very rapid and dangerous slowdown, just watch what happens when people see a trooper lurking on the side of the road. I've even seen a couple accidents happen as a result of this scenario. It's unnecessary.
Based on my experiences driving, and what I have read in this site, I
believe that the above ideas would do much to make our traffic better.
It would be nice to save tens of billions of dollars and 30 more years of
revolving traffic constructions zones that will buy us no
What a good article, and what a simple solution to our traffic problems. And, if truth were known, there are probably other sensible drivers like you out there with excellent suggestions -- unfortunately, all our state can think about is spend, spend, spend. Hey, yeh, let's build a monorail to cast shadows over out city, to create large pillars of cement traversing our streets. Hey, yeh, rather than retrofit the Viaduct, let's tear it down and spend billions of dollars to build some kind of rail system under the city, after we dredge out half the bay. And, while we're at it, let's build a second bridge over Lake Washington -- 520 just can't handle the capacity. Oh, and that's not even mentioning how long all these wonderful improvements will take to build. Oh, almost forgot, let's be sure to dig up and tear out the present transit rails in that very expensive, artful bus tunnel that is vacant most of the time, so that we can extend light rail through the city. Too bad we can't use the present rails, but, oh, well, who ever even thought they'd be obsolete so early.
Darlene Cox <email@example.com>
Seattle, WA USA - Monday, October 07, 2002 at 20:59:31 (PDT)
Maintaining a proper following distance will allow smoother merges, as you point out, and thus improve the flow of traffic. (It also minimizes accidents since there is more time and space to react.) Would it not be possible to "force" these driving habits onto people? What if maintaining a following distance were required by law and subject to ticketing?
Seattle, USA - Sunday, October 06, 2002 at 22:39:43 (PDT)
See article about jam-busting by a professional trucker: gravel-pit etiquette and guide to bottlenecks
USA - Sunday, October 06, 2002 at 22:30:22 (PDT)
I saw the article in the Oct 6, 2002 Seattle Times. You're right on! Your intuitive approach can be "beefed up" by studying "longitudinal wave theory" but you are qualitatively correct. A slinky is the most common device which will demonstrate the effects you are talking about. Take a long (preferably metal and larger diameter) slinky. Fix one end and rhythmically pulse the other end at various speeds in line with the stretched out slinky. You will see standing waves and traveling waves as you change the frequency and amplitude of the pulses.
I've seen a
reference applying longitudinal wave theory to traffic, but it was so long
ago that I have forgotten the reference entirely. I do remember some of
the principles and conclusions, however.
Traffic has "regimes"
defined by the traffic density... At very low traffic density, no
smoothing is needed since the large spaces allow drivers to operate
independantly. Speed is usually at or above the speed limit.
As density increases, a critical density is reached where longitudinal
waves can form and speeds tend to be bistable or pulsing depending on the
density and persistance of the standing waves. In the bistable mode
traffic either moves near the speed limit or approximately 26 mph if there
is no accident or other actual obstruction. In the pulsing mode, stop and
go persists. I have seen stop-and-go be 0 to 70 mph near San Francisco
with multiple pulses!! These tend to be at the back of large slowdowns
where the fast traffic "runs into" the back of the jam and is adjusting to
a slower speed ahead. These are the traffic densities that driver
behaviour affects the traffic flow. Basically, tailgating and other rude
behaviour causes the waves to develope. Smoothing the flow, by
maintaining the average speed of the pulses and a large space to buffer
your movement so you don't stop can be effective on smaller waves, but not
very large ones unless a higher percentage of drives do as you suggest.
In higher density, the traffic moves only at the lower bistable speed;
typically around 26 mph. This is near the maximum capacity speed of
multiple lane freeways. Smoothing by maintaining space will probably not
do much if the flow isn't stop-and-go. If stop-and-go is present,
smoothing can help.
The on-ramp metering lights work by smoothing out the "pulses" of merging
cars caused by traffic lights. In Canada, say Vancouver BC, people used
to be much more courteous and "zippered" together really well, having a
similar effect. I haven't driven much up there for more than 10 years, so
that may have changed. "Zippering" at on ramps never has worked reliably
here in my observation.
My impression is that, in the last 5 or 6 years, we are at critical
density a much higher percentage of time; and the average driver is less
courteous, tailgating more, and causing wave effects frequently. The
frequency of multiple rear end collisions on the freeway seems to have
doubled or more in the last six years!!
Traffic courtesy and your smoothing techniques should be required content
for driver education programs and driver's examinations.
I've been a Metro Vanpool driver for the last 11 years. Most of my
driving is with a loaded 12 passenger van in rush hour traffic. I often
have to cross multiple lanes traffic to get to the inside HOV lane. It
used to be that, in the Seattle area, I would seldom have someone cut me
off when I used the turn signal to indicate a wish to change lanes. Now,
it seems that I can count on 2 or 3 drivers accelerating to cut me off.
I have had as many as 6 in a row! I am only wanting to reach the HOV
lane; which will not slow their commute! I soon pass them when I finally
reach the HOV lane. Buses have the same problem, only worse because of
their more limited acceleration and mobility. I see them in the right hand
lane, generally slowing traffic, because they can't fight the discourteous
Boston is classic case of aggressive driving causing horrendous traffic
conditions. As we approach their driving habits, we too will suffer as
I want to comment on professional trucks and stick shifts. It uses up a
clutch to start and stop. I typically try to avoid stopping and drive
like a pro trucker in heavy traffic because I don't want to use up the
clutch and brakes. I sold my last car with 160,000 miles on it and the
original clutch still fine!
Mr. Beaty: I just read the article in the October 6th Seattle Times. God, ow nice to hear a voice of common sense and sanity...
I used to have a friend who was a designer at Metro in the early eighties.
At that time, they had another name for traffice waves, they called it
"catepiller traffic", because of the inch worm style of movement.
I have to drive. I'm a consultant living in Brier and working in Tukwila.
It would be amazing if our driving programs would concentrate more on
courtesy and less on rules; we might make some headway.
Without knowing what to call it, I have been practicing what you've
described for the past year, with varying degrees of success. I too
thought of cars as molecules and have noticed that slowing down and
driving in the right lane at or slightly under the posted speed limit will
get me to my destination faster. Let everyone else fight to get in the
left lane: it crawls anyway...
My stress level is significantly LESS if I drive on the right in this
manner, instead of fighting my way through the clumps.
My only complaint is that it seems there are aleways the pieces of rude
slime that will take advantage of the space I try to leave in front of me
to dive in at the last moment. Oh well. (My prime example is the cars and
Canadian trucks that will dive in to the lane exiting North-bound I-5 for
the Express Lanes in afternoon at Yesler. They just slow everyone else
I read the article in the Times today and could not agree with you more. This is the solution I have known of for many years. I live in Seattle and commute to Redmond everyday via 520. I've practice your theory with great results. It amazes me that people don't see that this does work. I'm a 16 year driver with UPS and I know all about traffic and what causes backups and jams and how to lessen them. I'm happy to see this hit the paper and hopefully people will start to follow your suggestions. A little bit could make a big difference. Keep up the great work and thanks for having this site.
Seattle, Wa USA - Sunday, October 06, 2002 at 10:28:54 (PDT)
I loved these pages, it was something I noticed but never had put into print, though I had mentioned it to others. The below is slightly long.
I also agree with the person about the manual transmission, but
something similar applies to automatic transmission cars. The problem
though is that most people dont care/or think about what it does to their
cars. An automatic that has just had the brake released begins
accelerating prior to the gas being applied. Then the brakes need to be
applied to slow it down. Since most automatic drivers actively accelerate
This causes transmission wear from the jackrabbit start and increased
brake wear from the sudden stop & reduced gas miledge too! It's a loose -
In a manual a slow down can be accomplished by depressing the clutch removing power from the wheels rather than by tapping the brakes. Since I drive a cargo van I can slow down steadily by letting wind resistance slow me down. This gets the drivers behind me to slow down with out their thinking about it. Ive gotten tailgaiters(drafters) down below 40 on the highway(limit 65) with this method and get much enjoyment from it. A person responding in the AI mode while driving(see much below). They will even do it if there is multiple completely open lanes. Some people are more comfortable as followers though most follow so close I cant see they are there unless they have headlights on and its dark. Even then it isn't always possible to see them.
The main thing I see wrong with this page is the merge stopages. The steady merge is only attainable under two conditions. One, that the offramp maintains a constant velocity. Most don't, there is a light or multiple lights at the end. Two, the merged lane becomes a new lane on the new highway not a new merge. The problem is that drivers are really not used to the second case. Ive seen them but usually people come off that offramp and put on the brakes because they are not expecting to be in a new lane on the new road. A nearly perfect example of this is in the upper left corner of the picture on this link: http://www.empirestateroads.com/week/week29.html. It is marred by the ramp that pops very slow moving traffic into the end of the highspeed ramp (westbound to northbound).
The merge stopages usually occur when the road system is at or near peak capacity.
Here are my basic driving rules:
#1) If you cant see the face of the person ahead of you in the drivers side mirror you are TOO CLOSE! Dont be a tailpipe inspector.For a lighter side search for '20020528.swf' on a search engine, it's about 1.1M.
The Virtual Kid
Rochester, NY USA - Friday, October 04, 2002 at 23:39:01 (PDT)
Motorist are afraid of having an accident is one reason for traffic jams at a bottleneck. So they slow down to avoid an accident. Everyone can't drive as good as us and merge perfectly everytime. Some of us drive better than others. One bad apple spoils the whole bunch!!!!!!!!! Rubbernecking as you suggested will do it everytime. Let a cop pull someone over!
James Nelson <firstname.lastname@example.org>
John C Stennis Space Center, MS USA - Wednesday, October 02, 2002 at 13:03:41 (PDT)
If everyone had manual transmissions: Once again, I am not picking on John Richard, but I think the idea of everyone having manual transmissions is a bad idea. In traffic jams you don't save gas. Two, you're doing more work than normal shifting and clutching which causes arm, back, leg, neck, etc. pains (sitting in traffic is enough alread). That would cause more accidents in traffic jams. Three, you may not be inclined to ride on someone's bumper, but you may pause just that instance longer because you know you have to shift causing that one second ripple effect. Trucks leave a gap because they take off slow for one and they also know that traffic cuts in and out in front of them. So they do this to keep from rear-ending cars (where they would be 100% at fault). Not to disagree with anyone because we all have good points! Look at everything on both sides first.
James Nelson <email@example.com>
John C Stennis Space Center, MS USA - Wednesday, October 02, 2002 at 12:55:19 (PDT)
When you say you slow down and encourage cars to move long in front of you to keep traffic from freezing up and you keep a two second gap: Eventhough it is a good idea, your idea kinda contradicts itself in that when you stop to encourage cars to move ahead of you, [Where do I talk about STOPPING?!! To open up an extra car-length of space, just drive 1mph slow for 30 seconds. -Bill B.] the cars behind you will want to and more that likely will move to the other lane to pass you because traffic is moving faster in the other lane. This alone will cause an incident and a ripple effect of brakes lights going back until the end of traffic. Not picking on you but you would be creating that incident. Don't forget the person behind you wants your space and will stop at nothing to get it whenever they can! [Please read the FAQ section.]
James Nelson <firstname.lastname@example.org>
John C Stennis Space Center, MS USA - Wednesday, October 02, 2002 at 12:41:48 (PDT)
At work, there is a long single lane to get out of the office car park. This lane can get rather slow-moving during hometime. I've been playing the "average speed" game since I started work for the fun of it (also, I imagine I have passengers who get car-sick easily). While I've been playing the "average speed" game, I cannot see the traffic behaviour behind me (beyond one or two cars) because the road bends. However, I've suspected that the immediate cars behind will have no choice but to mimick my "average speed" behaviour. I found this site because it was featured on www.bbspot.com on 27th September 2002. Thanks for sharing your insight (: Regards, Kevin
Kevin K. Woo
Sydney, NSW Australia - Thursday, September 26, 2002 at 17:56:36 (PDT)
WOW! I have been manipulating traffic like this for years, and occasionally I would talk to people about it with many of the same conclusions that you have reached. I had no idea that this was here. You have done a fantastic job with this topic. Hitting the average speed of traffic is like a game with me and has been for years.
I have a few extra ideas that I would like to share. First, I believe that the ease of automatic transmissions contributes to the traffic jam problem. I drive a standard, and constantly clutching in stop and go traffic is a major pain. I started the "average speed" thing so that I could stay in one gear as long as possible without having to clutch. I imagine that this is also why truckers always leave large gaps infront of themselves. It is less a case of helping traffic and more a case of personal convenience. People who only have to push stop/go are not inconvienced in the least by riding someones bumper. Basicly, I feel that if everyone drove a standard (where it actually requires effort to ride someones tail), few people would be inclined to "guard" the gap in front of them. The harder the clutch, the better the traffic.
Second, you mentioned at one point that you found yourself causing the problem. I think we all do this. I call it the human AI (not Artificial Intelligence, but Autonomous Idiot). It is that part of us that subconsciously takes over the chore of whatever we are doing while we think about something else, and it happens ALL THE TIME while people are driving. Look at people in their cars from time to time and you will notice that you can tell when they are in 'AI' mode. I almost slammed into a woman who was making a turn in a wet intersection. I was about 1 foot from her car when I stopped, and I could see her face as clear as day... she was in another world, and never saw me. She never acknowledged that the almost got creamed... just went on her way without noticing my presence. I am willing to bet that a large percentage of wrecks happen when a person in 'AI' mode meets another person in 'AI' mode ("I didn't see it coming" or "they came from out of nowhere"). But there is a nice side to 'AI' mode... It makes drivers very predictable. They just cruise along, without doing anything inventive, risky, or creative. I guess that a constant of 1/5 to 1/4 of traffic is people in AI mode, and that all drivers do this from time to time. If we could combine this with maybe a constant 10% of people who are averaging the gaps, I think the system would be smooth enough to eliminate jams regardless of how the other ~65% behaves.
Finally, there was a comment about increased density of traffic on the up side of hills, and near the banked areas of freeways. I travel on several of these and have come to a conclusion that does not include reference to any off/on ramps. The up sides are always slow, and I think that there are a few factors contributing to this. Construction vehicles, and diesel rigs frequently do not have enough power to maintain their momentum up a long incline, and are forced to shift down to a slower gear with more torque in order to make it. Sometimes this slowdown is extreme, and it is very predictable. If I know I am approaching one of these areas and I see a large truck that "looks" underpowered for what it is carrying, I switch lanes so that I don't get stuck behind it. This works to my advantage 95% of the time, regardless of the lane I am switching to. Second, I think people naturally slow down on the up sides of hills because they are just holding their gas constant without realizing their slowdown in speed. Finally, I think that a significant number of people are not comfortable driving on freeways, and having high-speed turns and hills just magnifies their discomfort. They slow down to navigate these "obstacles" safely. I routinely see unnecessary brake lights near turns in the road and hills.
I think driving classes should teach this stuff so that more people would understand that they contribute to many of the headaches they hate. Dallas is constantly budgeting more and more for highway expansion. I feel like they could save a lot of money by teaching people how to properly use the roads we have.
Keep up the great work.
John Richardson <email@example.com>
Dallas, TX USA - Wednesday, September 25, 2002 at 17:01:22 (PDT)
This is one of the most in-depth, reliable, interesting and LITERATE sites ive accessed. Its great to know that there are people who emphasize the practicality of the study of Physics in everyday life. I know this site has been around for a while, but hey, better late than never. I'll be trying out this theory of yours on Philippine traffic. We could use that here in Manila.. "congested" is many times an understatement to the state of our car flows. Thanks alot and more power!
Manila, Philippines - Sunday, September 22, 2002 at 08:37:28 (PDT)
Bill, I have finally gotten around to reading your 'traffic waves' article . . . very good! In fact the subject has been formally treated and is the 'theory' and factual parts have been (and are being) used by many many computer performance and capacity analysts around the world: only we call it queueing theory. The traffic waves you describe are, in fact, queues - structures which pop data (cars) from one end while pushing data (cars) at the other end . . . the latency which occurs once the accident has cleared we call 'wait time' or 'latency' and the time for a driver to enact a function is 'service time'. This theory has been used to explain response-time problems with computers, to design queueing systems at banks nd movie theatres for optimal customer flow and also to handle (though it does not seem as if too many people use it these days!) to figure out the seating capacity of restaurants given the speed of the cooks in the kitchen and the size of parking lots at chain stores. Thanks for the interesting article . . . maybe you could write about the speed of cars on the turnpike as opposed to the 'jams' at the ticket booths . . . that is queueing theory also!!!
USA - Sunday, September 08, 2002 at 06:13:44 (PDT)
What a great site! And the links to the Seattle driver stuff made my day - and other friends in the transport planning community once I e-mail the site to them ...
Ottawa, ON Canada - Tuesday, August 27, 2002 at 08:59:22 (PDT)
A very good explanation about an every day event. The only problem is if you drive slower in a two lane (or more) street everyone in the other lane will rush into the gap that you are trying to create... [You'd think that would happen, but in reality it doesn't. If I drive 1mph slow for only a very short time, I will back off from tailgating the person ahead, yet nobody behind realises this is happening. Even better is to never slow down at all, instead just preserve your space when you approach congestion. See the FAQ section. -Bill B.]
Carlos Crespo <firstname.lastname@example.org>
Lisbon, Portugal - Sunday, August 04, 2002 at 11:06:32 (PDT)
Your ideas equally work well for toll-booths and ticket gates. Also, at least here in Tokyo, there are numerous road side displays showing where and when the traffic jams occur.
Tokyo, Japan - Tuesday, July 16, 2002 at 01:26:45 (PDT)
What a neat site. We plan to alert our subscriber's list about this site. Very informative.
Road Warrior Ezine <email@example.com>
Seal Beach, CA USA - Wednesday, June 26, 2002 at 21:08:12 (PDT)
The mystery is: why has it taken so long for traffic engineers to notice these obvious patterns? (it is only very recently I read some researchers in Germany started to simulate individual driver behavior with a computer program, an easy task for anyone who has ever played with a BASIC home computer.) My hunch is that most traffic engineers studied GEOGRAPHY at college, because they were too stupid to study anything more interesting... [Many engineers are convinced that each driver behaves totally different than all the others, therefore traffic has no patterns and obeys no general rules. This is wrong. In reality, the many similarities between drivers allow an entire repetoir of "traffic physics" phenomena to exist. We blind ourselves to it by assuming that it cannot exist. It took some physicists specializing in Cellular Automata and Criticality theory to set us on the right path. And it only happened in the mid 1990s! -Bill b.]
Varese, VA Italy - Monday, June 24, 2002 at 09:24:08 (PDT)
Hmm. I've always had an aversion to being in tight traffic and never really understood why -- just thought it was because I might need to change lanes suddenly to avoid an accident or road debris. Turns out I just like to drive fast, huh? So I guess I'll see if I can try this lubrication strategy next time I'm in urban traffic.
Kevin McGehee <firstname.lastname@example.org>
Newnan, GA USA - Sunday, June 23, 2002 at 14:05:28 (PDT)
The London Orbital M25 system is more sophisticated than Brenda says. It uses variable speed limits set automatically by sensors that detect the volume of traffic. The higher the volume the lower the limit. The maximum UK limit is 70mph, this can be taken down to 40mph in the variable spped section of the M25. It is enforced by laser speed cameras hidden on the back of the overhead gantries that display the speed limits. (Not on every one, but you cannot be sure which ones have them!) This has the same effect as your rolling State Trooper road block. It works very well until the traffic density exceeds the capacity of the road, at which point everything grinds to a halt, and no amount of calming will fix it. This happens at least twice every day, and sometimes more often.
You may think that the slower the speed of traffic, the greater the cars per hour. This is not so. Some basis math shows that there is an optimum speed for maximum traffic flow.
The time interval between vehicles passing a point is the distance between the front of one and the front of the next divided by the speed. This distance has two elements:
a) A Fixed Element - which is vehicle length plus the minimum gap people will leave while still actually moving
The time interval is the sum of the time for the two elements. The time for the first element is INVERSLY proportional to speed, while for the second element is DIRECTLY proportional to speed.
Thus the speed for greatest throughput is when Fixed element = Variable element. Below that speed the time taken for the first element increases more then the reduction in the second element, and above that speed the time for the second element increases faster than the reduction in the first element. Apparantly it is around 40-50mph which explains why the M25 usually jams solid when the 40mph signs are showing.
In order to increase traffic flow on the M25 (the main London orbital motorway) speed limits have been REDUCED from 70 to 50 mph. Everyone gets there quicker as there is less braking/acceleration and lane changing.
Brenda Jolley <email@example.com>
UK - Friday, June 21, 2002 at 05:42:58 (PDT)
Around the Washington DC beltway, there are occasionally police cars apparently doing what it is you suggest. It doesn't require several abreast, because people slow down in order to not pass a police car.
A note about your animated example of forced lane merging - the 'fast'
example showing the cars leaving merge-space in front of them doesn't
actually show what you describe. By the description you give, the cars in
the right lane, when someone merges in front of them, should slow down to
allow their gap to reappear. When this slow-down happens, the effect on
the traffic flow is pretty much the same as in the left animation -
everyone has to slow down, the whole thing compresses, and the theory
You might argue that a reasonable driver would know it's okay to not
reopen the gap for the short duration of single lane traffic... But what
happens when three lanes merges to two, then two to one? That happens
quite often around here, and if people have let the gaps stay closed in
the first merge, then there are no gaps for the second merge. If they
*haven't* let the gaps stay closed, then we're back where we started, as
all the drivers are forced to slow down to reopen the gaps, and we have
standing waves once more.
I may have to write some simulation programs to test several theories -
lights, forced merges, exit and entrance ramps, stop signs... And
roundabouts - I believe roundabouts are a hugely superior device to stop
signs or lights for crossroads (and even more-so for turning a pair of
close-knit crossroads into a single six-exit roundabout).
Great idea! I have frequently seen semi trucks do something like this when two lanes are merging. There are always morons who want to drive 90mph in the lane that is merging right up until the last minute, when he HAS to merge. Then he swerves over and creates a wave. I frequently see semis that will just slow down to the average speed and drive so they are blocking both lanes and preventing idiot drivers from doing that. It clears the problem up pretty quickly.
Sherman, TX USA - Tuesday, June 18, 2002 at 07:52:49 (PDT)
I can't believe they are getting rid of the camareo. The F-Body is the best. They have lost their minds
Chilhowie, Va USA - Tuesday, June 18, 2002 at 07:48:20 (PDT)
When I was a kid, we were on a two-lane road that was going real slow, and I asked my dad why. He said it was merging down to one lane up ahead. I said "Oh great. I'll really slow down then." My dad, a mechanical engineer, said that the opposite would happen, because in order for our lanes to be going say, 10 mph, the single lane up ahead would have to be doing at least 20. Hmmm.. That got me going about this traffic engineering stuff.
Your idea about slowing down ahead of jams is right on, IMO. I do it all the time, and while I haven't noticed its affect on the jam itself, it keeps me much more calm. The same with upcoming red lights. Slow down, and coast up to anything on the road ahead that is a stoppage. You'll breeze through with less stress, and help the overall situation.
I used to commute to Redmond from Federal Way. I have thought about traffic flow concepts for a while, and you did a good job of putting them on paper. My technique during "black" stop and go traffic is to drive as slow as possible - I would rather drive 3mph steadily; instead of driving 10mph for a few seconds, and stopping for a few seconds, and repeating. Also, when approaching red lights, I slow way down in the hopes of it turning green. It's a lot easier to accelerate to 30mph if you're going 10mph rather than stopped.
Auburn, WA USA - Monday, June 10, 2002 at 15:28:06 (PDT)
Nice explanation of some traffic behavior. But there is one "big picture" item that is missing: the total capacity of the road. If a road can only carry 100 cars per minute per lane, it doesn't matter how people drive if 1000 cars per minute want to use that section of road. In that case you're going to have backups no matter what you do. The only solution in that case is to convince people to wait a while before taking their trip. The techniques outlined here work best when the road is nearing its capacity but is not yet at capacity. I guess you could argue that delaying your trip is just an extreme form of friendly merging behavior... ;-) BTW, the idea behind on-ramp traffic lights is to space out the cars that are trying to merge (one half of the problem). The benefit of the lights is not obvious, so lots of people complain about them. Interestingly, Minnesota tried an experiment in 2001 where they turned off all the on-ramp traffic lights for several weeks. At that point it became obvious that the lights had been helping because traffic immediately got worse. So everybody voted to turn them back on and now traffic is back to where it was before.
Jeff Martin <firstname.lastname@example.org>
Seattle, WA USA - Monday, June 10, 2002 at 13:45:32 (PDT)
It's funny how the most simple concept is so difficult for others to grasp: TAKE TURNS! Didn't their mothers teach them anything?
KLS < >
Seattle, WA USA - Monday, June 10, 2002 at 12:45:24 (PDT)
Interesting reading, here's an idea, if you explained it to a large number of people that spend a lot of time on the road, (Like delivery drivers, cops, and roadkill) and they did this, then a large, long lasting improvement could be seen in overall traffic might be seen.
The large scale application of this theory relies entirely on education though, and I doubt that the majority of drivers are inteligent enough to understand it.
Well of course your ideas make sense. The problem is that most motorists don't. They will threaten your life if you don't fill the gap. Especially in California where bumper-to-bumper is the only acceptable mode regardless of the speed and use of signals is perceived as a sign of weakness. The solution is not a small cadre of knowledgeable drivers but rather for law enforcement to attack the most dangerous driver behavior which is tailgating. I have never seen any be stopped for this and I don't believe traffic officers ever do unless an extreme example. However the vast majority of accidents, injury, and misery are caused by following too closely. I'd say about 80% of drivers here in CA follow at a distance that is dangerous to themselves and others. Traffic congestion is probably the most benign side effect of this stupidity. Enforcement's favorite target -- speeding -- is a distant second in terms of danger. One can exceed the speed limit safely but one cannot follow too closely, safely. Hand out a few million citations for following too closely and you may change something. Yeah, dream on. Cars suck.
arecibo, puerto rico - Wednesday, May 15, 2002 at 06:47:02 (PDT)
interesting theory and the state trooper thing caught my attention. i dont think it would work in a single line. but if they were out there singley doing what you do then i think it would work. I'll use steel as an anology. Steel is iron with carbon added at about 1 part per thousand (might have been million). The carbon prevents slipping between atoms and stregnthens the metal. The state troopers could be like the pieces of carbon within steel. Not in a single line but flowing within the trafic and softening the jam throughout the problem. More continously than a single line would do
Seattle, Wa USA - Thursday, April 18, 2002 at 22:13:11 (PDT)
A good, and interesting read.... I do a round trim of 100 miles a day, and am frequently stuck in a Jam. Your reasoning will, as you say, reduce the 'wave' effect, but I think only as far back as, say 50 or so cars... The point is that people will 'queue' up behind you, urging you to 'put your foot down' when they realise that there is a big space in front of your car.... Eventually, after a certain amount of distance behind you, people will start cutting in front of each other, etc, and the wave will start all over again.... Now If you Look at the whole situation from above, you'd find that by slowing down, you had not changed the situation at all, but the cars 'in uniform' behind you were all part of just another, Peek (or is it trough) cycle of the wave..... This is true, even when the gap in front of you has disappeared when you catch up to the people in front.
When [on my travels] I come to the end of a 20 mile or so spot of congested traffic, I have noticed that there seem to be a certain percentage of drivers (say 10-15%) Esp. HGVs!!) who have adopted your anti-traffic methods to, in their opinion, increase the traffic flow, however, by doing this, they actually slow down the flow, as at this point, the congestion is coming to an end . After the 20mile or so of stop-start, they are not immediately aware of the increasing traffic speed, and allow huge gaps to appear in front of them before accelerating away (Im talking 60+ cars gap here!). The point is, that in my opinion, these large gaps become like actual obstructions, with drivers queuing to get around them, in the same way they might try to get around something more physical, e.g. a rolling roadblock. Maybe there should be automated signals that inform the driver that the Motorway (Freeway) is becoming clear to prepare the driver to pull away?RubberNeckers Those people who slow down to gawp at an accident. Has anyone ever seen a Rubbernecker? I recon, that even a fleeting glance by a driver to an accident, and a speed decrease of say only 1% will be enough to cause problems, as the person behind will have to also slow by 1%, and then may also have a quick look, slowing and additional 1%, and so on, until theres a big queue. This is human nature Again, Motorway signs urging drivers to speed up at the accident, may be the answer?
To Sum up, I think that your analysis is, at first glance, valid, but there is much more to motorway congestion than you have put forward.
Tony Weston <email@example.com>
Wolverhampton, UK - Thursday, April 04, 2002 at 04:48:24 (PST)
I learned about traffic flow in 1985. Quite fascinating. I believe it started in its modern form by a physicist (1950s) modeling traffic flow the same as a one-dimensional gas. There are shock waves and everything. The defining feature is the dispersion relation (flow vs density curve), if I remember correctly.
The book I used was "Mathematical Models: Mechanical Vibrations, Population Dynamics, and Traffic Flow" by Richard Haberman. Lighthill & Whitham wrote about kinematic waves in the traffic flow context in Proc. Roy. Soc. A, 229, 317-345 (1955). But there are no cool animations in those old articles!
Thanks for the great website!
bob jones <firstname.lastname@example.org>
USA - Sunday, March 31, 2002 at 22:40:09 (PST)
Your comments on traffic waves and the need to merge freely at speed rather than compete are spot on in my opinion. If even 5% of drivers were intelligent, considerate and had thought some of these ideas we would see far less persistant congestion. In the UK we see long slow moving lines of traffic beside empty lanes with the odd car flying down the empty lane and trying to cut in at the front. On one occasion I saw the approach to roadworks (two lanes to one) signed USE BOTH LANES for miles before the lane closure and merge in turn signs close to it. People were obeying the signs and although the traffic was heavy it was flowing fast and freely, why don't we see this more often?
Nick Lewis <Nickelplates@yahoo.co.uk>
Rugby, UK - Saturday, February 23, 2002 at 10:36:21 (PST)
Wise words. I've been doing just that for years and so far haven't understood why 97% of drivers won't get it. What's most important about this anticipating way of driving is all the saved energy. Especially in America where smaller cars don't seem to catch on. Every time a car stops, the inertia gatherd so far is thrown away as heat and has to be regained converting the enrgy from fuel. So as I save my nerves I save fuel and the atmosphere too. In a car with manual transmission it's also easy to save fuel by using gears for slowing down. In a motor with fuel injection no fuel goes in while enginebraking. I want to bring this up since I've noticed that just the sight of a pair of brakelights flashing can cause an epidemic of brakelights, slowing the traffic down for no reason. But you can allways overdo things. A few years back I actually had to change rear brake discs in my car because for not enough braking they got corroded.
Jussa Nieminen <email@example.com>
Espoo, Finland - Thursday, February 14, 2002 at 12:37:57 (PST)
Great to find others think the same. Fascinating reading.
cheltenham, UK - Wednesday, February 06, 2002 at 11:12:33 (PST)
This is pretty much the way I drive, without realizing how great it was for the traffic pattern - it just seems to me that braking and accellerating as a pattern is very ineffcient and wastes gas and brake linings. My mechanic tells me that other people with my car complain about the need for annual brake jobs at a couple of hundred dollars each. I've had my car, bought used with worn brakes, for 5 years, and have never needed a brake job. Still don't - I just don't use my brakes much. Thanks for another good reason to keep up my moderate driving approach.
Judith Haemmerle <firstname.lastname@example.org>
Seattle, WA USA - Friday, January 18, 2002 at 19:13:50 (PST)
In traffic up to moderate levels, it seems that the most throughput will occur if drivers allow faster cars to pass on their left. As traffic increases, there's probably a point where the theories described in this site apply to all lanes. I encourage you to expand your study to address the "passing lane effect".
Bay Area, CA USA - Thursday, January 10, 2002 at 15:24:30 (PST)
Fascinating site. Referred thru Rob Morse's daily column. I encountered the same theory on merging years ago in L.A.; where by not getting territorial you get merged,faster. It worked reliably enough that I could demonstrate it to others. Unfortunately it isn't working as much any more. Where people once used to realize that you weren't going to contest their "right" to be there, they now see a space as an opportunity, and in the selfishness of the day, two or three extra cars and trucks speed up to cram in and none of us go anywhere. Any suggestions for a scientific solution for growing rudeness and selfishness? Responding in kind would be little more than going back to the stone age of driving.
Paul Tominac <email@example.com>
San Francisco, ca USA - Tuesday, January 08, 2002 at 09:31:52 (PST)
I wonder whether a cadre of "shepherd drivers" who drive in the wave-eating, merge-flow-promoting manners you describe, would do more to help speed traffic than a bunch of new freeway lanes.
Greg Corning <firstname.lastname@example.org>
San Francisco, CA USA - Monday, January 07, 2002 at 20:51:46 (PST)
I am fascinated with your site as I too commute 60 miles every day 30 each way. They need to enforce this especially in the convention center area of the freeway as it gets intensly frustrating to get off on the right exit or left exit. Police unfortunetly get annoyed when one goes 5 miles under the speed limit around here in the rain. Thank you for your insight.
Seattle, WA USA - Saturday, January 05, 2002 at 11:19:03 (PST)
Hi, I more than agree with you analysis of the MERGING-LANE TRAFFIC JAMS.
You say you dont need many people to change their behavior and to eliminate the traffic congestion.
Do you have an idea of the amount?
A simulation program can find out the exact amount. By defining the percentage of cars who space out the room in front and the rest who dont you will see if a there is a change.
Do you know a work that tries to answer my question?
I think that 20-30% of the people can be convinced to change their driving habits. If a bigger percentage is needed, then Im not sure the change can be achieved.
Jim Lader <email@example.com>
israel - Tuesday, December 25, 2001 at 08:11:22 (PST)
Nice [qualitative] discussion of the phenomenon and our ability to affect it. I've thought (on occasion) that this should be amenable to mathematical analysis, but my training is not sufficient in that area. Regardless, I also drive with space ahead of me and find it less stressful than my earlier technique. I have also found that it works on traffic-light controlled streets, where I try to pace myself to hit the lights when the traffic which has rushed to get to the red first finally is moving. Saves gas, too!! Nice work, Donald.
Donald H Locker <firstname.lastname@example.org>
Chelsea, MI USA - Friday, December 21, 2001 at 19:46:19 (PST)
I read your artical and it all seem very obvious as i was reading it. I know i've been caught in that merge situation in bumper to bumper traffic, especially when i dont knows its coming, and thought this whole thing could be a lot better if everyone drove like they were supposed to. I think most of the ideas (i.e. staying back a couple car lenghts) you had, were taught or should have been taught during drivers ed. And it does sound very familiar to defensive driving. I think cities with traffic problems need to emphisize this. I suggest you take your ideas to the driving schools in you city and try to sell them this idea. If a bunch of new students are taught this today, then just think about how traffic can potentially be changed in about 10 to 20 years.
Columbus, OH USA - Thursday, November 01, 2001 at 20:50:53 (PST)
I drive a nasty stretch of road daily with one truely awful bottleneck with two major highways comingtogether and two lanes of solid traffic merging into three lanes of solid traffic at the worst times. I felt today that I have to quit my job because I just can't take the selfishness of the road another minute. I, like you, try and analyse traffic patterns and I know what you are saying is true. You've inspired me to be proactive and do me and others some good.
Tri-state area, NY USA - Friday, October 19, 2001 at 19:24:47 (PDT)
This has all already been extensively studied by actual scientists and modelled in computer simulations. At least it shows that what they come up with at MIT can also occur to the common commuter, even one in Seattle. One point you missed and should have gotten was the phenomenon occuring in the absence of a block to traffic flow, that is, slowing due to volume.
Dean Obsy <email@example.com>
Portland, OR USA - Wednesday, September 26, 2001 at 15:12:24 (PDT)
I had only observed the phenomenon, and on a lark searched for any web pages that described the phenomenon. Thanks. Now a note from my evil twin: Once in Austin, TX, an idiot behind me hit his horn the instant the light turned green. By using the antithesis of your recommendations, and careful realtime analysis of each car around us, I was able to prevent him from passing me or anyone else on I35 for several miles. He finally pulled off the road entirely in frustration.
Robert Lane <firstname.lastname@example.org>
USA - Friday, September 21, 2001 at 17:00:51 (PDT)
Many many good points. I've been thinking about this sort of thing for a very long time..yet never thought to experiment. Great work.
Bellingham, WA USA - Friday, September 07, 2001 at 01:06:20 (PDT)
i think youre thinking way too hard on something that will only get worse as the years go by and the # of drivers on the road goes up ......too many damn cars and too many ignorant drivers .....its amazing how you figured it out tho lol ....like ...the "science" of it all
USA - Thursday, August 23, 2001 at 21:07:24 (PDT)
Totally correct, I've been observing this phenomena for years. Just one addition. The point you made about the 'shock wave' (as I like to call it) occurring for no apparent reason. There is a reason, in the UK we have too many drivers who tailgate, the car in front touches his brakes for some reason, the tailgater slams his brakes on, the tailgater behind him does so a little harder. Yup here it comes the shock wave.
Bob Jones <email@example.com>
Halifax, UK - Wednesday, August 22, 2001 at 07:42:47 (PDT)
wallpapers <ffveikw@millionaire intraining.com>
USA - Wednesday, August 15, 2001 at 14:14:13 (PDT)
Nice, great, like the descriptions, yadda yadda. ;) No, really, I enjoyed this site. What makes me post, though, is this: I do not drive. I utilize buses quite a bit here in Austin. People zoom around the 'slow' buses all of the time - but what springs to mind after reading your page is the fact that I have never, ever been caught in a traffic jam while on the bus except when the bus is an express/commuter bus that takes a highway for a long stretch without a stop - meaning they can accelerate to the speed of the local traffic. Interesting.
D. Johns <kallistiREMOVE@morpho. dar.net>
Austin, TX USA - Tuesday, August 14, 2001 at 23:33:28 (PDT)
Really enjoyed the website. Thank You!
MO USA - Saturday, August 11, 2001 at 15:44:05 (PDT)
Thanks! For years, I have intentionally done this without really understanding how or why it actually worked. But I can attest that it often does work! There is a two lane section of freeway that bunches up as it approaches the end of the freeway where a traffic light is. I've always slowed downed considerably and have found that I and the drivers behind me very rarely have to stop. We move slowly but we don't have to stop. It's cool! Try it yourself!
Frank Paiano <WonderNerd@nerds.com>
San Diego, CA USA - Thursday, August 09, 2001 at 20:50:15 (PDT)
I have noticed, a time or two, that some trucks pair together and form the rolling roadblock that you are talking about and that it seems to work like you said. I will try and follow some of your suggestions.
Steve Archer <firstname.lastname@example.org>
NY USA - Thursday, August 09, 2001 at 10:33:41 (PDT)
Sometimes its very easy to cut your time stuck in a traffic jam, stay in whatever lane that is closing or has the accident, this lane always tends to go faster than the other ones. Why? Because everyone is trying to merge into the other lane leaving less cars in the one that is closing, leaving more space to to speed up. Dont merge until you get to the very end, it might make people mad but they really have no choice but to let you over.
Dallas DeVries <email@example.com>
Troy, NY USA - Thursday, August 09, 2001 at 09:01:07 (PDT)
What a way to kill boredom. I have already realized most of this stuff through personal experiences every day, but it is still nice to read about it and about someone else having the same experience. Interesting that you mentioned the GPS, I actually work with a company researching in that. One day when they have automative drivers, traffic should be able to be fixed instantly. May the traffic be with you.
Douglas Lee <firstname.lastname@example.org>
- Wednesday, July 25, 2001 at 12:04:30 (PDT)
Great Website ! Im a young driver in UK, but do a 100 miles daily on the Motorway (Freeway). I use to be a "Racer" commuting to work. I've had the time to think what this has done to my driving situation.
1. Im not getting to work or home quicker during peak times as I may accelerate past one driver only to be slowed down again a few car lengths ahead i.e. No point doing it I just waste Petrol.
2. It is less stressful driving slightly slower than constantly thinking "I must take, I must take". Also their are always others driving faster than you, then a "Race really happens". Not good.
I guess the point Im making is the past 3 months my driving has changed and seeing this site has formalised what I'd been doing already. Having space in front and keeping a consistency and watching Traffic in front prevents you from getting involved in t
he Jam. This Works whether your in US or UK and I expect alot more Countries !!
London, UK - Tuesday, July 17, 2001 at 04:00:35 (PDT)
Your method is so obvious to anyone who takes time to think (about 0.1% of motorists). Here's a tip: If you are going to use this approach, do it behind a large truck. The bulkier the better. The sight of the large vehicle sends the message "this lane slowest" to the brain stem of the average motorist. This tends to offset the effect of the non vehicle-filled gap which sends the "this lane fastest" signal. In this way motorists are dissuaded from mindlessly filling in the gap. It also greatly reduces motorists' behavior where they try to make you go faster by either a) tailgating you very closely or if that doesn't work b) accellerating towards you, almost hitting your car, then dropping back a bit to begin again. They are focused on "must ge t around truck" instead of "must punish non-tailgater". If the truck is truly going a few MPH slower than the average traffic the effect is almost total. I find this situation the most preferable. It minimizes your exposure to hostility and takes the least effort while making very little difference in travel time.
A GUY <AGUY@justaguy.net>
Garfield Hts, OH USA - Monday, July 16, 2001 at 22:14:58 (PDT)
Having lived in California where everyone lives three cities away from where they work (the prices are all the same, but the people will tell you it's cheaper to live that way), wouldn't it just be more productive for people to live near where they work, and maybe just get up a little earlier and leave earlier? I know when I lived in Aurora Colorado, I would have to drive home in the a.m. rush, and it took up to 45 min for me to go the 3 exits that only took 10 min when I went to work at midnight the prev evening. as I only had three exits, I'd only move over one lane (so as not to have to cross back accross the blocking traffic to get off). I tried to get home by driving the city streets instead of the freeway, but since everyone waited till the last minute to leave for the job they didn't want to go to, it was even longer to go that way. on hte rare occassions I was able to leave even 5 min early, I could cut up to 30 min off this drive! maybe the REAL answer is for people to get jobs they LIKE, and work hours that their bodies can deal with. I am unable to work mornings...I just can't, so I always try to work a job that's graveyards. I just handle it better, and at least the right shift makes a job you hate better than hours you hate as well. What I noticed in California, everytime they added lanes to the freeways, they didn't break up the blockage because everyone thought they could leave later to get to work, since there would be more space and there wouldn't be traffic blocks. People just need to plan ahead better. Even here in small town USA we get the morning jams...but I found that if I left to get my daughter to school 5 min earlier, I got accross town without the problems. My mom found that if she didn't leave by exactly 7:45, she might as well take the longer (but faster by this time) backroads to work in the morning. Amazing what literally ONE minute can mean. In your area it probably wouldn't help you alone, unless evdryone using the highway woudl think the same way, but it's a thought. Leave early for work...what a concept!!! get a job youy like that has hours that work with your body's clock! HAH! what a radical thought!
No Traffic Here, ND USA - Thursday, July 12, 2001 at 07:42:44 (PDT)
These are the things I have realized myself while waiting in traffic. Leaving the space ahead of me and constantly moving.
It makes me even more angry when the solution to solving traffic problems is so easy yet people are to bull headed to take an extra 5 minutes out of their day to implement it. Never realising by reducing their speed and not tail gating they will actually
get home (or where ever) quicker!
Oh well that's life I suppose.
St.Louis, MO USA - Tuesday, June 19, 2001 at 12:26:16 (PDT)
Hey, we have the second worst traffic in the nation (only behind LA) and I can clearly see how your ideas would help alleviate some of the problems. Except on the beltway, if you drove anywhere below 20 mph over the speed limit when you didn't have to, you'd get shot by some a-hole behind you. Keep up the good work anyways. -MAS-
near Washington DC, NA USA - Friday, June 15, 2001 at 17:08:11 (PDT)
This is a superb site. Ive been thinking about this for a long time, but had never thought of doing experiments.
One added thought- in traffic that is not quite thick enough to truly stop a car, a traffic wave can travel forwards rather than backwards (if people are doing 60mph --> 40mph --> 60 mph, then the speed of the slowest speed (40mph) can overcome the time delay of reactions, etc. Ive noticed these waves travelling forwards on the highways near chicago
Noah Freeman <noah_freeman@ post.harvard.edu>
Boston, MA USA - Thursday, June 14, 2001 at 07:27:01 (PDT)
Absolutely great website, I'm glad I've found it again. I drive as seldom as possible, preferring to ride my bicycle, but when I drive in nasty traffic, I always keep it in first (maybe second) gear, no brakes, and no clutch, makes it feel like an automatic! What I've noticed a lot is the anger that is generated by the people who think I'm slowing them down, I really don't believe that we will be getting anywhere any quicker if we're all boxed in. My favourite time was crossing the 520 bridge, right next to a large semi, who're always accelerating slowly, the semi and I kept an enormous gap between us and the cars ahead. Of course this isn't the objective but it was humourous at the time. Driving is madness, I'm glad I've ran across this site, because it gives me a clear mind when I'm stuck in traffic. Where are we all going? . . . "from where we didn't want to be to where we didn't want to stay"
Michael Webber <email@example.com>
Seattle, WA USA - Sunday, June 10, 2001 at 22:23:09 (PDT)
After being rear end twice in my driving career I decided that I would not use the exuse that it was not my fault, and call it an unavoidable accident. I now use an improved method in my attempt to never be re-ended again. Maintain a distance with the vehicle in front that no matter how close the vehicle behind you is, you will be able to slow down at a rate that the vehicle driver behind you will not hit you. If the space that that you leave in front of your car bothers the vehicle behind you, they can pass you. You then revert to the same driving pattern as that you had before. Keep cool, smile and do not let your heart rate go up.
Manchester, WA USA - Wednesday, June 06, 2001 at 14:48:55 (PDT)
It was great to look at your web site. I have also driven in traffic where I kept a distance from the car ahead, and paced my speed to avoid braking and speeding up. The problem can sometims be the car in the lane next to me. They like to fill in the voids. I let them do it and then enjoy watching the lane that they came from start to speed up. It brings a smile to my face.
Manchester, WA USA - Wednesday, June 06, 2001 at 14:37:25 (PDT)
Two thumbs up! A terrific website! Thanks for enlightening us!
Cincinnati, OH USA - Thursday, May 03, 2001 at 09:56:29 (PDT)
Wow! I just stumbled on this site and am *so pleased* to see that I'm not the only person who realizes how traffic works! I've been ranting for years about "brakelight propagation" to anyone that will listen. I wrote my own simulator to prove it all mathmatically (I used pretty realistic actual physics in it for accel. and decel., braking distance, etc). I even traded in my automatic for a 5-speed just so that I could slow without braking... I've made a personal vow that if I ever win the lottery, I indend to use a part of that money to run "public service announcements" on TV to educate the masses about this. Anyway, I'm just pleased as punch to see that there are other enlightened individuals out there. Maybe we should arrange for a mailing list or something to share our findings, tools, and advice....? Any plans for updates to this site?
Brian Bloom <firstname.lastname@example.org>
Vancouver, WA USA - Wednesday, May 02, 2001 at 15:12:04 (PDT)
These techniques really do work. I've played with them for years, and first realized the how bigger spaces worked with my first car that had a manual transmission and a stiff clutch. It was the only way I could get through traffic, but it seemed to help everyone around me. Sometimes relaxing the space around my car causes other drivers around me to do the same, and it doesn't take long before we are all happier and even moving a little faster. It works!!! One other thing I would like to add here - traffic waves occur in lighter traffic as well. In light traffic the amplitude of the waves is smaller (cars don't cram as close together) but the waves are still there. You can actually use this to your advantage by driving a very steady speed. Within a few miles, you will often find that you are in a total clearing and have the whole road to yourself.
CA USA - Friday, April 27, 2001 at 13:42:19 (PDT)
I have been using "antitraffic" principles in heavy traffic for years. I must have read about it somewhere, don't recall where. It used to work fairly well, but recently I find it difficult to maintain a large enough space in front of my car. People behind see the space I am trying to maintain and assume I am a problem driver. They rush around me and fill up the space. When they do this, I slow down to create more space, and that makes people want to get around me even more. I guess the space creates a brief illusion that I'm going "too slow", when in fact I have merely evened out the stop and go into a steady pace. This is in addition to the impatient lane hoppers, who try to get through the impaction with lateral movement from lane to lane. They find my "buffer zone" irresistable, and fill it up quickly. They gain little advantage that way, but maybe they had too many coffees and simply CAN'T just sit there and wait in one lane. Truck drivers often use antitraffic techniques. Sometimes they follow me while I'm doing it, but that makes the "rush arounds" even worse. I guess thats because the people behind the truck can't see around it.
Los Angeles, CA USA - Friday, April 27, 2001 at 11:19:53 (PDT)
I hate to tell you this but this is not very original. Back in the early 1960s Popular Science had an article describing standing waves (in traffic) and I believe it was called something like Forces That Screw Up Your Driving or some such. It also included things like bow and stern shock waves from trucks etc. While I was at Illinois Institute of Technology I saw several Thesis on this and similer subjects both on caes and train traffic. Many people in urban areas try to break up standing waves to speed traffic. I have done so for years. In Chicago the road taxes don't go to the roads and as a result we have 39 of the 50 most overcrowded roads on the continent. Mike
Michael Ugorek <email@example.com>
Chicago, IL USA - Friday, April 27, 2001 at 10:41:35 (PDT)
I've been driving a long commute from Bellingham to Seattle and have noticing the very traffic phenomena you describe. If only everyone on the road were required to take a class in traffic dynamics to improve the situation! One observation that I have is that it seems to me that exits should always occur on the opposite side of the freeway, this would get rid of the need to merge and than backtrack over the same traffic to get off. The new rule would be that you can drive forward or go into the lane left of you but never to the right. This would make the traffic flow much more predictable. I would also get rid of HOV lanes, they are simply underutilized patches of freeway that could be carrying more cars than they do.
Jeremy Dunn <firstname.lastname@example.org>
Bellingham, WA USA - Tuesday, April 24, 2001 at 20:24:48 (PDT)
Thanks for a great insite into the physics of traffic jams. Being a regular 2 hours a day plus commuter, I too have spent many hours pondering the cause and effect of humans and motor vehicles. I would agree wholeheartedly with all your theories. The main problem here in the UK is the pure selfishness of other drivers. No one will let you in, out, across their path in case you hold them up for a nanosecond! I may persevere with some of your experiments. Great page. Regards Steve.
Steve Rogers <email@example.com>
Bracknell, UK - Friday, April 06, 2001 at 04:21:50 (PDT)
Dear Sir, I just happened to come across you page by accident. I have enjoyed your thoughts very much. I can see how it works and it is interesting. One Porblem in England apart from suffering the exact same problem is this one, At Peak times lots of traffic needs to exit from the motorway and then tails back on the the Motorway causing a as i call it reverse flow on to the Motorway, simply because the planers did not see the hign demand for the use of these exits at certain peroids. Naturally it is very hard bearing in mind we can eat into the Gren spaces we all naturally wish to keep, so I guess it is a read off. However I am sure that if we could use the emergency lanes at some of the exists with the assistance of the police I feel it would st least ease the problem, until someone does need the emergency lane. I would love to know what you feel. I also agree with you over the space in front. As I have worked as an instructor for racing schools and do find that speed is not dangerous if you know what you are doing , by that I mean you really do get to hear what the car is saying to you so you know when things are going to get out of hand and can driver fast without getting into a dangerous situation. As I have had to drive a car on the limit for the high speed rides at the end of a clients session with me and you can drive really fast and be safe. I promise you it can be done. As you can be easly tuned to the feedback from any car, to know what is happening. Anyway I look forward to reading more from you site. I do feel that people should give more space between the cars in front and not get bothered with people jumping infront. Yes that is easier said than done I do know what it is like! Just as you pointed out it would make driving safer and flow quicker too.
Thanks again for an interesting read on this grey saturday Morning
here in towcester England
I've been a hobby-type observation of traffic waves for many years and enjoyed Bill's comments on these pages. I recently ran across a news commentary on traffic simulations in Science (D. Helbing and M. Trieber, Science vol. 282, p. 2001-2003, Dec. 1998) describing in non-technical language some exciting new advances in computational fluid dynamics theory that pretty accurately describes on-ramp standing wave behavior. According to the writers of this article, a lot of excitement was generated by the theoretical work. Seems like the researchers have touched on a universally interesting and sensitive subject! Hey, if you're forced by the traffic to take time to contemplate the traffic...
Jim Remington <firstname.lastname@example.org>
Eugene, or USA - Sunday, February 18, 2001 at 13:08:16 (PST)
Further to my previous remarks; I've been doing an awful lot of motorway driving recently, during which I have applied the theories suggested here. They work remarkably well! I've noticed that by creating a large gap between me and the car in front even in traffic jams, a localised gap develops in all three lanes of the M6 around me (even when traffic is at a standstill). Giving myself more time and space to react I tend to feel less rushed, thus motorway driving has become an almost relaxing experience! Thanks for your insight, it's made my commuting life far more pleasant than it otherwise would have been.
Birmingham, UK - Thursday, February 15, 2001 at 15:16:28 (PST)
EXCITING INTERNET OPPORTUNITY!!!!! Just kidding. Very nice work here. Like the other manual shifters below I too have experienced the pleasure of brake-free driving. What's needed is more behavior research into brakelights, braking, etc. I wonder what the effect would be if everyone responded not to the brakelights in front them, but to the brakelights of the 2nd car in front of them...3rd car?...4th car?...Please direct me to the software that will enable me to model these behaviors. Thanks.
Chicago, IL USA - Wednesday, January 31, 2001 at 14:16:45 (PST)
USA - Friday, January 19, 2001 at 13:10:32 (PST)
Very good observations. We have a very heavily traveled freeway 696 - 5 lanes and I have observed the following. Cars leaving the freeway in the far right lane are applying brakes. the lane next to them and behind are applying brakes. This continues behind them for several miles until all cars come to a complete stop for no good reason. I believe this brake tapping when you see a red light is almost a subconcious reaction. I believe the only thing that will be able to cure this whole traffic problem is electronic control of the cars - speed and distance control by computer/radar system. Oh yes - more telecommuting wouldn't hurt.
Jim Gregerson <email@example.com>
Roseville, MI USA - Thursday, January 11, 2001 at 11:46:41 (PST)
This very subject was on Scientific American Frontiers a few months ago:
http://www.pbs.org/saf/ transcripts/transcript904 .htm#5
And here is the web site of the TRANSIMS project:
But neither of these sites is as good as this one!!!
http://www.geocities.com/phordee heres the link, thanks s
S <stillmefrombelow ifany1isinterested@ mywebpage.com>
LA, CA USA - Wednesday, January 03, 2001 at 09:58:40 (PST)
hats off to you sir. Great info to think about. I actually enjoy time spent in traffic, great way to really see human behavior at its best. The only downside about traffic for me is thinking about how much of my life will be spent in traffic. Just one of those things. Someone send this to the president, maybe he can do something with it. I have a parody of how to survive la traffic, please check it out, thank you for your time s
la, ca USA - Wednesday, January 03, 2001 at 09:56:41 (PST)
Very interesting page, I never thought of traffic like that. I wonder what patterns apply to large crowds of people as well... hmm.
Chicago, IL USA - Thursday, December 28, 2000 at 10:52:17 (PST)
The December issue of the Atlantic Monthly contains an article by Stephen Budiansky on this very topic. Fortunately it is now online as well: http://www.theatlantic.com/issues/2000/12/budiansky.htm. He discusses a series of journal articles publised by German theoretical physicists, suggesting that traffic jams can arise spontaneously. A very interesting read.
USA - Thursday, December 21, 2000 at 13:51:16 (PST)
My complements and appreciation; your observations reflect my experience. Here in New Jersey on Interstate 78 the commuting traffic moves at 75mph with little or no space between vehicles. I follow what I believe are your principles, which have the added benefit of reducing MY stress - I suspect it's a function of increasing reaction time available to me and therefore allowing my muscles to relax.
1. I maintain a gap of about two to three seconds between my vehicle and the vehicle ahead - the distance varies in proportion to speed and can be judged easily after practicing by timing the 1/10th mile markers at the highway's edge. The distance is adjusted for weather conditions and tailgaters (the latter to allow me to safely slow down since the tailgater is part of my "stopping-unit.");
2. always allow drivers to move into my lane, then readjust interval;
3. always signal my intent to turn or change lanes before assessing when and how to safely accomplish the maneuver - as opposed to the drivers who signal AS they begin to initiate the change;
4. at commuter-level densities, disregard the speed limit and move at the rate of the prevailing traffic;
5. when changing lanes toward the exits - as you noted - anticipatory changes into the exiting lanes must be adjusted to the traffic density. Never change lanes if it will cause the vehicle in the lane you are entering to need to brake. In dense traffic, signal the change, match the speed of the vehicles in the lane you intend to enter and enter behind the next vehicle in that lane that passes (this sometimes requires allowing more than one vehicle to pass, but that is where the anticipation reduces the
stress of I've-gotta-get-over-there.); and
6. If you are not comfortable with the prevailing traffic speed then move to the lane for the slowest traffic and pace the vehicle ahead of you. This is problematic as slow lanes have a habit of becoming exit-only lanes and on some roads (California Free
ways in particular) you will find lanes being added on your right as you pass entrances. California Freeways use the lane dividing marking as indicators, telling you if the lane you are in is changing to an exit only.
Thanks for getting me thinking about this topic. While human nature is a bell-curve, these methods can affect the quality of life for those participating, not to mention our increased safety and that of our passengers.
I'm doing Cellular Automaton model simulations of traffic flow as a stundent at the Departament of Physics at URI. Some of the results I have come to in the past months support your many of your ideas. ie. as maximum velocity increases, the critical densty for maximum flow decreases. Therefore, you are supported through tested simulations in saying that, "moving slower in dense traffic will increase overall flow." i notice many people refer to the fact that people will rush to get in front of you if you leave space. i'd like to agree with the important point that you make in response; not everyone drives super aggresively. if you wait long enough you will be driving parallel with drivers as patient as yourself, thus blocking aggressive drivers from continuing to foil your scheme of maximizing traffic flow. this sheds new light on the turtle and the rabbit story. how quickly we forget so many things are we learn as children. if we just remembered so many of those simple rules, life would surly be easier.
kingston, RI USA - Thursday, December 14, 2000 at 10:22:25 (PST)
Thank you for your article on traffic. It is amazing that you actually took time to study it. I can't wait for my next traffic jam (read tonight) so I can test out your theories.
atlanta, ga USA - Monday, November 27, 2000 at 10:28:21 (PST)
Interesting. Many years ago I read an article in New Scientist by Tom Margerison regarding the wave theory as applied to traffic flow. He suggested a scheme to use traffic lights to control the flow of traffic which also allowed vehicles to merge from side roads without disrupting the flow. It seemed like a very good theory at the time - shame it wasn't put into practice. I've also noticed a similar phenomenon when driving into the mountains, particularly during the skiing season. Traffic crawls for miles on a two lane highway. The only explaination I can come up with is that it is related to traffic slowing to allow the occasional vehical to join or leave the highway. It appears to be a driver behaviour problem rather than a capacity problem.
Bob Starr <firstname.lastname@example.org>
Denver, CO USA - Tuesday, November 21, 2000 at 12:41:46 (PST)
Absolutely spot on. This works even in the UK - I have tried to practice this same approach. Actually it works even if the traffic is flowing at a reasonable speed. Leaving enough space to brake and the same again means quite often means you never have to brake as you can just slow down into the space. This saves the brakes and keeps the traffice moving.
If traffic has come to a standstill - at a junction say - I have always applied the rules of a) trying to keeping the car rolling (however slow that might be) and b) if there is another filter let a car go and then move forward yourself this keeps traffic
flowing all round.
If you are unlucky to be stuck in a traffic jam my observations suggest that what is normally the slow lane moves fastest. This is for two reasons. Firstly you get cars exiting from this lane so the volume of traffic can be reduced. Secondly when "rubb
ernecking" takes place everyone in this lane moves at a steady speed whilst the faster lanes accelerate and brake - ultimately going slower.
As a final thought why do communters take part in runnernecking? We are all trying to get from A to B as fast as possible why add a delay.
The wave theory is spot on, but unfortunately your remedy is completely impractical.
1) The problem cannot be solved by slowing yourself, because other drivers
will fill in any gaps (just as any fluid would). Human nature is how it
is. You cannot hope to change it.
2) How do you know whether the traffic is going to bunch again? I've come
across many people (especially when I lived in LA) who would let large
gaps open out in front of them in order to avoid the constant stop/start
cycle. However, once the hold up has passed, these people still refuse to
accelerate, and thus perpetuate the jam for everyone behind them.
3) Jams can cause delays not only to a single road, but to any that merge
with it or cross over it. This affects many people who are using
completely different routes. If everyone adopted your system, a short
section of seriously jammed road would become a longer stretch of road
travelling at a low speed. This would still cause (smaller) delays to
surrounding roads, but over a much wider area. Basically your solution
would inconvenience a greater number of people.
I suggest you try adding a bit of realism to your theories -a naive idealistic simulation has no practical application.
[Hi Mal! I've been following my own suggestions for years. These are not "theories." These suggestions are what I use every day during commutes. I've watched them work time and again. You can theorize that they will fail, but I have real-world experience that they do not. Also, please see the FAQ, which contains responses to these objections (which have been raised many times before.) -billb]
Mal Lansell <email@example.com>
London, England - Tuesday, November 21, 2000 at 08:20:04 (PST)
It's great that you've researched this so well. I only passed my driving test a few months ago and I observed that a combination of speeding, poor driving, and irresponsible road use seems to cause traffic jams (e.g. a few weeks ago we had a fuel cris is in the UK, what did everybody do? get on their bikes and conserve fuel? no! they queued for hours at filling stations to get the last available drops!). The problem in the UK is exacerbated by the fact that nobody actually knows how to drive on motorw ays (there isn't a separate test so very few people get proper tuition). Add that to speeding and tiredness and UK Motorways are a very dangerous place to be.
Birmingham, UK - Tuesday, November 21, 2000 at 05:34:03 (PST)
The BBC recently ran your theory on their web site. Although i understand your argument about slowing down - and have observed this effect myself, in the UK lane discipline would simply nullify the effect. As soon as a gap opens up in front of your car, those to either side will fill it in and eventually your speed will continue to slow in order to clear the gap in front of you. Furthermore those behind you - sensing the gap to your front - will overtake to get ahead. As I say lane discipline in the UK is non-existant. It is common to find the 'fast'lane being the slow lane and the speeding traffic to be found in the 'slow' lane from which those drivers - observing the rule to overtake on the right - have pulled out of in a vain attempt to overtake.
[Hi Pete! It is certainly possible that driving behavior is different in the UK. However, drivers everywhere will change lanes to fill a space. Therefore my ideas cannot work! Yet they do. Why? Try them during a commute, and you'll see. Or read the FAQ, which explains why lane-jumping drivers are not as big a problem as you might expect. -billb]
pete e <firstname.lastname@example.org>
UK - Tuesday, November 21, 2000 at 05:11:10 (PST)
I noticed the same things a few months ago while commuting to school every day. However, your analysis is much more complete. I think you have done a great job noticing this!
James F. Cerra <email@example.com>
Pittsburgh, PA USA - Sunday, November 19, 2000 at 19:38:15 (PST)
Re: Rolling barrier of State Troopers.
I travel a lot on the "Autostrade" in Italy. Occasionally I've seen a police car join the motorway, take a central lane position, with it's blue lights flashing, and slow the traffic to about 30 mph. The mystery is solved when about 20 minutes down the ro ad we arrive at the scene of an accident, which has just been cleared. Result: no queueing! Maybe somebody should tell the highway police in other countries?
DOMINIC RICCI <DOMINIC@ DRICCI.DEMON.CO.UK>
CARDIFF, WALES - Friday, November 10, 2000 at 02:54:02 (PST)
You can engage in experiments while in traffic too. Try swerving left and right, accelarating hard toward the car in front of you, and slamming on your brakes. Note the resultant behaviors of cars all around you, just like the particle physics of overheated liquids. No cell phone necessary.
Pj the mad scientist <madscienceinaction @hotmail.com>
San Francisco, CA USA - Friday, November 03, 2000 at 21:25:16 (PST)
Interesting Stuff. I commute on the M6 Motorway in northwest England, which is like a US Interstate only with twice the traffic in half the space, and i see all kinds of traffic jams.
There is some sort of crash or problem almost every single day, and they
try to slow the oncoming trafic with those flashing overhead signs which
everyone ignores. Lately they have started installing speed cameras on
those stretches where there is some sort of temporary speed limit due to
roadworks. You can't ignore these, and they seem to keep things moving
reasonably well when traffic is heavy.
Of course when the traffic is only moderate people just slam on the brakes
when they hit the camera zone then speed off again - in this scenario the
cameras maybe cause rather than prevent jams. Around London they're
introducing some sort of system which measures your average speed between
fixed points, which will probably work better.
Thing is, these cameras are about as popular as the bubonic plague,
because everyone thinks they're some cynical revenue-raising scheme. Maybe
if it was explained that they prevent jams drivers would be more
The referenced URL is http://www.media.mit.edu/starlogo/
Seattle, WA USA - Friday, October 06, 2000 at 19:36:16 (PDT)
Those interested in modeling traffic jams may want to surf to the referenced URL, download the free StarLogo code and read "Turtles, Termites, and Traffic Jams," MIT press(1994/98) by Mitchel Resnick.
Seattle, WA USA - Friday, October 06, 2000 at 19:33:50 (PDT)
Augusta, ME USA - Tuesday, October 03, 2000 at 12:39:43 (PDT)
Traffic flow was explained at VMI by physics head as similar to a pendulum. You can in fact travel in the dead spot for awhile if you pay attention. Yours is another good way of looking at the problem.
Tom Wilson <firstname.lastname@example.org>
ca USA - Monday, October 02, 2000 at 15:14:37 (PDT)
Cool theory. I wish they taught this stuff in traffic school or in my physics class. Also, I've seen some truck drivers employ the same method the FAQ described state troopers using. If there is a major backup on a two-lane highway, sometimes two trucks will pull up along each other and block the traffic behind them. They'll go slowly and let a large gap build up in front of them. It's the whole anti-traffic thing in effect. Maybe truck drivers are smarter than people think!
Jennetti Spaghetti <pastaqueen2[nospam] @hotmail.com>
Lousiville, KY USA - Monday, September 25, 2000 at 23:39:57 (PDT)
Your argument is flawed I am afraid. Leaving longer means traffic moves slower. Imagine a que of cars waiting to get on the motor way, the larger the gap required before entering the motorway the longer it takes everyone to get on. Also imagine merging traffic, the only way you can make a gap is to slow down. Sorry to pee on your parade.
NZ - Sunday, September 24, 2000 at 22:49:25 (PDT)
The entire idea is so amusingly simple. I love it. Who would have thought driving patiently and with concern for proper merging tecniques that we could have eliminated traffic jams, obviously not hte average joe. + Do you mind if I print this off and mass snail mail every one in my city, because I have already emailed it to every one I know and about a hundred I don't, haha. I just love it!
dayton, OH USA - Monday, September 11, 2000 at 12:04:40 (PDT)
What an excellent discourse on traffic dynamics - and remedies. The essence of this phenomenon should be distilled and incorporated in every driver training course! I have browsed only ~20% of comments, but see no reference to other real benefits of steady driving - improved gas mileage and longer brake life. Stop-and-roll eats gas and heats brakes.
This will be lost on the whizzing idiots who failed, or never
physics. Only with their first accident will they perhaps dimly grasp the
concept of inertia and the fact that, Fitzgerald contraction not
withstanding, a speeding automobile gets longer in terms of its physical
length PLUS the space it WILL occupy, or earnestly try to, in the next few
seconds, no matter how good the braking system and the driver's reflexes.
Perhaps if Public TV were to do a special . . . ?
Civilization ended when the automobile became the focus of our society. Please do not try to alleviate traffic jams; they use up gas! The sooner all the gas is gone the sooner we can live in a rational world again. Thanks!
WHIT WALTMAN <YAHOO@AOL.MSN.COM>
VALLEJO, CA USA - Saturday, August 26, 2000 at 18:33:28 (PDT)
Excellent article! I wrote something similar in 1995 (http://reality.sculptors.com/ ~salsbury/Articles/traffic.busting), and always wanted to add diagrams, animations, etc., to it, but never did get around to expanding the original article. Your page is wonderful, and really helps to convey the idea of how the traffic flow works. I also run a mailing list about autopilots for vehicles, and would encourage interested folks to join the list. Info is available at http://reality.sculptors.com /lists.html Keep up the great work!
Patrick Salsbury <email@example.com>
Santa Cruz, CA USA - Friday, August 25, 2000 at 19:47:58 (PDT)
Your observations are dead on. Not suprising that you are getting contrary comments. "Slow down, you'll get there faster" is absolutly true, regardless of how counter-intuitive it may seem. Three additional things: 1)You can reduce the waves in city traffic on streets with sychronized signals by practicing the same techniques as you use to reduce waves on the freeway. In this case it should be obvious to the nay sayers that getting to the light at 35 MPH just as it turns green is a lot faster than driving 45, and being at a dead stop when it turns green. In spite of this, I will often pass the same "hurry up and wait" fellow 3-4 times, and he still won't catch on. 2) On the merging problem: You can reduce _Your_ wait, as well as helping those behind you, by using a different technique than you suggest. The trick is to stay in the lane that ends untill very near the "pinch point" but drive the same speed as the lane that continues. Because much of the slowdown is caused by drivers that run up to the front of the line, then force thier way in, you can break a whole jam loose by doing this. Truckers seem to be savey to the technique, and will let you in when you get up to the pinch point. B y blocking the route for the problem drivers, the jam will quickly dissipate. Warning: The above is nearly gaurenteed to piss off anyone driving a camareo of corvette behind you. Even in moderte traffic, I try to pick my hole early, but pace along with it untill the last minute. 3) Large trucks, due to thier high inertia, and highly experienced drivers will naturally tend to "average out" traffic waves. If you operate in an adjacent lane, then you can leverage this natural tendancy to produce "anti-traffic" in two lanes. This i s also a non-linear effect, as having two adjacent smooth lanes tends to improve traffic much more than twice as much as having just one. -KF-
Kevin Ferguson <firstname.lastname@example.org>
Albuquerque, NM USA - Monday, August 21, 2000 at 09:51:50 (PDT)
I pretty much arrived at your same conclusions and even noticed that I could eliminate traffic wave my moving at a slower speed. I'm still not sure if I agree that the overall throughput would be greater if volume were moderated. Cars would spend les s time on the road between points A & B, however fewer cars would also be allowed. Throughput aside it would certainly be less frustrating then sitting in traffic at a stand still.
USA - Friday, August 04, 2000 at 13:25:02 (PDT)
I liked your idea of state troopers equalizing the traffic flow - but how about speed and traffic density sensors placed in the road and networked to "optimum speed" traffic signs placed strategically on the highway just like speed limit signs. This w ay the road conditions are self regulated by the drivers - who would obey the 'optimum speed limit' signs since by doing so they know they would get there faster.
Joe Sixpack <email@example.com>
Seattle, WA USA - Thursday, August 03, 2000 at 00:14:43 (PDT)
Excellent paper. I've tried doing this since I started working in the business park. One suggestion though, instead of a wall of police cars, sprinkle patrol cars down the chain. I noticed that they have a tendency to reduce traffic speed to whatever t hey are doing. It may not work completely but would help.
Durham, NC USA - Tuesday, July 11, 2000 at 09:07:32 (PDT)
I've also noticed that in cities with a larger number of aggressive drivers (LA particularly) who switch lanes frequently, there often arise left-to-right waves of traffic which seem to be roughly in phase with the occurrence of on and off-ramps. I'm s ure these also tend to slow down traffic as a whole and tend to eliminate the gains of leaving gaps in front of you.
Matt Hine <matt@hineDELETE.cx>
Austin, TX USA - Tuesday, July 11, 2000 at 08:09:45 (PDT)
Your theories are completely false. They are based on EVERYONE on the highway being courteous. Since there are a large number of people just weaving back and forth just to get that "one car ahead", no amount of space can help...they will just fill th at space until you end up stopped while everyone goes around. If everyone on the highway were courteous, we would have smooth flowing traffic. Nice try, but you have to add the A@@ H#$% factor into your calculations.
Nampa, ID USA - Tuesday, July 11, 2000 at 07:38:45 (PDT)
Be Especially Kind to Trucks. As mentioned above, the laws of physics compel big rigs to employ "smooth" driving practices to the N'th degree. I would just like to add that it's good karma to help them out as much as possible. Just for starters, be aware of their visibility limitati ons and strive to always remain visible. One good rule of thumb is "never drive next to a truck." In other words, stay behind it or pass promptly. Maintaining this space is also helpful for them when they need to change lanes. Finally, if a truck want s to be in your lane, drop back and turn your lights on and off to let them know you're letting them in, they'll appreciate it.
Mike Schmelzer <firstname.lastname@example.org>
Washington, DC USA - Tuesday, July 11, 2000 at 07:31:18 (PDT)
It gives me a lot to think about...when stuck in traffic. With no basis in fact, I always told my wife that the waves were examples of "Brownian Motion" just because I liked saying it. At first she thought I said "Brownie in Motion," so it's become a r unning joke. My question: I know there are people who analyze these things for a living; what kind of feedback have you received from them?
Bob DuCharme <see web site>
New York, NY USA - Tuesday, July 11, 2000 at 07:17:55 (PDT)
An extremely interesting article. I actually started doing this about 10 years ago, when it just made sense to me that if I could manage to not have to stop in a full-stop traffice jam, it would ease the condition just that much. I never really analyze d it, though, which is rather ironic since wave theory is something I'm quite familiar with as an old Navy Radioman. I got an idea reading this, though. We should have "traffic busters" -- vehicles with specially-trained drivers, in communication with eac h other and perhaps with a traffic chopper, who roam the highways and break up traffic jams. Perhaps they could be specially marked vehicles, and it would be illegal to pass one when it had its lights flashing, or something like that.
Alan Little <alanATholotechDOTnet>
Ann Arbor, MI USA - Tuesday, July 11, 2000 at 05:48:34 (PDT)
This is absolutely wonderful !
You have made an independant discovery of a profoundly counter-intuitive theory of systems-scheduling.
The theory is called Theory Of Constraints - was prompted by the exploration of an Israeli physicist into production line scheduling.
Good introductory book: "The Goal" - Eliyahuh Goldratt, you can get it from Amazon.
Or you can look at the organisation he founded to promote his ideas: http://www.goldratt.com - follow the 'library' link for some interesting articles.
Check it out, I think you may find it quite interesting and I would love to hear your comments !
Antoine van Gelder <antoineATegeneticsDOTcom>
Cape Town, South Africa - Tuesday, July 11, 2000 at 01:56:41 (PDT)
Very interesting info on traffic waves. I have personally seen a "moving wall" of hiway patrol cars act as an effective deterrent to what would have almost certainly become a standing wave. This was done along a section of freeway that had very few onr amps (The I-15 south of Temecula in the mid 80's). The wreck was cleared before we got to it, thus denying rubber neckers a chance to ruin peoples day. I am not sure if this saved us any time but there was not the feeling of frustration that comes from sl owing to a crawl or stopping. The feeling was one of resignation due to the fact that because the hiway patrol was conducting the slowdown there was nothing one could do about it. Anyway, I drive a stick shift vehicle and consequently try to maintain a constant speed whenever possible. While I waiting on I-25 in Denver one day I had an idea. I was wondering if automatic transmissions actually contribute to stop and go driving by re ducing the driver's incentive to maintain a constant speed. Perhaps comparing traffic behavior in countries with different ratios of automatic vehicles could bear some fruit. Warning! Contemplating about traffic while in traffic can worsen traffic.
Lee Thornhill <email@example.com>
Boulder, CO USA - Tuesday, July 11, 2000 at 01:16:18 (PDT)
cool idea, but one fatal flaw. i do NOT want traffic to have a "mild slowdown for 40 miles" because in most cases i am only travelling 10. in other words, everyone bunched up at the front of the jam is CONSIDERATE to those of us who are getting off the freeway well before we get to the jam! so, by creating a traffic jam, those people are making traffic much better for those behing them. >;-)
USA - Tuesday, July 11, 2000 at 01:09:10 (PDT)
A very good presentation of interesting facts. I had read about these phenomena, however, several years ago. A whole "science" of traffic management has already evolved to deal with the issues of the "fluid dynamic" nature of road traffic. Interestingl y, in the UK, on some motorways there are automated traffic jam wave busters! They consist o electronic speed limit signs that ajust automatically to lower the speed limit when a jam is detected ahead. This has the effect of forcing all drivers to become slower without decreasing their gaps, thereby breaking the jam.
London, UK - Tuesday, July 11, 2000 at 00:24:25 (PDT)
I enjoyed reading your observations. They roughly parallel observations that I have made during the last 23 years of awful "Silicon Valley" commutes. My conclusion has been that traffic flow will greatly improve if drivers can learn to drive smoothly at the "best speed" for freeway conditions, rather than driving as fast as they can. Thanks for a good read. Art
Art Hicks <firstname.lastname@example.org>
San Jose, CA USA - Monday, July 10, 2000 at 21:36:57 (PDT)
Real nice observations. Have you realised that the unregulated traffic flow of a highway system behaves very similarly to an Ethernet network? As the traffic density approaches the nominal capacity of the carrier, the random (unpredictable) behaviour of particles (packets or cars) cause local ripples and jams in the traffic, so this kind system is unpredictable when it operates close to its capacity. It can even come to a standstill. Your idea of State Troopers makes the system work like a Token Ring network. There, traffic is partiotioned into units (one packet per unit on Token Ring, one or more cars between two State Trooper barriers on the highway) and the the unpredictability of individual elements is constrained by the partitions. This kind of network can always deliver its full capacity.
Laszlo O'Vari <Laszlo.Ovari@prophecy-open.com.au>
Adelaide, SA Australia - Monday, July 10, 2000 at 20:47:54 (PDT)
I have just moved to Florida about a year ago, and just last week I had the opportunity to drive I4 during rush hour. I had noticed that everyone was braking and stoping, while I was leaving enough space ahead of me to not have to hit the breaks. I w as able to cruise through 60% of the traffic jam without hitting my brakes while the other lanes were starting and stopping. I was curious how it would affect the traffic in the greater scheme of things. Thank you for taking the time to document this, s aving me lots of observation time!
Aaron Gerstenkorn <email@example.com>
Orlando, FL USA - Monday, July 10, 2000 at 20:27:49 (PDT)
Beutifull observation and explanation that I cannot disagree with! Thank you! I am very interested in your shared thoughts on "Chronic Left-Laners" (Those of us who for no obvious reason, will make a mad lateral dash to the Left or inside lane and sta y there) Particularly Left Lane "Hoosiers" who will speed up or slow their rate of travel or "pace" adjacent trafic in an effort to prevent those who follow passage or "right-of-way". Could it be genetics or is it chemical imbalance? Possibly something in the water. We may never know but I belive the subject warrants discussion. Looking forward to your hypotheses. ;-)
Louisville, KY USA - Saturday, July 08, 2000 at 13:25:34 (PDT)
When I was about 8 years old, I first began to think about the flow of traffic. I clearly remember sitting in the back seat behind my father at a red light. He was fuming. The light turned green and as the cars at the head of the line began to move but we did not, I said to my dad, "Why doesn't everyone just move ahead at the same time when the light turns green, and then we wouldn't have to sit here and wait for our turn?" My father answered in his usual clear way, "BECAUSE IT DOESN'T WORK THAT WA Y!" Ok... Anyway, I too leave good size spaces in front of me when driving. One reason is for safety... I learned in driver's school to leave one car length open per 10 miles of speed, so if I'm moving at 50 mph I leave 5 car lentgths open in front, etc . And if someone moves into that space, I back off further. It's no big deal. But I also do it for the very reasons you identified. And it works pretty good most of the time. It just never occurred to me that it could open the jam for everyone behind me too. I gotta' add... one way I love to mess with people is, while sitting at a red light, move my car up a couple feet. Invariably the cars behind me all move up a couple feet. I can do it one or two more times and sucker them in. People do it wit hout thinking. In the reverse, when I'm at a red light and someone pulls up a few feet, it is very hard to resist the temptation not to move up also, especially if it's a car length or more. It always feels unnatural to leave open space in front wheneve r at any kind of stop, but especially on the freeway. I sure enjoyed your website. Thanks for identifying many thoughts I've had but never thought out clearly.
CA USA - Sunday, June 25, 2000 at 12:57:00 (PDT)
In 1957 Lighthill and Witham (Mathematics, Oxford, UK) identified traffic patterns on the newly opened London-Manchester Highway M1 as "kinematic waves" and analysed them [Proceedings of the Royal Society 1957]. In 1958 Professor Sir Charles Frank (191 1-98), my PhD Supervisor in Bristol (1957-60), applied this mathematics to the way atomic steps grow and move on the surface of a growing crystal. Atomic steps are the counterparts of automobiles. An impurity atom (or a group of impurity atoms) is the cou nterpart of an accident or whatever it is that leads to transient jams that give rise to traffic waves, the road with automobiles is the counterpart of the medium in physical waves ... and so on. The waves are termed "kinematic", not "dynamic", because no physical forces act between parts of the medium. Besides giving this extra information to you and visitors of your site let me say that I enjoyed reading your whole site. I have driven extensively in India, UK, USA and Canada but have never come across the "merge, let merge, and move faster" pattern. Is it only your idea or does the pattern actually take place?
(Prof) Arawind S Parasnis <firstname.lastname@example.org>
Pune 411007, MH India - Wednesday, May 31, 2000 at 18:08:07 (PDT)
I had to laugh when I saw your site! As an engineer, I have observed the way the traffic "packs" and "unpacks" as various obstacles or events interrupted what was a relatively steady flow. I always called it a "ripple effect". I worked out the algor ithms for it in my head, and figured there must be some obscure governemnt agency that has done the same. I never reached the point of utter boredom you must have reached that led you to create this site with it's amusing graphics. Thank you!
Tom Spivey <email@example.com>
West Linn, OR USA - Tuesday, May 16, 2000 at 23:44:00 (PDT)
Your roving line of state troopers idea seems to have a fatal failing-- all the people behind the troopers would bunch up into this steady, but slow moving, pattern with no spaces behind them... they would feel that the speed isnt changing any time soo n so they could comfortably bunch up and then when the troopers left you would have a traffic wave behind them compressing... maybe some big signs on the back of the troopers or a new state law: when the troopers are sweeping you must leave a 2 length or more space between cars.
Chicago, Il USA - Thursday, April 27, 2000 at 15:10:37 (PDT)
Thanks for the very insiteful analysis presented here. I have noticed a number of them in the past myself. All of the examples you present here are of compressional waves. I thought you might be interested in a lateral traffic wave I encountered a fe w years ago. On a 2 lane road in the country with continuous traffic, I could see traffic ahead of me braking and then driving onto the shoulder, apparently to avoid some obstruction. As it did so they kicked up a cloud of gravel dust which obliterated the view ahead . Of course I just followed the car ahead of me, onto the shoulder and into the cloud of dust. I was astonished when the traffic started moving back onto the pavement although there was no obstruction of any type on the road we had bypassed. My theory is it started as a rubbernecker type wave moving around an accident or obstruction but the cloud of dust created obscured the view even after the accident was cleared and maintained the wave. From my perspective as a driver, I don't know whether it was a standing wave or moving backwards, but it could have persisted for hours after the obstruction was removed for all I know!
John Thompson <firstname.lastname@example.org>
Ottawa, ON Canada - Friday, April 07, 2000 at 14:05:05 (PDT)
Washington D.C. (and surrounding beltway area) has the worst traffic jams in the country, second only to Los Angeles. The highway system is way overcrowded and even secondary boulevards and county routes are congested far beyond their 2 or 4 lane capa cities. I have practiced the traffic-wave dampening techniques mentioned here with some degree of positive effect. I would typically find a fellow car (in the second lane of a two lane-divided highway) travelling slower than the rest of the pack and pac e them for a spell. This sometimes builds up frustrated drivers behind me, who angrily fly past when the barricade inevitably falls apart. The ratio of "type-A" aggressive drivers to "type-B" defensive drivers weighs heavily in favor of the reckless, fast-track types heading for their high-powered careers in DC and surrounding areas. The whole region for a thirty mile radius from DC is like this and it's a battlefield out here! The best thing would be if the local law enforcement agencies read this material and experimented with dampening the shock waves of traffic around the Captial Beltway to see if the jams would smooth out. During rush hour, I have yet to see a police car o n the road pulling someone over for speeding or driving recklessly. Rather, the cops only show up after there's been an accident, which is too late. There are also traffic light timings that could be improved to allow traffic to flow completely out of Tysons Corner (a booming high-tech business strip a mile long in northern Virginia) and let the general populace get out of DC in a more regulated fashi on. We routinely take 20 minutes to travel a mile or less across this stretch of otherwise fast road during 5:00 rush hour. The civil engineers who designed the highways here need to pay attention to the root causes they have built into their highway systems and how to correct them. In the meantime, the aggressive drivers we contend with need to develop better patience and if they are driven to act, they should act as the traffic flow regulators to effect a positive change on the traffic tie-ups. -Ticked Off Techie in DC-
David Holt <email@example.com>
Falls Church, VA USA - Wednesday, April 05, 2000 at 11:35:07 (PDT)
It is with fascination and chagrin that I notice an inverse application of Bernoulli's principle when applied to freeway traffic dynamics.
James Mize <firstname.lastname@example.org>
seattle, wa USA - Wednesday, March 22, 2000 at 20:45:15 (PST)
I have observed the same motion in cars, i have found that it the the driver who is in a hurry who pulls into your lane, causing you to brake ( to keep a safe braking distance), which then starts the flow back. It also happens aon slow moving traffic drivers see the other lane moving quicker and so a load o cars move ito that lane, hoping to reap the benifits, but al that happens is that they cause the traffic to slow as they force gaps into the traffic to let them in. What then happens is the lane th ey all came out now has fewer cars and so can close the gaps giving the impression its speeding up, and what happens then all those that moved lane move back again causing this slow occilation to happen in the lanes of traffic. If drivers stayed in there
Z Cheema <email@example.com>
UK - Monday, March 20, 2000 at 13:22:14 (PST)
Nothing could be worse than living in a city that is getting ready to host the 2002 Winter Olympics! You can't spit around here without hitting a piece of construction equipment! Most of the construction is being done on our highway systems. Our traf fic commutes are legendary for MISERY! Yet, every time my family or I have driven the 'smooooooth creep' home, we have witnessed this phenomonen! Come on people! Do your own empirical research. Meanwhile, without a doubt, we KNOW that a spirit of con sideration and safer driving spreads like a wave too! When this happens, it's nice to look at people doing nice things for others all the way home! Makes you think mankind might be worth saving after all.
'Well-Seasoned Traffic Jammer" <firstname.lastname@example.org>
Salt Lake City, UT USA - Thursday, March 02, 2000 at 02:53:02 (PST)
btw - that java simulator seems to work smoothest at lower res (800x600)
- Monday, February 28, 2000 at 14:21:36 (PST)
I liked the site! I've observed the traffic wave effect myself and I reckon the factors as being speed, distance and reaction time - I guess we're stuck with human reaction time though... I've seen efforts on the British highways to reduce this effect - the electronic variable speed limit on the M25 as mentioned below (It's enforced, by the way by automatic speed cameras behind each sign) and a scheme on the M1 where chevrons are painted on the road surface (say 100yds apart) - and the drivers are inst ructed by roadsigns to leave 3 chevrons between themselves and the vehicle in front, both seem to have improved matters (but it's a couple of years since I used either road) - the wave like behaviour of traffic is not only frustrating but dangerous and ca uses you to burn off more fuel and wear your brakes and tyres with pointless stop-starting. there's an interesting simulation of the traffic wave effect (for java enabled browsers) here (bu t only driver reaction time and speed are variable)
Liverpool, England - Monday, February 28, 2000 at 14:17:32 (PST)
My compliments man, You make a lot of good points. I've driven in some of the worst traffic in the country, and the fact is, there is little or no reason for the "wave" phenomenon here in Washington or anywhere else, except for poor driving habits. I have always driven in th e manner you describe. Not necessarily in the form of any experiment, but for the sake of safety. I for one, attended driving school before I recieved my driver's license some 17 years ago, and certain rules I learned, I have always followed. One of them being, to maintain two carlengths between you and the car in front of you. This fact doesn't change just because the traffic has come to a halt or slowed down for one reason or another. If people drove with the slightest bit of safety skill, and kept thei r distance from the car in front, it would be as you say, and there would be very few of these "waves". although I've never noticed the effect my driving was having on the traffic behind me. I have although observed with much humor, the anxiety it causes to some unskilled drivers. It cracks me up to no end to watch some idiot freaking out behind me, because I won't "catch up" to the car in front of me. Sometimes it overwhelms them to the point that they have to roar out and around me just to catch up to t hat other car. This effect is quite entertaining to me. You just know they are flabbergasted at the fact that there are cars just changing lanes at will in front of me....LOL....The funniest thing is, in my observation of the usual traffic patterns on I5, I have discovered a perfect path to take into Seattle. No matter what else seems to go on, I have a set pattern for lane changes, and I just drive along in these lanes no matter what, ignoring the lane changing idiots trying to get that extra car length that I or anyone else even cares about. In doing this, because of the normal waves that take place in the same areas every morning, I am in the correct lane when they do come and I go right by all the ding dongs who raced around me when I was only doing 7 5 when the traffic opened up. As I said, it is quite entertaining to me more than anything. I think it's hilarious how most people would just follow the herd right into the slaughter chute, without looking 50 feet ahead and watching what's going on. CATCH UP!!! they cry as they are led right in....LOL.....I used to get frustrated because I, being the courteous person I am, let people entering the highway into the lane without a thought. That's what I was taught. But when it came time for me to get over, s ome idiot would actually speed up so I couldn't. Now I just drive like I learned in NJ, you put on your blinker and change lanes, especially when cars cost 20, 30 thousand dollars...they make room. My little car has 250,000 miles on it and it's worth abou t 50 bucks, so what do I care. Besides I had two different idiots put nice dents in either side of my car (neither was my fault) and I left them there because mainly I couldn't do without my car, but I think it has the added effect of warning the drivers of the pretty cars that I'm coming over whether they like it or not. If they have to slam on the brakes well it's THEIR fault not mine. When someone puts on their turn signal in ongoing traffic it's not to ask permission, but to tell the other drivers of their intention. When someone speeds up to prevent a car from "getting in", that is the offense not the changing of lanes. (except of course in the case of the habitual lane changers who are just a pain in the ass trying to get 10 feet further, but of cou rse if you leave a proper gap it doesn't matter). Well I could go on and on, but I just wanted to let you know you're not alone. If everyone would just drive safely and not try to be their favorite NASCAR hero, (how fucking stupid is that?)traffic would be so much smoother. Personally I think it's tiny p enis syndrome that makes most of them drive as they do. Combined with tiny brain syndrome you have a dangerous recipe for traffic. Sincerly, Fixer Good
Fixer Good <email@example.com>
Tacoma, WA USA - Saturday, February 26, 2000 at 11:39:39 (PST)
Right on the money. I came to the same conclusion when I used to commute on Rte. 30 (Massachusetts state route), eastbound into Newton. There's an area with a long, winding downgrade followed by a level and a slight upgrade into an intersection with a traffic light. At morning rush hour there would be 5- to 10-minute backups extending all the way back up the hill, with waves of "move ahead, slow to 1 mph" flowing back from the light. I got tired of changing speed, so I would leave a gap when the car ahead of me started up again, and just coast forward in neutral, on gravity or momentum. I tried to time myself to close the gap to the car ahead of me just as it started moving forward ag ain. While I couldn't see very far behind me because of the grade and the turns, I know that the cars close behind me also shared my slow but steady motion.
Mark A. Mandel <Mark_Mandel@dragonsys.com>
Framingham, ma USA - Tuesday, February 22, 2000 at 08:28:28 (PST)
I love that you have actually taken the time to write down these observations as a scientific study. I commute every day and have found myself playing with the traffic in similar ways. One reason is because I drive a manual shift car, and as others hav e already commented, doing exactly what you suggest actually makes driving easier by elimating the constant need to shift gears. Another reason I always leave plenty of space in front of me is to help prevent accidents. Since I began commuting around Bos ton, I have been rear-ended three times, all on the way home from work in heavy traffic. These accidents, which I see all the time, add to the traffic jams (rubbernecking, blocked lanes, etc) and cause all kinds of damage and stress that nobody needs. The y are also completely preventable by leaving some space in front of your car and driving at an even speed rather than rushing to catch up to the car in front of you. Now if only I could get the car riding my bumper to back off......
Boston, MA USA - Friday, February 18, 2000 at 09:58:35 (PST)
Thank you for your observations and comments. I too have experimented with traffic after driving several hundred thousand miles. Please add one more observation. Brake lights... Human nature is to touch your brakes soon after the car in front of you rs brake light comes on. Avoid the temptation and anticipate the slow down. Then the car behind you won't see your brake lights and so on. Result; smooth traffic behind you. I love being the driver smoothing the wave. Remember, drive way ahead, obser ve slow downs and don't rush to hit your brakes. Thanks..fxm
Frank Mollo <TheresaA@aol.com>
Moorestown, NJ USA - Friday, February 18, 2000 at 07:53:40 (PST)
the only problem i saw with your theory in regards to bottlenecks caused by the big flashing arrows? people who refuse to merge until the LAST POSSIBLE SECOND. even if traffic is flowing quickly and smoothly, this can slow it down. they come flying up in whichever lane is ending, and then SLAM on the brakes because they haven't bothered to even put their blinker on, and nobody wants to let them in. any suggestions for that?
venice, fl USA - Thursday, February 17, 2000 at 15:45:24 (PST)
Excellent analysis of the situation. It makes perfect sense that empty space in front of you will reduce waves behind you, even if the net speed has not improved at all. However, I think that once cars get spaced out behind you, their speed is likely t o climb because of the feeling that everything is moving right along, which improves the rate of flow and spaces out the cars even more. I had a few thoughts: -Increasing the speed would improve the throughput of our roads, and therefore reduce traffic density and improve traffic flow. But much longer on and off ramps would be needed to ensure that merging was done at high speed. -Automated throttles/braking could solve the whole problem automagically. If a robot was placed in control of brakes and steering, it could be programmed, for example to brake at 1/16th of a G, and accellerate at 1/8th of a G. Because it takes twice as lo ng to get stopped, the inflow rates to these traffic jams would be half of their evaporation rates. It would be very difficult to maintain a traffic jam in these conditions, as they would all dry up quickly. It would also promote even spacing, and there w ould always be room for high speed merging. What we need are automated cars that will run in special lanes and use special on-ramps which allows for high speed, evenly spaced traffic.
Ivar Thorson <firstname.lastname@example.org>
Olympia, WA USA - Wednesday, February 16, 2000 at 19:17:34 (PST)
While, in theory, the idea of getting over the idea of "winning the race" is nice, there's a lot of reasons why creating a wave-cancelling space in front of you might be a really bad idea. I've experimented with this approach a number of times, and at least in my commute it's just not practical. I drive a very happy little piece of asphalt nightmare called Highway 17 that winds and twists for about 18 miles through the Santa Cruz Mountains, with a 2000' climb and descent. Some of you will likely rememb er the old Jerk du Jour web site a couple of years ago-- that was all Highway 17 commuting. It brings out the worst in drivers. On the majority of occasions when I've tried leaving a space cushion in front of me to alleviate traffic impact, the following things have occured: 1) The cars in the other lane (it's two lanes each direction) quickly begin speeding up and flowing into my space. They quickly accelerate up to something much more like highway speed. 2) Because 17 twists like a shoelace, they rarely go more than another 100' before they hit a curve, where traffic has slowed down again, and suddenly they have to panic stop. 3) So do the twenty other cars who were accelerating into the space ahead of me as well. Golly, people, 100' is not enough room for twenty cars to accelerate from 20mph to 60mph and then back down to 30mph. String of panic stops ensue. On one occasion, th ere was a substantial impact ahead of me as cars slammed into the divider. What *had* been a tremendous space ahead of me was quickly swallowed up as cars went from 60mph to 0mph. What little space remained was just enough for me to squeak over into the other lane and not end up part of the accident. It was, however, pure coinci dence that there was space for me in the other lane-- because, as I said, that lane had been shooting forward at much higher speed to gobble up my safety space. Had there not been a conveniently timed opening in that high-speed lane, I would have had nowh ere to go, and would have had to rely on my anti-lock brakes (and the ALB's of the idiot tailgating me, angry as hell that I was driving slower than the other lane) to keep me from a crushing impact. I no longer give anyone a break on 17. Given a choice between leaving just enough space ahead of me to control my own degree of safety, or leaving enough space ahead of me that others can move in and control my degree of safety instead, I no longer even h ave to think about it. Good theories, applicable under reasonably straightforward field conditions, but not so good when subjected to stress. --The Elder Dan
Dan Johnson <email@example.com>
Santa Cruz, CA USA - Wednesday, February 16, 2000 at 17:06:29 (PST)
We almost never have jams that we actually stop, execp near trafic lights. We have laws to dictate minimum space between drivers. We have a good public transport system, and I have seen a 3 lane freeway.(1/4 of mile long) And 90% of our gas price is tax. (more economic driving since it costs). And only barbaric idiots in here try to cut between others in switching lanes or protect the space in front of them, from others. Okay we must admit that we have 1 bad apple in 100 c ommuters.
ESPOO, Finland - Wednesday, February 16, 2000 at 15:50:25 (PST)
My favorite analogy to traffic waves are the density waves among stars in spiral galaxies. The spiral arms are actually traffic waves among the stars. Remembering this analogy and visualizing it often helps me to relieve the stress of bad commutes.
Tractor-trailer drivers use your techniques by necessity. Observing them, I long ago realized that just one tractor-trailer rig in a lane smoothed out the flow of traffic in the lanes behind them. Plus they have the added effect of having a sort of "psy chological intertia" in that they're SO big that if they travel at a constant speed, often so will neighboring traffic. So in a traffic jam, I would get in the lanes behind a big rig, and eventually practiced moving at a constant speed in heavy stop and go traffic regardless of the presence of big rigs.
It's most effective in the fast lane, where the impatient drivers gather. I used to abandon the fast lane in heavy traffic, since it was full of "jackrabbits" that continually accellerated and braked. (I'm impaitient, too, but I hate abusing my vehicle
like that even worse.) Lately, I've stayed in the fast lane and nullified the waves caused by the jackrabbits ahead of me. (Of course, more than once, a jackrabbit behind me became enraged that I wasn't closing up the space ahead of me fast enough.)
Regarding Traffic Waves... This is something I have observed before as well. I even tried the same techniques of leaving gaps to iron out the wave. Canberra is strange in that it has a small population and a good share of traffic problems, but they move quickly. I have actually seen the waves moving towards me. I like the ideas presented, they make perfect sense and do actually work. For those who doubt the veracity of the concepts, consider the evacuation from a burning building analogy. All little office workers (anybody for that matter) are all taught that in the event of a fire, everybody is to "remain calm and file out one at a time". Anybody care to guess why? Partly so that people don't get trampled. Partly also so that people don't get burnt. A single file of slow moving humans gets more people out faster than a rush at the doors. Why should cars be any different?
Canberra, ACT Australia - Wednesday, February 16, 2000 at 14:42:33 (PST)
Maybe I'm weird, but I've thought of traffic patterns as waves for many many years. The one that I was most aware of is the effect of a red light on a long oncoming stream of traffic. Basically the red/green oscillation of the light acts as the driving force, with a damped wave of traffic going back along the oncoming traffic. (Full stop-start at the light, with more and more minor slowdown-speedup as you are further away from the light.)
Davin Milun <firstname.lastname@example.org>
New York, NY USA - Wednesday, February 16, 2000 at 13:04:50 (PST)
Very interesting. Your ideas will save on gas, brakepads etc, but if the main aim is to get home in the minimum time they don't help at all. The thing to concentrate on is throughput. By slowing the speed you smooth the speed distribution, but do nothi ng to affect the throughput. There may be a minimal gain due to eliminating acceleration lags, but I suspect these will be erased by drivers merging into the gap in front, and entering the freeway into your gap. Having said all that, your technique can be summarised as "be polite and calm", which is bound to make driving more pleasant for everyone. Of course, you may get stressed at being polite and still getting cut in front of. If too many people left big gaps anti-traffic would dominate traffic and there would be no traffic jams, but also a net slowdown of traffic flow due to surplus gaps.
David McWha <jadm@ SPAMFREEcs.waikato.ac.nz>
Hamilton, New Zealand - Wednesday, February 16, 2000 at 11:44:48 (PST)
Wow, and I thought I was the only person who spent his time analyzing traffic flows! From what I have read this is very astute observation. I too have experimented with this behavior thing. Have you ever noticed that certain spots on the freeways al ways behave the same way? The gravity slowdowns going uphill? The blinding-sun coming around a bend at a certain time of day? Now, as the author has pointed out, all we need to do is behave in an anti-traffic manner and we can all reduce the traffic wav es. We will never eliminate them because we are all just people and we do stupid things. Just our nature. But if we create the space and allow others to use that space we shall reduce the traffic waves! Anti-traffickers unite!!
Bradley K. Downing <email@example.com>
Bakersfield, CA USA - Wednesday, February 16, 2000 at 09:42:39 (PST)
Two points... Highways around here curve, allowing drivers to see ahead to the jams encourages slowing down rather than a headlong rush to join the congestion. Also, you left out another factor which seems to be important, proximity to a Wal-Mart, espe cially during the daylight hours. This is a known attractor for the blue-hair crowd, the ones who think that going 15-20 mph slower than the posted speed limit is actually safer than paying attention to what is happening around you and "going with the flo w".
OK USA - Wednesday, February 16, 2000 at 07:34:00 (PST)
Along with the guy a few posts prior re: the M25 and variable speed limits I have a story re: UK traffic. The other day I was driving along the M1 (major N-S m'way) and got stuck behind a police car doing about 40 (limit=70) after about twenty mins he pulled off and let everyone go. "How strange" I thought. Just after this I passed the scene of a new acciden t (glass all over the side of the road etc) and almost immediately joined the back of a flow of traffic doing about 65. Sounds like he was leaving a MASSIVE gap in front of us so that we travelled slower for a while and therefore allowing the stationary traffic in front to clear instead of us screaming up to them and having to stop. The great thing is that he seemed to time it to perfection! Didn't really think about it until I looked at this site. Good work fella. Cosmo
London, UK - Wednesday, February 16, 2000 at 07:11:30 (PST)
Wow. I thought I was the only person to have noticed and tried this. I used to commute here in the DC area down I-270 to whats called "the Spur", where it merges with I-495. I found that by leaving a little space ahead of me I could ease my amount of s hifting (manual transmission) and get my traffic lane moving smoother and faster. If I tried to leave a BIG space ahead it would get filled up with all the surrounding, impatient drivers, who would start doing the lurch and stop thing, and the lane would freeze up again. Every morning was a chance to experiment again, to find that balancing amount of space that would smooth out the lane. When I got it right it was like magic. If you're ever stuck in a non-moving lane and notice one next to you flowing freely, maybe now you know why? Enjoy.
Kirk Wagner <firstname.lastname@example.org>
Washington, DC USA - Wednesday, February 16, 2000 at 07:06:24 (PST)
I've driven with at least a 35-minute commute for almost 10 years now in Boston and in Detroit, and I'm pretty familiar with traffic jams. I've done some experiments like yours, and you're right on the money.
I first started tinkering quite unintentionally. My car has a manual transmission. If you're a driver in a long commute a manual is about the worst thing you can drive.
clutch, 1st, clutch, 2nd, clutch, 1st, clutch, brake, clutch, 1st
So what I discovered is that I'd rather do a constant speed (i.e. remain in the same gear) as long as possible. Not so fast as to rev the engine up high, but not so slow as to stall. So when I get in a jam, I pick a gear and try to stay in it. This has the effect you described of leaving a large gap in front of me which grows and shrinks as the traffic wave approaches me. But, if I pick the right speed I never quite catch up to the oncoming wave. My 'column' of traffic never quite comes to a stop and I've looked back and seen a uniformly-spaced column of cars behind me going at this speed.
The effect is great, I wonder if anyone appreciates it. :) I certainly like not having people rush up the bumper of my car just to hit their brakes every few seconds in the midst of a wave.
The other thing to note, though, is that this uniformly-spaced column of cars doesn't last. Invariably, someone will 'force' their way into the column. Even though the column has some gaps in it, they're ususally not prepared for an "invasion" like this
and a mini-jam wave will form (and spread back through the column). However, since I'm usually in the leftmost lane someone will leave the column to go to his exit and the uniform distribution will slowly return.
Here in sunny England we have a motorway called the M25 or Magic Roundabout which circles London. On some sections where commuter traffic is heavy they have introduced a variable speed limit. By lowering the speed limit the traffic flows at a more cons tant speed, which almost eliminates the "wave effect" we used to get.
Colin Bridger <colin@wychway889. freeserve.co.uk>
Guildford, England - Wednesday, February 16, 2000 at 05:17:37 (PST)
Interesting .... Years ago i developed a phylosophy ... in a traffic jam leave as much room in front of you such that 10 mph * the number of open car lengths = the speed i would like to go ( e.g. 3.5 car lengths = 35 mph (anticipated) It WORKS!
Dave Price <email@example.com>
Denver, CO USA - Wednesday, February 16, 2000 at 01:06:40 (PST)
Check out our gouvernments website about "zipping" http://www.ritsen.nl/
USA - Wednesday, February 16, 2000 at 00:58:39 (PST)
HeHeHe, You must realy have been bored to come up with all this... But anyway, Here in holland the gouvernment is promoting to let two merging lanes act like a "zipper" through tv commercials. It realy works. Also there are policemen waving lighted sti cks at cars at the end of a jam meaning "move faster!" I don't understand their use but it looks funny and it makes you think about where gouvernment funds are going....
NL - Wednesday, February 16, 2000 at 00:54:26 (PST)
I think your ideas on traffic flow as liquid are interesting. The proper term is, I believe, a "slurry", which is solid particulate matter conveyed in a liquid stream. I agree with some of your observations, but have come to somewhat different conclusions. I worked and drove in the Seattle area for three months, from March thru May 1996. In that time, I was astonished at the driving habits of Seattle drivers. First, there is basically only one way to travel, North/South. However, there are several alterna te ways to enter the North/South freeways. Whichever method is used (enter in the center of the freeway, or enter from the edge), the entrant driver IMMEDIATELY turns and drives to the extreme opposite lane(outer or inner), aggressively cutting in front o f whoever is there, without regard for accident.Secoun: it is ALWAYS raining in Seattle, therefore 1) drivers there will know how to drive in the rain, and 2) will ALWAYS have full tread depth tires. Yea, right! In the first month I was there, March 1996, Seattle area POLICE were involved in 29 accidents in 30 days! (check on it) Moving on. People are simply AFRAID to accellerate! You do the math. If a given intersection is 50 feet across, the average car is 15 feet long and the average light stays green for 65 seconds: How many cars can go through at an average speed of 30mph? Now time one. How many DO GO through? How many are lo st if the first car is doing dishes? talking on his/her cell phone? just doesn't give a shit? There is simply no other job you will ever do which takes as much concentration as driving in rush hour traffic, (unless you are a brain surgeon or a fighter pilot). Yet the drivers treat that job with impunity, as if it is just DEAD time which must be fi lled with activities OTHER than driving, and the auto manufacturers aid and abet them by providing cell phones and drink holders. In short, the phenomenae you observe are valid, but the CAUSE is ass holes who drive, and the CURE is much more rigid driver training, and fewer licensed drivers. I have taken driver training many times because of tickets. At each of these I ahve asked the participants some questions. 1) How many here are licensed pilots? (More than half) 2) How many here participate in other high risk sports such as skydiving, ra cing, skiing or SCUBA diving? (same people raise hands) 3) How many of you have had an accident in the last 10 years? (None of those raise hands) In my opinion, although your wave mechanics theories are ablolutely correct, they are symptoms. The causes of the traffic anomalies themselves are the drivers, and god forbid, (because they are NEVER given blame in any official investigation)the highways themselves (design/construction/materials). I haven,t finished reading your whole series, but it is good thought. Thanks, Garrett
Garrett Waddell <firstname.lastname@example.org>
Augusta, Ga USA - Sunday, February 06, 2000 at 14:14:38 (PST)
Interesting to read the experiments on the anti-waves. I used to do this without knowing that it had any effect on the traffic following me. My last home was in Reno, Nevada, and my commute passed through a daily traffic jam created by the city's main (only) freeway interchange. If anything out of the ordinary happened, the traffic would back up for usually over a mile of stop-and-go misery. My truck is a stick-shift, so this got old remarkably fast. I used to just put it in first and idle through the jam. Trying to maintain constant speed in stop-and-go traffic was an interesting challenge that made the miserable congestion more bearable, but I never thought to look behind and see what I was doing to the traffic flow. Now I wish I had - that could hav e been interesting. But I walk to work instead of driving now, so my personal traffic-jam problem is solved :-)
Seattle, WA USA - Wednesday, February 02, 2000 at 15:20:04 (PST)
Your pages are very interresting. I agree with your solutions, to my mind they can be summarized in one sentence "keep cool behind your wheel".
Mathieu Berland <email@example.com>
Quimper, France - Wednesday, February 02, 2000 at 06:47:02 (PST)
Your pages are very interresting. I agree with your solutions, to my mind they can be summarized in one sentence "keep cool behind your wheel".
Mathieu Berland <firstname.lastname@example.org>
Quimper, France - Wednesday, February 02, 2000 at 06:46:49 (PST)
This is really good stuff. I think that Car and Driver might be interested in it. Please consider submitting it.
Michael Dunn <email@example.com>
Mountain View, CA USA - Tuesday, February 01, 2000 at 11:28:35 (PST)
Fascinating stuff. The wave theory of traffic jams seems right on the money. I will try the merge thing next time I am out. Disturbing to think I am part of the problem not the solution. I've noticed that in europe they leave the left lane open for faster traffic, whereas in north america it is everyones god given right to drive the speed limit in any lane they wish. I think the the european approach is much more sensible. Any thoughts on that?
toronto, ca - Sunday, January 30, 2000 at 14:27:18 (PST)
Great information on traffic waves here. I have been commuting around the M25 (London, UK) for many years and have observed the same basic principles. Traffic behaves like a fluid in a pipe. Where pipes join (intersections) you can get pressure buildup s (Jams). For a laugh some friends and I tried modelling traffic using Shroedingers equations with jams as potential wells. Surprisingly we found jams can be persistent or spontaneously dissolve, wave function collapses. One problem in the UK is that our motorways are 3 lanes only, four at most. Think in terms of the population of Calif. commuting on 3 lane freeways. One section on the M25 is particularly bad due to 2 other motorways insecting the London Orbital and Heath row airport in the middle. So they put in a fourth lane and an enforced variable speed limit. Traffic is monitored, where a jam starts building the speed limits are decreased behind the jam and signs are put up warning people to slow and not change lanes . The signs also say where the jam is. It largely works. Maybe the engineers read your page first? To the critics who say use a bike or bus; You try cycling 40 miles at 5am in the morning with air temperatures of -3/-5 degrees celsius. Buses, hah, it would take about 2-3 hours to do that on a bus as there are no direct services. Dave "I try to be a good driver and leave gaps" Barlow
Dave Barlow <firstname.lastname@example.org>
Basingstoke, Hampshire, UK - Friday, January 28, 2000 at 01:36:30 (PST)
Do everyone a favour and get out of your car and onto a bus or a bicycle. If this isn't realistic then you should move closer to work.
Jess Bossert <email@example.com>
Victoria, BC Canada - Tuesday, January 25, 2000 at 12:55:32 (PST)
CHP generated BIG wave! Just before Xmas I was North bound on 101 in the fast lane moving at the limit in dense traffic. A few cars in front of me a grey MB ML SUV had a xmas tree tied to the roof (remember going the legal limit). All of a sudden the xmas tree was in the fast lane and the car in front of me swerved trying to miss the tree. I had no choice except to drive over it. Shortly there after there was a CHP car with red lights flashing coming up behind me in the fast lane. I pulled over and let him pass. He found a break ahead of me and started weaving between the 4 lanes of traffic, slowing down the whole freeway to may be 20 MPH. We kept at this low speed for a number of minutes. Soon there was no traffic in front of us and we were joined by another CHP car that backed up to meet us! During this cars were entering the freeway from on ramps ahead of us. (In other case s I have seen the on ramps closed when there is an acident ahead.) I think what was happening was that the CHP had created a wave that stoped the traffic where the Xmas tree was in the road so that it could be removed.
Brooke Clarke <firstname.lastname@example.org>
Ukiah, CA USA - Sunday, January 23, 2000 at 09:39:13 (PST)
OOPS! I DAMAGED THE GUESTBOOK PROGRAM. IT'S FIXED NOW.
Reading this article I realized that this was exactly what I have been doing for the last couple of years. In the Netherlands the problem with traffic-jams is becoming bigger and bigger. The solution explained in this article seems simple, but I have t ried this several times and must conclude that it works. When you stop making the 'wave' movement and begin to slowly drive your car, leaving gaps in front of you, that the traffic jam behind you seems to dissapear as everybody starts to move on slowly. In the Netherlands the solution would be to have a maximum speed during traffic jam hours. As an experiment we should test this on one of our traffic jam roads as the 'A2'. Every single morning there is a traffic jam on exactly the same place. I wonder wh at would happen if we could convince people to actually drive 50KM/ph on this part of the road. Ciao, Leon
Leon Kuunders <email@example.com>
Den Bosch, Netherlands - Monday, November 29, 1999 at 00:46:25 (PST)
Many years ago, before traffic jams and e-mail, I studied the flow of water in the streams of my native Italy. You may have observed that when water flows and arrives at a narrow spot, if actually flows faster. I used small pieces of wood or straw to m easure how fast they moved in the broad section of a stream and also in a narrow section. I found that when the width of the stream went from a two lane to a one lane the pieces of straw moved twice as fast. This makes sense, because the same amount of wa ter moves downstream in the broad and the narrow sections of the stream. Thus if we could double the speed of traffic in the narrow sections, no traffic jams would occur.They do because we do the opposite. We reduce the speed in those sections to one half (to 30 mph down form 60 mph). This makes the jam worse. Nature is wiser than we are and knows how to keep the water moving in narrow sections. What can we do to solve the problem? I propose that we imitate nature. First we need to slow the traffic before the narrow section, say from 70 mph to about 20 mph, I say slow, not stop. Then as we approach the narrrow section we allow the traffic to move at 40 or 50 mph. This can be done without making any car stop. Finaly, the cars will resume their normal speed . This needs to be studied by the traffic engineers in more detail to determine the best combination of speeds to solve teh problem. The solution used in Holland, as mentioned by a dutch engineer in these comments, is probably not too far from what I prop ose here. The most painful thing for me is to see that after several centuries my studies of flow dynamics, straws moving in a stream, now cars in the highway, are not known by the "apparently" hypercivilized age. My children, you all need to look at the past, you all need to go to the library and read about traffic, physics, fluids and straw floating down the river. Ciao you all.....
Michael Angelo <firstname.lastname@example.org>
Florence, Italy - Wednesday, November 24, 1999 at 17:00:32 (PST)
Thanks for your thoughtful insight on trafic jams. Having lived in L.A. for almost 15 years I have seen traffic go from bad to worse to impossible. In the last few months I have been practicing "friendly" driving. I have nocticed that I seem to get lot s of positive feedback from other drivers and sometimes traffic appears to ease up when I leave large gaps in front of me. It works! I was doing this before I found your website in order to curb my "roadrage" tendency. I realized it was only a matter of t ime before I got pissed off at somone who had a gun. DRIVE HAPPY!
Los Angeles, CA USA - Sunday, November 14, 1999 at 16:44:00 (PST)
Your ideas sound great but living in Chicago area for some 25 years, if you left several car lengths open in front of you , at all times you might as well park. The problem in this town is pick up trucks ( mostly from Ind.) in the left lane doing 90 or 25 ,same truck only 2 minutes apart. Any space left in fromt of you would be filled in a matter of seconds, this space could be as large as 2 feet and some suv will cut in and you had better hit the brakes or hit him. Your ideas are better out in the land of grey skys and rain. willie
chicago , IL USA - Saturday, October 30, 1999 at 16:53:55 (PDT)
I live in Portland, OR. Where I think our drivers are pretty polite on the freeway traffic jams. Most of us are quite willing to make room for someone who wants to change lane and cut in front. There's always a few *$heads who cause problems and I keep wondering if these people came from California to escape the traffic problems there. General rule at all congested on-ramp merges:
Bill Rogers <email@example.com>
Portland, OR USA - Thursday, October 28, 1999 at 15:58:44 (PDT)
Get a life people. Take the F%#$ing bus!
San Diego, CA USA - Wednesday, October 27, 1999 at 00:30:16 (PDT)
The only reason traffic builds up due to merge lanes is because MOST DRIVERS HAVE THEIR HEADS UP THEIR ASSES!!! Logically, every OTHER car should go when merging onto a roadway. But OF COURSE those FRIGGIN BMW drivers think their shit doesn't stink and they can cut in front of 3 cars. SCREW EM! I always go out of my way to cut off BMW drivers now. =) haha Let em eat shit!
Craven Moorehead <firstname.lastname@example.org>
Newville, AL USA - Tuesday, October 26, 1999 at 09:07:06 (PDT)
i hate the cars they all suck take it off
willie <jon is a skin email@example.com>
freedom oh, al USA - Friday, October 15, 1999 at 09:57:36 (PDT)
Wow, were you bored!
Lawrence Levesque <firstname.lastname@example.org>
USA - Thursday, October 14, 1999 at 10:27:52 (PDT)
I have spent the past years living in three different metro areas: San Fransisco Bay Area, Columbus OH, and Phoenix AZ. I've been in too many traffic jams in all three areas and from personal experience it seems your observations about traffic are correct.
Tucson, AZ USA - Sunday, October 03, 1999 at 02:22:38 (PDT)
I've noticed the same phenomenon occurs in freely moving traffic, notably on freeways. Cars moving faster than the mean speed tend to clump in the two fast lanes. Occasionally they are held up when squeezing past a car moving too slowly in the left min us one lane, but the pack reforms. As a practical matter, if a driver stays at the rear of a pack, the car will travel faster than the ambient traffic but with less risk of being ticketed. The pack demonstrates considerable but evolving coherence as drive rs drop in and out. I've wondered if there can be derived a coefficient of affinity for drivers in a pack of cars, based on their inclination to stay together, and whether there is a different coefficient for drivers of different speeds.
F. Tomkins <email@example.com>
Sitka, AK USA - Monday, September 27, 1999 at 14:47:23 (PDT)
G'Day and great site Mate.... Its good to see that there is a forum in which such discussion can occur......I have a few comments on your work however. Firstly, Your methods to resolve traffic problems sound fairly convincing, and if cars were like water molecules, then your model wou ld work. However, having a medical background, i can say for sure, that humans being what they are driven more by emotion than rationality....In a modern 'capatilistic' society we expect more with less wait....everything is instantaneous..... the modern m an waits for nothing......hence in a traffic jam, expectation and emotion mean that he can never drive slowly continuosly but rather make gallops of distance, the stop go effect (perhaps Ive explained this too simplistically, but anyway). Hence even if on e man should slow down to eat up the traffic jam, it is sure that another Jam will most probably build up further down from impatient drivers angry at having to slow down earlier than normal, having to drive their cars at less than their potential. Second ly, I believe that even though your methods will not completely resolve traffic jams, the reduced 'stopping ang going' will ahve a positive effect on the environment. I have no figures to support me, but Brake pads will last longer at least. Also, to the traffic engineer who 'obliterated' your views. The views presented on this site may not fix traffic problems completely, but a small significant impact in terms of the environment is a good imapct enough. And also the will to look beyond your own frustrat ion to make a change for others and the good of society by large is admirable. Good on ya mate and keep it up Dr. Shannon Thomas
Shannon Thomas <firstname.lastname@example.org>
Sydney, Aussie - Friday, September 24, 1999 at 07:52:56 (PDT)
I wholehearted agree with your basic theory. I personally believe that maintaining an adequate following distance is key. That way, drivers can simply let up on the gas, as opposed to braking when traffic slows. I believe that brake lights cause braking in 90+ percent of following drivers. The (following) dr ivers will actually slow to a greater extent than the cars they are following. This process repeats until gridlock occurs. Following distance is taught in driver's school already, but there needs to be more emphasis on the fact that you can control your cars speed without using the brake by simply letting up on the gas. So many drivers feel they must be depressing one of the peddles at all times. Not so!
S P Gass <email@example.com>
Alexandria, VA USA - Monday, September 20, 1999 at 08:53:46 (PDT)
I believe Albuquerque, New Mexico did a study a few years back, in which the tested different speed limits during rush hour traffic to determine how it affected traffic flow. I remember seeing the changeable speed limit signs between the Rio Grande ri ver bridge and the I-40/I-25 interchange on Interstate 40. It would be interesting to know what info the collected. Talk about a terrain problem! Albuquerque is downhill for about 15 miles going west into it, then about a 10 mile climb back up the west side of the Rio Grande valley. Traffic always to seem to flow fairly smoothly there. I still feel if people would start thinking of the vehicles sharing the road with them as neighbors, instead of competitors the flow would be smoother. It definitely would help with controlling road rage.
USA - Saturday, September 18, 1999 at 21:40:47 (PDT)
Roads should be thought of as communities rather than a racetracks. In high density traffic, one has to think in terms of the good of the greater group, not as an individual on a mission. People in crowded situations tend to handle like cattle going to the chute. Everyone wants out as quick as they can from the confinement of slow or stopped traffic. As for resloving the traffic wave situation experienced in heavy traffic, terrain can play an important part with commercial truck traffic. Once the t ruck is stopped it does start quickly. The Law of Inertia comes into play here. The best way to keep the trucks moving is don't get them stopped. A slow creep in a low gear at an idle works best from the cab's point of view. Once I can get a pace I tr y to maintain it to buffer out the stop and go wave. The spaces in front of me help relieve the anxiety of the cars around me. It is interesting to observe them from above, they rarely gain much turf in the long run. The traffic wave eventually will move completely away from where the original problem was such as an accident or stalled vehicle and can take many minutes to disappear, especially if the volume of traffic behind the wave is increasing.
USA - Saturday, September 18, 1999 at 21:23:15 (PDT)
This is a great site. Every morning this summer I've been communting along Rt 15 to New York and wondering why there needs to be traffic. I've thought about it a lot and I agree with all of the things that you've said. I noticed that every day there was traffic in the same spots and I started thinking about what would make those particular locations prone to more traffic. Strangely, the cause did not turn out to be cars entering the highway and merging with traffic, but were related to the actual t opogrophy of the road. For example, say everyone on the road is moving along at 60 mph. If just one person going around a sharp curve or coming to the bottom of a hill decides to hit their breaks. Often curves and hills make us feel that we are travelling too fast, so this person slows down to 50 or 55 mph until the curve is over or they are on flat ground again. As a result of that person's slowdown, the car behind him is forced to slow down even more. For example, the car behind him notices breaklights but doesn't know exactly how much he needs to slow down so he reduces his speed to 45 mph to leave enough distance, especially on a part of the road like that. Consequently, the car behind him ends up in a similar situation, he must slow down even more than the car ahea d of him. This effect multiplies until traffic is just stopped for no real reason. I think that this is why there is so often traffic at the bottom of hills or around sharp turns. You are right in saying that leaving a reasonable distance between you and the car ahead can absorb these effects and keeps traffic at a constant rate. Thanks for such a good website! Adam Nelson
Adam Nelson <ASNjazz@aol.com>
Trumbull, CT USA - Tuesday, August 10, 1999 at 08:34:41 (PDT)
Well you've solved the traffic problem in front of you, but you've totally neglected the situation behind you. Consider this: You're on the freeway, and you spot a traffic jam a mile or so up the road. You slow down to create the huge space in fron t of you, and by the time you get to the jam it is gone. BUT, while you were slowing down the traffic behind you, more cars were entering from the on-ramps at a normal rate. Since traffic is now moving slightly slower than usual, cars aren't leaving th e freeway at the same rate as they normally would. You've created a "zone" of slower-moving traffic where the population of the freeway builds up more rapidly than usual. As the "zone" extends further and further behind you as you approach the traffic j am, more and more cars enter the freeway and aren't able to leave at the same rate they are entering because they don't reach their exits as fast. And just like magic, you make the traffic jam in front of you disappear, and then make it REAPPEAR BEHIND Y OU!! I'll give you a big hand for effort, but in all of the thought you put into eliminating the traffic in front of you, you completely forgot about what was going on behind you. You eliminate one traffic jam, and replace it with another. At least you're no t creating a whole new one. Unfortunately, though, your experiments do nothing to actually ease the flow of traffic.
Chad Spencer <loser007@ hotmail.com>
Bloomington, IN USA - Saturday, August 07, 1999 at 09:24:44 (PDT)
At last! An excellent science article about traffic dynamics. It's in SCIENCE NEWS, and discusses spontaneous emergent patterns, phase changes, nonlinear dynamics, and cellular-automata simulations. It also points out the war between those who study the traffic physics, and those who are trying to cure the problems of congestion:
Some of the new mathematical models, advanced mainly by physicists, depict traffic as more complex and unpredictable than traditional traffic experts, mainly civil engineers, believe it to be. These models have sparked intense controversy....but then we've already encountered this problem, right here on this website. A person who reads Gelick's book CHAOS will see traffic with new eyes. A traditional traffic engineer might be angered by the very idea that new discoveries can be made whic h aren't already in their textbooks. (Even worse that a total non-expert might already know about them!)
Bill Beaty <>
Seattle, WA USA - Thursday, August 05, 1999 at 23:18:15 (PDT)
I've been doing this sort of thing on the NJ Turnpike for years, with very very positive effects, especially around the Newark Bay extension area.
Daniel Drucker <firstname.lastname@example.org>
Plainsboro, NJ USA - Thursday, August 05, 1999 at 15:09:18 (PDT)
I've found that it helps to think of lane MERGING as if zipping a ZIPPER -- the most efficient way to merge two rows into one.
Newport Beach, CA USA - Monday, July 26, 1999 at 07:53:54 (PDT)
I live in California, Land of the Commute. I drive home from work on a two-lane highway; not enough to handle the volume of cars, causing a slowdown every evening. I have to make the comment that (at least on this freeway), leaving a nice, large, emp ty space ahead of you, although a very good idea, can be counterproductive if it's too big. Drivers in traffic jams are attracted to large holes in the traffic like moths to a light. People in front of you in the other lane, seeing the large space ahead of you, will use it as a drag strip to get ahead of the other cars in their lane, and then often merge back into it. Even worse, people behind you will grow frustrated with the moron not going as fast as possible, merge into the other lane (which will p robably pass you by at some point), and then merge back into your lane to use the empty space to get ahead as well. In scientific terms, it is osmosis: blank spaces of traffic tend to be filled with other traffic, until it stabilizes back into one big, ev enly spaced parking lot again. So you're no better off than before, and with all the merging, it may even hurt the situation. Leaving enough space ahead of you for someone to merge is very important, but too much will be exploited by other people, even if it doesn't really get them anywhere faster. Another interesting thing I've noticed (and this may only apply locally to this two-lane highway), is that when traffic gets bad, it's almost always better to use the right lane. While the cars in the left lane are suddenly stopping, the right lane will continue to slowly, gradually, move, putting you far ahead of the cars in the left lane. There are two reasons for this I can think of. One is that because of the "left lane = fast" mindset most people have, enough cars merge into the left lane that it is actually more clogged. But perhaps the best reason is that trucks use the right lane. Trucks, unlike cars, brake gradually and leave a safe (although, once again, not massive) distance between them and the car ahead of them. That is where your theor y of stop-and-go slowing everything down comes in. The right lane has less stop-and-go traffic, and as a result, flows at a constant, and faster overall pace. Not to mention less stressful. However, this would no longer work once traffic is naturally f lowing at a speed faster than the slowest truck travels, and sure enough, that is what happens. Once the speed of both lanes recovers back to around 45 mph or higher, the left lane is once again faster. Obviously, this is not necessarily true for all hi ghways, and I'd be interested to see if it's still valid once you add more lanes to the equation, or in other locations for that matter. But still, my advice in slow traffic is to get ahead of a truck, behind one, or both. The visibility might suck, but your ride will at least be smoother, and you can get off at the next exit if necessary with no problem. Anyway, this site was great. Traffic can be so fascinating. Regardless of how scientifically true anyone's theories are, it's fun to hypothesize about the nature of our clogged highways.
J. Smart <email@example.com>
CA USA - Thursday, July 22, 1999 at 20:31:16 (PDT)
You can't spell "therefore." It's not spelled "therefor."
bigtown, NC USA - Sunday, July 11, 1999 at 17:02:08 (PDT)
Musings of a scientist over a decade of traffic: consider the parameters involved in the wave phenomina related to trafic flow: wavelength, velocity and how they relate to the system properties: density, flow rate etc. the on-ramp effect is a resonant system that can be tuned or detuned and several such on a highway form an interference pattern. Judicious choice of flow rate and density conditions combined synergisticly with the design of certain interchange give a beautiful ossilation ineight lanes of homicidal commuters. very like double slit experiment from freshman physics. More significantly are the quantum effects noted. Consider the Heisenberg constraints when position and velocity of the motorist is correlated with those of thousdands of others. uncertainties result in the spontaneous propagation of waves in the system, avalanche effects are noted and the traffic jam forms. Behold the brake light waves! now, the observable phenomina link the motorists together and the physical effects link to psychological effects. Can road rage be a collective behavior? how do waves of emotion (Darn! stuck again!) <=> (Whee I can go!) correlate with the brake light waves? How about the blood pressures of the subjects? I see another comment from the "close the ranks" (a.k.a. tailgating) crowd. "2 second rule"? That's not enough time to react. Try 4 seconds or more. Lots of people out there have no clue as to the physical and physiological principles involved in c ontrolling heavy machines at high speeds. For most people the science of driving boils down to 2 points. 1) if there's enough room for a car, put a car there. 2). the left lanes are for speeding. Crunch! there goes another one. The requirements for a drivers license should be at least as stringent as those for a pilot's license. Physics, mechanics, human factors, emergency procedures, situational awareness, etc. All sadly lacking with predictable results. Go ahead, close the gap. At 0 mph the re quired following distance is 0. Total gridlock, but at least there's no "wasted" space, nor is "slower" traffic in the holy left lanes. Traffic nirvana just the way we like it.
Gridlock, USA - Wednesday, June 30, 1999 at 23:02:12 (PDT)
One thing I haven't seen commented upon (in a cursory glance through the other comments) is the dramatic effect on reducing road capacity caused by the drivers who think that the right lane is for wimps only. They generally drive at below average spee d, turning a 3-lane Autobahn into a 1-lane trunk road and tempting others to overtake on the wrong side. These people are wasting an important resource and should be encouraged to keep to the right (education rather than coercion). If this doesn't work, how about fitting all cars with a TV camera like those in Formula 1 / Indy racing? A few minutes sampling the view from cars on critical sections of road would identify the wasters and provide evidence to 'encourage' them to change their ways. But they d o allow others to merge from both sides!!!
Tracey Runciman <cecilia.runciman@ hamburger.netsurf.de>
Hamburg, Germany - Tuesday, June 29, 1999 at 08:22:29 (PDT)
Interesting site. A few personal observations and comments:
1) Adding extra lanes (inclined sections for example) causes traffic jams when the lane is removed downhill.
We Northwesterners are a competitive bunch (just ask My Buddy Bill G.). Not being satisfied with having the fourth worst traffic in the nation, we moved to Number One this year! Naturally, to attain this status we had to elevate Stupid Driving to an ar t form. Being an Aeronautical Engineer involved with control systems on a daily basis (not Boeing, by the way), I must state unequivocably that Standing Waves CAN BE eliminated through focused effort. And no, it won't be by Slow Poke/Average Speed activ ists, who inevitably will see car after car change lanes in front of them while their blood pressure rises (Zen or no Zen). The answer is Feed Forward, and in addition to reducing or eliminating the Standing Wave effect, it also reduces the start/stop syndrome. Here's the trick: Instead of releasing your brakes after the car in front of you does, look through their glass and release your brakes when the car in front of THEM releases their brakes. The third tailight allows you to do this. No, you shouldn't floor it and ram the car in front; use a modicum of restraint. Likewise, when the car in front of THEM brakes, you likewis e BEGIN to brake. I have no doubt that if the programmer who fashioned the animation modified the program in this manner, the wave would be much less entertaining because it would be difficult to detect. My pet peeve: STAY RIGHT EXCEPT TO PASS! In the Northwest, our lab rats seek out the "Fast Lane" in droves only to find (surprise!) it's slower than than Slow Lane. Luckily for me, nobody else seems to be aware of this, which is why my commute is much sh orter than my automaton neighbors. One last comment: here's a brainstorm for the engineering-impaired, the recommended Two Second Following Distance WILL BE DIFFERENT at 10mph than when you're going 60mph. You do the math. Let's close the ranks out there, people!
Roger Clarke-Johnson <firstname.lastname@example.org>
Kirkland, WA USA - Tuesday, June 22, 1999 at 05:09:34 (PDT)
I used to commute 60+ miles per day in the Chicago area, and had actually begun to formulate such a theory. Naturally it came from indignation: 1) Tailgaters. Very often I'd be tailgated by some person who, after passing, went only 3-5 MPH faster than I'd been going. SO I did a little math: I'm going 70, the other guy 75. Over a 50 mile distance, he manages to save a whopping 3 minutes. 2). Traffic jams: as you (CORRECTLY!!!!) point out, people who try to get through the jam in the minimum time don't help at all. Ever seen a semi truck block the closing lane, driving as fast as the still-open lane? He is doing what you suggest (I alwa ys leave room for those guys to get in--because I know exactly what he's trying to do, and I approve, not just because he'd squish my Escort into tinfoil...). The guys who zoom up the empty lane and duck over at the last instant are only making things wo rse. 3) Speed: the notion that we can eliminate traffic jams by raising the speed limit misses the point. The problem is not that cars are going too slow; the problem is that the fashion in which they are driven causes jams. It is a proven fact that increas ed speeds lead to more accidents--accidents block lanes, and lead to rubbernecking, and all this does is slow things down. 4) Exits/merges: I run under the "allow one in" rule: when approaching an on-ramp, each car should let one car in front of it. And if you're merging, you should not try to squeeze in behind the car in front of you who just merged. (this falls under t he "good luck" category--like eliminating murder or pay toilets). 5) "Anti-traffic": The wave/particle duality of traffic does not lend itself to easy analysis, so one must use the paradigm one was trained with. As a BSEE, I think in terms of "cars" and "holes" (similar to semiconductor theory, wherein you have "electr ons" and "holes). But it applies--cars want to fill in the holes, and so on.... 6) Good site. Keep it up.
Ed Hering <email@example.com>
Cedar Rapids, IA USA - Thursday, May 27, 1999 at 13:22:40 (PDT)
I agree with most of your findings. I also have spent time thinking of ways to improve "flow". Here is another for your list. When a traffic light turns green if everyone would excellerate together it might be possible to get maybe 40% more cars thr ough the intersection. If we could some how link the cars to the light so this could happen we would have it made.
Mark Banks <friends>
Snohomish, wa USA - Monday, April 26, 1999 at 13:13:29 (PDT)
I thought I was the only one to try the "anti-traffic" approach! I used this method in 2 hellish commutes - San Diego to Escondido along I-15, and later on I-80 in the SF bay area. I definitely agree that congestion is worsened and amplified by peopl e trying to create headway by tailgating and constant lane changing. My intent was different from yours - I was not trying to erase any particular jams, I was just tired of the brakes-gas-brakes-gas-brakes routine that is required to achieve the coveted "no-space-in-front-of-you" state. It's an awful lot of work going from 0 to 60 back to 0, then 30, 0,50,0,60,10,0.... why work that hard when you're only going to average, say 10 mph on the trip? So I started trying to figure what the actual average spe ed was for a given section of the road, and just travelled at that speed. Of course the drivers behind me became horribly irritated as they would see first one empty car-length of pavement in front of me, then 2, 3... by then they could hardly restrain t hemselves! It took a lot of determination not to give in to their horn blasts and vulgar gestures. There was no way I could explain to them that they'd still get where they were going just as quickly. Maybe they'd realize it after a few minutes? The g ap widens, 10 carlengths then 20! 30! Ahhh. but now the group ahead has hit the next crunch. As they sit idling, I putter along without need of brakes. I take my foot off the gas and coast effortlessly. And as I near the pack, they begin to move. Th ey've hit another "hole" and they of course accelerate as needed to keep their snouts firmly in the butts ahead. I've been driving effortlessly, minor speed adjustments, no braking, no constant monitoring the few inches between my bumper and the one ahea d. It's actually enjoyable!. Maybe now the folks behind will realize this hidden truth. But no, they are still fuming. Travelling at a constant speed makes no sense to them. To them, one can only maximize speed by going as fast as possible during eac h instant of the drive. The sweet bliss of gentle coasting is marred by the constant barrage of horns and flashing headlights behind me. They just can't stand it! They begin taking dangerous chances in order to swerve around me and demonstrate that THE RE IS EMPTY PAVEMENT AHEAD - what am I, an IDIOT not to see it? Not to OCCUPY it? Well, over time I begin to relax and enjoy even this aspect of my new-found power. Yes, every driver is motivated by self-interest. But I have transcended the traffic ja m. I have realized that the ultimate goal is not speed. On this road, at this time, speed is impossible. All of the power and maneuvers in the world will not achieve it. It should be obvious - look at that yellow Honda that has made 20 lane changes in the last half hour, cut off a dozen drivers and nearly caused five accidents. Ah yes, there he is. 20 yards up. No, I've realized that an additional 20 yards is nowhere near worth that amount of work. By travelling at the average speed, letting the g aps grow and shrink in front of me, I alone have achieved what these thousands all lust for. Open road in front of me. Ahhhhh. Bliss! Well, I no longer submit myself to these crazy commutes. But it was nice to discover the zen of traffic. Congratulations to you too! Don't worry about the prickly highway/traffic "engineers" and their sour grapes. Those engineers created the whole mes s to begin with. Keep spreading the enlightenment! Peace!
USA - Saturday, April 24, 1999 at 10:33:21 (PDT)
The problem of traffic congestion is, YES one of physics. Take 50% of the cars off the road and the problem goes away. ie: Don't give woman licences.
Brad Gray <firstname.lastname@example.org>
Sydney, Australia - Wednesday, April 14, 1999 at 04:20:36 (PDT)
I think that many of your observations are correct. It definitely is a very complex fluid dynamics problem. I agree with you that one car can make a difference. Like you, I've done it myself. There was an article in Scientific American about 10 years a go that discussed some of this. I know it appeared sometime between Dec 1988 and Sep 1991 (because I know where I was living at the time), but their Web site doesn't have issues back that far.
David Reynolds <dreynolds@systemware- inc.com>
Dallas, TX USA - Tuesday, March 23, 1999 at 19:20:27 (PST)
Holland is a small country with lots of traffic jams so i was reading your website with great intrest. It is nice that the knowledge of chemical engineers can be used in all parts of society. I want to tell you about an experiment they did in Belgium. In summer a lot of people go to the coast and the beach. In the evening when everybody is heading home they had huge problems with the traffic heading to Brussels. To solve ( or make less worse) the problem, police motorcycles were used to force the speed down ( they call it "block rijden" (rijden is driving in eng.) I don't know if there is scientific data evailable on these experiments. Maybe the ministry of transport in Belgium knows more. It would be nice for you to have some data you can present on your webpage. It is nice to see that people still have some natural behaviour in them although this causes trafic jams. Nice subject for the psychologists.
i actually studied this in 1980 when i started seeing it on 1-17 in rush hour traffic in phoenix. i noticed it seemed to get worse in homeward bound traffic at the end of the day, i supposed it was because people were tired and their reaction times were slower. i also noticed it happened each day 5 days a week and the prevailing traffic conditions (someone on shoulder, etc.)did have an effect but were not responsible at all times. in fact, on i-17 it was all the time and what the conditions didnt matter at all. by timing it for years (what else is there to do anyway)i found that the leading cause matched the road itself. as each major crossroad is approached 1 mile apart the road dips down to go under the overpass and each car would ever so slightly speed up and then slow down as the road elevation changed. what was causing the effect was a traveling harmonic motion (like a slowly moving standing wave) going in the opposite direction to the flow of traffic. the stiffness of the propegating medium corresponded to the average spacing between cars, average reaction times of the driver (varies throughout the day) and the average traffic speed (also varies throughout the day). the difference between slowdowns and real bogdowns were related to the cop on the shoulder, the lady with a flat, the random kid on the overpass with bricks, etc.. i actually in 1983 almost brought this to the attention of phoenix traffic engineers. realizing they might haul me away as a real schitzoid, i never did. how refreshing it is almost 20 years later to find out that i would at least have had some friends in the rubber room they had waiting for me. (or would i have sat there alone for 2 decades waiting for all you guys and the internet?) im glad im not the only one that sees it as it makes me feel that maybe im not so whacko after all to notice things like this. or at least im not alone in the world. since i left the city years ago ive never had a relapse of this strange observational behavior but if i ever go back ill study it with a vengence. all kidding aside, i think that working on little observations like this could actually help rush hour traffic problems, if you could actually get traffic engineers to look at things like the speed limit, not having everybody work 8 to 5, etc. well anyway, thats my 2 cents worth.
holbrook, az USA - Thursday, January 21, 1999 at 01:03:21 (PST)
My apologies for raising your ire, Mr Washburn, and thanks for the response. (You really are a full bottle on the whole subject of optimization.) Now, don't get me wrong, I agree with most of the things you've written. Sorry for confusing you with my previous comments on "traffic flow optimization"; because I'm not talking about vehicle throughput, I'm talking about energy consumption, and I've clearly used the wrong jargon. Let me put my point another way, and see what you think: From my own observations, this "go-slow" technique allows vehicles to travel at a more or less constant (albeit slow) speed along the congested road. The average speed along a section of road is probably no different than it would be in stop-start traffic. But vehicles are operating more efficiently. So in terms of the energy consumed in moving a specific number of vehicles from point A to point B in a specific time, less energy has been expended. Doesn't this mean that we are optimizing energy consumption for a given traffic flowrate? You tell me. And perhaps you can also tell me what the correct terminology for this kind of optimization is? By the way, I leave journal articles to the experts like yourself, Mr Washburn.
John Sanderson <john.sanderson@ eng.monash.edu.au>
USA - Tuesday, January 05, 1999 at 23:08:01 (PST)
For some experimental data on traffic flow and how jams form and propagate, see the article 'Experimental Features of of Self-Organization in Traffic Flow' in Physical Review Letters, vol 81, pp3797-3800, 26 Oct 1998. It can be found in any university library.
Norfolk, Va USA - Thursday, December 17, 1998 at 10:59:07 (PST)
"But what he fails to realise is that this go-slow tactic is an attempt at OPTIMIZING traffic flow on the existing roadway and it does appear to be successful. Optimizing an existing system is as important as designing it in the first place. As an engi neer he should know this." Well, this comment is so ridiculous that it almost doesn't deserve a return comment, but I just couldn't resist. My specialty happens to be freeway operations, and most of my graduate work has dealt with methods for optimizing the traffic flow on freeways. When traffic engineers talk about optimizing traffic flow on a freeway, they usually mean getting the most car s (or sometimes people) through a certain segment during a specific time interval (optimizing in terms of accident reduction is a different story). What I do fail to realize is how one driver can single-handedly achieve this optimization of traffic flow (in terms of vehicle throughput). If you are convinced that this technique works in this regard, I suggest you perform a more rigorous technical analysis and get your study published in a scholarly journal. MAYBE then I will realize the merit of this te chique for traffic flow optimization.
Seattle, WA USA - Tuesday, December 08, 1998 at 18:03:58 (PST)
You are absolutely correct! I am a Police Officer in Seattle, and can tell you that if folks would just leave a bit of space between themselves and the car/motorcycle ahead, traffic jams would surely be reduced. Also, so called fender-benders would b e reduced as they are usually a result of following too closely, and not paying enough attention to ones driving. I've been doing what you suggest for years. I think of the S/B on-ramp to I-5 from 45th in the U District. It'll be bumper to bumper, so I let a space or two open up in front of me, and behold the traffic pace picks up . I've also noticed that if I do this, then 80% of the time the cars behind me start to let others in as well. Keep up the good work.
Steve P. <email@example.com>
Brier, WA USA - Friday, December 04, 1998 at 17:43:34 (PST)
Dear Bill, this is the first time I've seen this site and I found it very interesting. I tried out a "go-slow" tactic in stop-start traffic a few months back. I was actually doing exactly the same experiment you described; trying to keep my vehicle's speed the sam e as the average speed of those ahead of me to iron out the compression waves. Well, sure enough it worked, and vehicles in the lane behind me were soon travelling at a fairly constant speed. I do not believe it resulted in me getting to my destination any faster, but it sure was a lot less stressful. It also means my vehicle and those behind it were operating more efficiently, which has some environmental merit I suppose. I now apply this technique frequently in stop-start traffic. So this is an inde pendent verification of your own work. Lastly, I would like to say that I am most disheartened to hear the some of the views put forward in the comment book by Scott Washburn, the transportation engineer. I accept his viewpoint that people don't all drive in a uniform, rational manner. I als o agree that simply criticising the state of existing roads is not helpful. But what he fails to realise is that this go-slow tactic is an attempt at OPTIMIZING traffic flow on the existing roadway and it does appear to be successful. Optimizing an existing system is as important as designing it in the first place. As an engineer he should know this.
John Sanderson <firstname.lastname@example.org>
Melbourne, Vic Australia - Friday, November 27, 1998 at 02:05:38 (PST)
All the traffic analysis on the site seems exactly correct to me. The one thing I would add to this waves analogy is that you can think of each car linked to the one in front by a spring. As the spring extends (the gap increases) the "pull" between the car increases and the car behind speeds up into the gap. The solution to reduce the impact of waves is to make sure the spring in front of your vehicle is short and stiff i.e. Keep up with the vehicle in front at a constant short distance. Long, elastic springs exacerbate wave problems.
Andrew Blee <andrew@familyblee. freeserve.co.uk>
Fairford, Glos UK - Friday, November 20, 1998 at 12:09:51 (PST)
Well, I just looked up this web site again to see if there were any new "interesting" comments, especially in light of my first comment. While there aren't many new comments, I thought I would address a couple of points brought up in the previous comm ent. First of all, this person asks the question about whether one person/vehicle can disrupt traffic. Well, this point is obvious and really goes without question. I am sure that every driver has experienced a traffic congestion situation that was ini tiated by a vehicle incident (whether they knew it or not). In fact, in the Puget Sound region, 60% of all traffic congestion is due to vehicle incidents (disablements, collisions, etc.). The real question is, "can one driver/vehicle single-handedly IMP ROVE overall traffic conditions?" The answer to this is obviously NO. While driving in a rational manner (i.e., reasonable speed and headway for the conditions) may make traffic conditions a little safer for yourself and those around you (and thus reduce the chances of a congestion producing incident), it will not improve the average travel time of the traffic stream as a whole during the peak period. Again, as mentioned in my earlier comment, if you could get every driver in the traffic stream to use t he same rational driving model, that would be a different story, but this is a driver education issue, and still highly unrealistic even with the best driver education. Barring incidents, congestion is a result of demand exceeding capacity (of course, co ngested conditions often times lead to incidents as well (e.g., rear-end and side-swipe accidents)), and on an individual level, you cannot prevent this or lessen the impact of this condition. To improve traffic conditions, changes in driver behavior hav e to made at a "global" or macro level. Realistically, the only way sure way to do this is to actually take the vehicle control away from the person and have it computer controlled. If all vehicles in the traffic stream were computer controlled, all veh icles would use the same optimized algorithm, and more capacity could be realized from the roadway system and traffic congestion could avoided (this would also entail controlling all entrances to a roadway and controlling the departure time and route choi ces of vehicles). Believe me, I would love to see everyone on the road drive in a rational manner--it would certainly make things better for everyone, and if somebody knows how to accomplish this (while still the individual in charge of the vehicle), I w ould love to hear how. Unfortunately, there is always someone who is late to somewhere (e.g., racing to the hospital for a child birth or injury, or any number of things), and in general people act in their own best self-interest. The issue is one of us er optimal vs. system optimal. Often times, what is best for the individual is not always best for the system. It is difficult (if not impossible) to convince everybody to act in a certain manner such that the conditions to everyone on the whole are at an optimum, if most people can see that acting in a different (self-serving) manner will make conditions better for them individually. My other comment(s) relates to the statement of "being a victim of a poorly setup highway system". This seems to be th e battle cry of commuters in almost every major metropolis. It is interesting how typical commuters are quick to blame transportation engineers for poorly developed roadway systems when they know absolutely nothing about the history of how or why the roa dway system was developed/built the way it was. There seems to be the general opinion by commuters that if transportation engineers weren't complete idiots, everybody would be able to travel wherever they wanted, whenever they wanted, without ever experi encing a second of delay, and without ever paying a penny. Well, wouldn't that be nice, even in dreamland. I don't know much about Jacksonville, but I am sure it is not unlike many other growing cities across the country in a number of ways. First of a ll, when the U.S. interstate system was built back in the 60's (and planned in the 50's), no one ever imagined the population explosion that occurred over the next several decades. And no one imagined the explosion in the popularity of driving for the fu ture population explosion. And even if the system designers did have this foresight, the politicians never would have allowed 12 lane freeways to be initially built because they realized they would never get re-elected after proposing the necessary tax i ncreases to fund this kind of construction. Believe it or not, traffic engineers usually know how to solve a particular traffic related issue, but what you probably don't know is that most times the optimal solution for the traffic condition doesn't get implemented for numerous reasons. Money is obviously a major factor. For some reason, people are amenable to paying for long distance telephone calls, and even water, but when you propose that they actually pay for using the roadway, you might as well t hem to give up their first born. The fact is that transportation financing has not kept pace (or even close to) with the explosion in drivers and driving (vehicle-miles traveled). But for some reason, many people think the driving is a God-given right, and that paying for it, or at least a fair amount for it, is somehow sacrilegious. Of course, politics is another huge factor. Asking politicians to propose revenue producing schemes (i.e., taxes, tolls) for transportation purposes is often deemed as po litical suicide for the politicians. Additionally, special interests also play a role. As a hypothetical example, maybe the freeway that was described was routed over the river so that it wouldn't disturb the neighborhood where a local politician lived if it were routed in the optimal location. And then maybe the bridge was built as a drawbridge so the politician get through in his/her sailboat. Anyway, you get the idea. Another key factor is other interest groups. Since land is not an abundant comm odity, especially in large cities, the design options for a facility are often constrained by the available land choices. And the available land choices become constrained because environmental agencies step in and tell the transportation engineers that they cannot encroach on some environmentally sensitive land (e.g., wetlands) in the vicinity. Or maybe an area of historical/archeological significance (e.g., indian burial ground) is discovered during construction. Anyway, you get the picture. The lis t goes on. To revisit the bridge example, as yourself, if it was built high enough to avoid the use of drawbridge, what other factors would have come into play? Just one aspect of money is certainly a key factor. In order to elevate the bridge, the roa dway that meets the bridge would also have to be elevated. Raising roadways is a very expensive proposition, especially if you do it for a long enough distance so as to avoid steep grades that would limit heavy trucks to the speed of a slow crawl. This i s just one of many significant considerations. So let me assure you, the reason that a roadway gets built with 2 lanes merging into 1, for example, isn't because the engineers are idiots and don't realize that this constrains traffic flow, it is for anot her reason that constrained the design from being optimal in terms of a traffic-only standpoint. Life is full of compromises, and transportation systems are definitely not exempt! I hope this provides a little more clarity and understanding of these iss ues. Sincerely, Scott Washburn, Transportation Engineer
Scott Washburn <email@example.com>
Seattle, WA USA - Wednesday, November 18, 1998 at 14:26:06 (PST)
I stumbled on this page by accident.. browsing Yahoo for traffic/driving related information. I must say I was very impressed when I read it. For the past several months I had been forming theories about traffic and you were able to explain it in a very scientifically sound manner. For those who think that it's a bunch of hogwash and "unscientific", consider that traffic is never 100% exactly the same in a given situation. There is no way you can reproduce to the driver the infamous 10-95 split we have here in Jacksonville. But I have tried Bill's technique and whether I get home any faster or not I don't know, but I feel a heckuva lot better when I get home. We do have our nimrods here in the Southeast, but for the most part most drivers are just normal people. We are victims of a poorly set up highway system. Picture this: I-10 eastbound terminates at I-95. Unfortunately, the southbound exit starts as two lanes (out of four). Then, some brilliant person decided to have those two lanes merge into one approaching I-95 south. Then, (get this, this is the kicker), some brill iant fool routed I-95 over the St. John's river, and the bridge was built so low that it was built as a DRAWBRIDGE. Now picture some rich, yuppy person in his tall sailboat coming towards the bridge, out for a pleasure sail (during rush hour, of course). You guessed it, the bridge goes up, and you wait for this sloooooow boat to pass under, and wait for the bridge to go back down. Generally, it takes a good ten minutes for a sailboat by the time they raise the bridge, the boat gets within two miles, sa ils under, and the bridge goes back down. So the traffic piles up on the southbound side like you wouldn't believe. In addition, half the traffic on the northbound side of I-10 is trying to get over. It's not pretty. But I've found that by taking it easy, leaving space, and so forth makes thin gs go a little smoother. some people get stuck on the southbound side wanting to go north because there is an entrance ramp from a major highway. So I make it a point to let over as many cars as possible and maintain a steady speed and gap. Works prett y well. Traffic moves on the northbound side, not just because of me. I'll tell you what makes it really bad is when a pigeon gets stuck in the mechanism and the bridge is stuck in the up position for several hours. Those are the days you just want to stay off the interstate. What I find is that in most traffic situations, like you have aptly pointed out, that it only takes one car to ruin a moving traffic pattern. No number of lanes on a highway is going to ever alleviate traffic jams. It is practially all driver behavior, and something called "situational awareness". Give yourself room to maneuver and you are giving yourself time to think and react to 99% of the situations that come across. Some fool weaving through slow-moving traffic like a bat out of hell is going to learn that 12 inches is not enough room to stop if traffic comes to a halt. If he's lucky, he'll live to see another day where he does what, the exact same thing that nearly got him killed to begin with. I would like to ask all the naysayers about Bill's theories and ideas one thing. (two things, actually). 1) Is there any doubt that one car stalling, running into a median, blowing a tire, or just driving really slow, can cause a traffic jam, especially in a merge situation? No, there is no doubt. One car, one driver, can cause a traffic blockage like you wouldn't believe. Let's take the northbound exit from I-10. Two lanes across. What if the car blows a tire, spins to block both lanes, and gets hit by oncoming traffic? One car just stopped traffic. The shoulders are wide, but everyone rushing for those shoulders will cause just as many problems. This brings me to question #2. 2)Doesn't it make sense then that one driver starting a trend can also impact the traffic situation? It's not just good karma and altruism and all that crap, but the whole situational awareness thing. A good driver will leave himself time and space to r eact, and be able to decide what to do in that time and space he has made for himself (this applies to "she" as well). It is not an easy practice, but if drivers would practice just a wee bit of patience (And realize, mathematically, going 65-0-65-0 will not get you anywhere any faster than going 35-35-35-35, and burns more gas to boot) and use their BRAINS on the road, not their gas pedal, they would enjoy driving much more. I love to drive. I prefer to drive at night or early, early morning (4 - 5 am), on secondary roads where no one is, so I can enjoy the drive. Rush-hour traffic on an 8-lane interstate is not enjoyable for me. I do relax as much as possible in that situ ation, but I don't enjoy the endless sea of brake lights. Ok, long enough message. silas daniel office associate music division Jacksonville University
Silas Daniel <firstname.lastname@example.org>
Jacksonville, FL USA - Wednesday, November 18, 1998 at 10:40:42 (PST)
Howdy Pard, Try slowin' down a Texan with your "Rolling Roadblock" theory on Loop 610, and you will wind up with the indent of the guy behind you permanently imbedded in your trunk lid. That's if your lucky!!! Hand guns are legal to carry in cars with a permit. The gun must be loaded, or it is considered a club, which is against the law. This cuts down on the chance you will be simply clubbed like a baby seal, as you are being dragged from your vehicle for gettin' in some dudes way as he is on his way to hold up a 7-11. Furthermore, I hear you folks up there in the Nortwest are fueled up on Prosaic. I guess that makes for some purdy sunsets, and a contemplative sensitivity. We like to combine whiskey, beer,and guns with our driving ,so as to be super dangerous. YEEEEHAAAAAA!!!!!!!!!! Regards to Ya'll,and happy trails, PAT PS Love Your Site!!!!!!
Houston, tx USA - Sunday, November 15, 1998 at 20:23:21 (PST)
I have no idea how I stumbled across this site, but I'm glad I did. While I'm not necessarily the most considerate driver at all times, I found myself agreeing with several of your ideas, theories, and suggestions, even to the point of recognizing som e I implement already. I'll be putting a link to your page on my own webpage in the hopes that enough random people get the ideas you've presented and we learn to control the flow of traffic through common sense rather than computers or heavier enforceme nt. Well done! Bravo! Wahoo! And a rousing cheer of 'Neat!' My hat's off to you.
Doc Wilbury <email@example.com>
Sacramento, CA USA - Wednesday, October 14, 1998 at 05:19:53 (PDT)
I just stumbled across this web site, and I must admit, it evoked a number of reactions. First of all, it confirmed what those of us in the business have long contended--that anybody with a driver's license thinks they are a traffic expert. However, I was delighted to see a web site that provided a forum for people to discuss their opinions about traffic and their own personal driving behaviors. Unfortunately, though, I was disappointed to see a web site that was purporting to educate people on traffic flow behavior be based strictly on personal "theories" and absurd personal "experiments". This web site contains a lot of conjecture and opinions about traffic flow and ways to improve it. But, it is very obviously missing solid mathematical and engineering based theories on traffic flow. Now don't get me wrong, I do appreciate a forum for discussion on a topic that impacts many people's lives everyday. However, I am concerned that people that might not know better, which, unfortunately is a lot of the people that have replied to this site, are led to believe that your proposed driving behaviors are what should be adopted by everyone in order to eliminate or minimize traffic congestion. Your personal experiments with speed regulation and headway regulation make for interesting stories, but that is about all. To think that your driving behavior alone in these "experiments" had a significant impact upon the overall level of congestion or the travel time experienced by commuters in this area is ridiculous. Now, it is true that if you could get every driver on the road to employ the exact same rational driving model, traffic congestion would be lessened. But there is the problem, and where your analogy to fluid flow breaks down (which has been explored by several researchers in the past)--every water molecule (i.e., driver) in this case has a different brain, and thus, every driver employs a different driving model. While it would be nice to homogenize driving behaviour (e.g., a surrogate method like vehicle computer control), and thus realize a system optimal traffic flow model, the fact is that people act in their own best self interest, so even the "good samaritan" actions of a few in the traffic stream will be far overwhelmed by the individual self interest actions of the many. Anyway, the fundamental problem with traffic congestion is that there are too many vehicles trying to use the available capacity of the roadway at the same time. And there is very little to nothing your individual games of headway and speed regulation can do to prevent it. As I just mentioned, roadway CAPACITY is one of the fundamental concepts that needs to be understood...and I did not see that mentioned anywhere in your "monograph". When demand exceeds capacity, congestion occurs-- it's as simple as that. Thus, your hypothesis of the wall of state patrol cars pacesetting the traffic is totally absurd. Trust me, this will do nothing more than waste taxpayers' money. Now just think about it, since every possible entry onto the freeway is not controlled (e.g., metering system), it is not possible to prevent demand from exceeding capacity at any particular point along that freeway. Thus congestion could occur anywhere along the freeway where demand exceeds capacity, regardless of where the wall of patrol cars are and what speed they are going. As the demand increases in a freeway section (through increased on-ramp volumes, less smaller off-ramp volumes), people will naturally slow down because of their increased discomfort with decreasing headways as a result of the increasing demand filling in the available capacity of the roadway. Now, I will acknowledge that your attempt to explain a generally complex phenomenon (i.e., shock wave theory) to the lay person is admirable, but you should make sure you thoroughly understand it before trying explain it. Unfortunately this is not the case, as is evidenced by your trying to base your crazy personal traffic flow "theories" upon it. I apoligize if this commentary sounds critically harsh, but I just had to shed some factual light on this subject, based on real mathematical and engineering research. Let me just close by saying this--drive in a manner that you deem rational and leave it at that; no matter how many of your friends you get to drive like you, it just won't make a difference when it comes to changing levels of traffic flow and congestion. Sincerely, Scott Washburn Transportation Engineer
Scott Washburn <firstname.lastname@example.org>
Seattle, WA USA - Tuesday, October 13, 1998 at 20:19:45 (PDT)
I would have to agree with Mr. Hartley from the UK below who states that it would be best to: 1) Get out of your car and onto train, bus, bike, ferry, feet, whatever; 2) Live closer to where you work; 3) Just stay off the road. Traffic is snarled the way it is because growth has outmatched the infrastructure's capacity. And this is only getting worse in the Seattle metropolitan area (which is why I moved across Puget Sound...the ferry ride, while inconvenient, is MUCH less stre ssful than sitting in traffic), and will continue to do so. People MUST get over their 'love affair' with their car and utilize public transportation. And, of course, FUNDING for public transportation needs to be increased. It is time to wake up and realize that adding lanes to freeways WON'T solve the problem. The easiest and least expensive way to solve the problem is to get more people into less space on the roadway.
Chuck McGowan <email@example.com>
Bremerton, WA USA - Monday, October 12, 1998 at 13:37:18 (PDT)
I drive a stick shift. Changing speeds means changing gears. I have intuitively adopted the driving style in rolling backups of guessing the average speed of the car in front of me, maintaining a buffer space, and driving at that average speed. This has always made my driving in traffic more pleasant. No matter what driving style I adobt, I cannot influence the traffic patterns ahead of me. I agree that I can influence the traffic behind me. When I initially slow down to create the buffer space in front of me, I am creating a wave of traffic congestion behind me which is greater than what would have existed had I not slowed down for as much, or as long. Now, as I maintain my constant speed, over time the cars behind me join that speed, and the traffic blockage has been erased. The average speed however, has not increased by my action. The perceived stress of myself, and most likely the drivers of the cars behind me has decreased. Many times, I have imagined what the result might be if a "walking" line of lights were strung along the median. The speed of the cars might be influenced to match their speed through some type of psychological entrainment. If the lights were synchronized to allow for hills, turns, exits, and such, then perhaps the average speed of traffic could be increased, delaying the need of creating additional lanes for growing traffic volume, at a much lower cost. I love your sight. I have built a static charge pump with help from some of your pages. I hope to publish my amateur science projects to my web site in the future. Your site has been an inspiration. p.s. I just noticed my domain name SpringStreet fits quite nicely with the fluid dynamics on roads discussion :-)
Peter Baumbach <firstname.lastname@example.org>
USA - Tuesday, October 06, 1998 at 21:06:14 (PDT)
I think of traffic behaving like longitudinal waves produced by a slinky stretched lengthwise and given a push from one end while the other is held still. Consider traffic lights on main street in a town. If there is enough traffic trying to progress Through town, the traffic pattern will appear (from above) just like the longitudinal waves passing along a slinky
USA - Monday, September 14, 1998 at 20:41:13 (PDT)
I just want to assure everybody that we have exactly the same problem in Belgium and in Europe. The problem is the behavior of some drivers and we can dream about a solution... Nice work but the texts are too long. So that, I purpose to cut them in pieces that we can access by cliking on. A little bit more structure would be cool. Keep cool to everybody!
Liege, Belgium - Wednesday, September 09, 1998 at 04:46:14 (PDT)
well computer controlled cars would probably help but as all computers do what happens in the big system crash????
keath rhymer <kgr1398@ ameritech.net>
indianapolis, in USA - Wednesday, September 02, 1998 at 16:45:45 (PDT)
Excellent theories! But as a professional driver, I have to tell you that some of us have been doing this for years. Next time you're in a traffic jam on the freeway, check out the tractor-trailers that always seem to block all lanes of traffic-you know the ones, they're going about 25 miles an hour, side by side, approaching a traffic jam (or in it), slowing down everyone behind them and creating a large gap in traffic ahead of them. They are working to help dissipate the jam and get traffic moving. Great to see someone else is finally catching on to this concept. Lets all work together to get home sooner-and in one piece.
Steve Cumbey < email@example.com>
Oxford, MI USA - Wednesday, September 02, 1998 at 15:32:27 (PDT)
Most traffic can be avoided by everyone just being smart. Most people that drive along side of me are complete idiots. Everyone wants to go 80 MPH to a RED LIGHT! I don't get mad about this, it just makes me shake my head and say what the hell are you people doing! Alot of traffic problems exist because of lack of respect for others. Not only drivers, but lack a respect for people in general. John
Southfield, MI USA - Wednesday, September 02, 1998 at 09:22:18 (PDT)
Yes! I like it... and it works as long as you can keep a free space... other drivers don't pile in and fill it up. I believe you mentioned that but I didn't get quite why that didn't happen in your situation. I lived in Los Angeles for 6 years, but even around here in Oklahoma the principles exist. Funny thing about "aggressive" drivers, to my observation, they rarely get any further ahead then if they'd just held position and drove sanely and curtiously. It seems they've got their nose too far up the butt of the vehicle in front of them to see what's going on... and so switch into a jam as often as a faster lane, if such even existed. They seem to drive by what I call the 6 foot rule ...if the car in the other lane is 6 feet ahead of the one in front - switch lanes! They seem to have a field of awareness of about 50 feet (maybe not that much on sides and back). They rush madly towards obviously red lights and rarely seem to notice anything more than 3-4 cars away. Social engineering(although that's a pretty vague and slightly ominous sounding term) or hardware based timing schemes, might help. It would be nice if drivers could take a calmer, longer view... That could be benificial in the rest of life also. Unlikely that will happen long as people drive in a rush... mind on other things mostly.
Bartlesville, OK USA - Saturday, August 29, 1998 at 08:15:58 (PDT)
New Yrok, USA - Thursday, August 27, 1998 at 13:22:44 (PDT)
Hi, Ive tried some of these ideas, unfortunatly kiwi drivers generally too pig-ignorant and impatient to allow any gaps to form, they just dive around you. The gereral complex is "i cant let anyone in front of me". cool site.
andyb < firstname.lastname@example.org>
Auckland, New Zealand - Tuesday, August 25, 1998 at 13:01:31 (PDT)
Your moving traffic is strangely hypnotic and I can`t bring myself to leave this site. By the way, if a tree falls down in an empty forest does it make a sound?
London, ? England - Thursday, August 27, 1998 at 04:20:18 (PDT)
Interesting site! Whaaat? Drivers slow down and get a little more space between cars here in Las Vegas. Never. Selective bazooka shooting is the only answer. The way to drive in Las Vegas, go fast as you can and jump up the leader's tailpipe. Then wait for the next movement to occur. This city is nuts.
Ralph Hamilton <email@example.com>
Las Vegas, NV USA - Monday, August 24, 1998 at 14:54:08 (PDT)
don't ask me how i arrived at your site, for i do not know! however, it must be an omen, because although i do not have the time i wish to have to look through all of your site, what i have seen so far is very intersting. ok, i must admit, that what i've seen on the section about traffic jams is something i've always known, it must be enlightning to some ppl who have signed your guestbook. oh, i have to say, it's very scary to think that some ppl who have signed your guestbook ACTUALLY do not understand, or want to acklowledge the facts that you point out! how could one of your guests say that when they raised the speed limit to 65 from 55 there was a subtle difference in traffic congestion?? oh, well, i do have my own little theory about that; when more ppl are being put off the road (example: hospitals, morgues) then of course the traffic will be lighter. my husband set his own speed limit (about 75 miles an hour) and, yes, i believe he did get out of the traffic jams faster, of course, that was because he was on the SIDE of the road, with one of NY's finest..... anyway, one thing to point out is that driving is not totally and completely a physical act. a percentage of driving is mental, and with that, irrational thoughts are bound to breed. not only do motorists not want to be left behind, to trail, but a few, if not most, wants to the ones to get out of the pack, to lead, so to speak. ok, this is getting too long, so to cut it short, until the day comes when we will stop trying to get ahead as individuals, and instead to try to get ahead as a whole, we will never get anywhere, out of the trafficjams included.
niki (aka Madam.nikly) < firstname.lastname@example.org>
USA - Monday, August 24, 1998 at 18:49:38 (PDT)
Having lived in Northern California (and now it is happening more and more here in Omaha due to construction), I have noticed many of the same things you have observed. I do think that things could be better if everyone maintained an "average" speed a nd thought about their travel plans ahead of time. Your I-520 experiment may have worked to some extent if you had a "good" set of drivers behind you, but the biggest thing that causes jams (outside of obstructions from collisions, construction, and on/o ff ramp merges) are poeple's inability to be like one another. You have your offensive and defensive drivers, your slow and fast drivers, your timid and confident drivers, etc... You will also have many "Oh, that's the Maple Street exit, quick, merge ov er there and we will visit Martha!" There are too many dynamics involved with fluid traffic dynamics. If only we could control . . . By the way, I applaud you for your excellent traffic animations!
Omaha, NE USA - Sunday, August 23, 1998 at 10:15:29 (PDT)
I enjoyed your site and your wave theory. We all talk about the "stream" of traffic or the "flow" of traffic, because we intuitively know that the freeway is a conduit for our vehicles, much as a riverbed is that for the water, and that we travel best if we can travel like that leaf gliding effortlessly on top of the water passing around all rocks and obstructions . . . but I have not seen your shock wave theory of traffic jams so nicely presented. Thanks for an enjoyable site.
Albert Hollan <email@example.com>
Houston, TX USA - Saturday, August 22, 1998 at 02:19:13 (PDT)
Cool site. Now if we could only get this info out to the rest of the world perhaps we wouldn't have so many traffic jams. I don't see how anyone could disagree with Bill's arguments or see his logic. I've been trying this "antitraffic" space technique for many years now, before I read this site. It's assuring to know somebody else is doing it.
laura #10 < firstname.lastname@example.org>
Philadelphia, PA USA - Wednesday, August 19, 1998 at 11:38:25 (PDT)
On your Traffic Patterns: I lived in Honolulu for several years in the Eighties. Even then, traffic on the one highway though the city was horrid. What made driving the highway bearable was the "island" attitude of the drivers. In that city (at that time) the dominant philosophy was, "We're all in this together." Traffic always filled all lanes right up until the merging lane ran out. Then the mergers and mergees simply alternated. This traffic pattern was a very firm rule and drivers in both lanes could pretty much count on everyone around them to conform, leading to slow but steady progress. The rare exception was given room for his/her insanity and perhaps a finger shaking through the windshield. (Horn blowing was strictly for real emergencies.) Anywhere, if a car signaled to get over one would open a gap and let it in! Everybody could expect the same courtesy when they had to change lanes. It took me a few finger shakings and nasty looks to learn the rules. After that it seemed the only sane way to drive.
Anna < email@example.com>
Zionsville, IN USA - Wednesday, August 19, 1998 at 07:27:29 (PDT)
While I appaluad your theory, and your attempt to alter the patterns you fail to take into consideration the selfish attitude of some people to drive a far down the merge lanes and then force thier way over in order to gain that few car advantage in line. Look at the I-90 to northbound I-5. There is only one lane at the exit yet two lanes all the way down to the merge point ( one is supposed to be an exit to an off ramp) The other example is I-90 eastbound there is a left lane that everyone knows will end at the tunnel yet people jet down this to bypass slow traffic only to jam into the lane to merge at the tunnel itself, so they have saved time and caused everyone behind to be caught in your "pressure wave". The solution has more to do with social engineering than physics.
Rich Dahm <firstname.lastname@example.org>
Issaquah, WA USA - Tuesday, August 18, 1998 at 15:55:18 (PDT)
Traffic pattern discussion matches my observations and I've been doing this for years. I've also found it very rewarding to "plan" my merges and keep them minimal. Sometimes I spend a few minutes more in traffic, resisting the temptation to jump into a faster lane, but it pays off so much more in peace of mind. Cheers!
Edward Martin III < email@example.com>
Portland, OR USA - Tuesday, August 18, 1998 at 12:49:48 (PDT)
I don't buy it.. I think people don't accelerate fast enough and that is the main reason for traffic jams. See I-5 Southbound right before the 520 exit.
Mark < firstname.lastname@example.org>
seattle, wa USA - Tuesday, August 18, 1998 at 12:04:20 (PDT)
One of the more fascinating Web pages I have ever run across. Thanks so much for it! I have my doubts you will ever see this, contending as you are with the E-mail jam!
R. G. Spindler, M.D. <email@example.com>
Sebring, FL USA - Tuesday, August 18, 1998 at 05:48:49 (PDT)
Interesting, I've always had an inkling that these sorts of processes are at work in traffic, though I think there are some regional variations between UK and US driving habits However, adressing traffic jams is all very well, but all you're doing is treating the symptom, not the problem. The problem is twofold: the excessive amount of cars on the road and excessive journey times caused by urban sprawl/unsustainable community design. The solution is fairly easy. Get out of your car onto train/bus/bike/feet. Live closer to work/other amenities so your journey times are reduced. Cars use an excessive amount of energy and resources and a a major cause of pollution. Use your car less, keep the resources available for other more deserving uses. All the best Mike
Mike Hartley <Michael_Hartley @yahoo.com>
St Albans, Herts UK - Tuesday, August 18, 1998 at 05:28:57 (PDT)
this site has really let me realize the physics of traffic. i think it is really interesting and a cool site.
Bree Coberly <daisy515@ bellatlantic.net>
PA USA - Monday, August 17, 1998 at 16:01:19 (PDT)
This site is very interesting. I think I'm one of the drivers who causes a lot of those traffic jams.
Detroit, MI USA - Monday, August 17, 1998 at 10:39:00 (PDT)
While computerized cars in metro areas would ease traffic flow problems, I believe the rolling wall of state troopers suggested would merely transfer the jam ahead (which would be erased by the time they got there) to behind them. If the average speed on the highway was 70 mph and the wall moved at 55, wouldn't an almost identical jam form behind them over time?
JD Parsons <firstname.lastname@example.org>
Rockwall, TX USA - Sunday, August 16, 1998 at 21:10:28 (PDT)
Yahoo's Picks sent me here.j For years now, I've been saying to anyone who'll listen: If they'd just teach chemistry alongside driver's education, we'd have fewer traffic jams!!! Anyone just needs to understand molecular structures of solids, liquids, and gases to understand traffi c flow! Why hasn't this idea been mainstreamed??? We'd all save hours, days, even months or years. What I'd really like to see is traffic flow control systems on busy interstates in urban areas which take merge/acceleration decisions out of the driver's hands. Put a computer into cars which will be activated by radio signal when drivers pull onto an e ntrance ramp...turn these interstates and our cars into a giant mass transit system. Fewer accidents (if any!), fewer traffic cops, lower insurance costs...on rural interstates, there would be no need...but the systems could be implemented into construct ion zones, or at major accident scenes to eliminate bottlenecks and backups. Somebody take these ideas and make a million or two!
Andrew Brenner <email@example.com>
Whiting, IN USA - Sunday, August 16, 1998 at 13:29:48 (PDT)
Thanks for a great site! I, too, have been trying to even out traffic waves by trying to drive at a more uniform speed than the car in front of me. I could not have explained what I was doing; it just seemed like something good to do. You have given me a clear explaination and heart to continue. Maybe this will be in Driver's Ed courses someday.
Walter Stromquist <firstname.lastname@example.org>
Berwyn, PA USA - Sunday, August 16, 1998 at 06:51:28 (PDT)
The URL is for those with an interest in human behavior and an ability to tolerate matrix algebra.
Aw, c'mon Bill, say you're not serious with your "simple" cure for traffic jams. Surely you just posted this in order to invite people to poke holes in this preposterous piece of illogic. No doubt your head has become muddled while sitting still, fru strated in that jam on the I-5. My guess is that your scheme would result in traffic coming to a complete standstill, but that everyone would be able to change lanes as often as he wishes, as long as he willing to back up. Of course, this is just a gues s, but then so are the results of your proposed scenario. Unless, of course, you have the results of an experiment where you have convinced thousands of people to behave in the manner you prescribe. Naturally, you don't have such experimental results, because the only way to modify the driving habits of this many people would be through the use of somewhat draconian coercion. One must always be suspicious of simple solutions to complex problems of h uman behaviour, especially when it involves surrendering rational self-interest to "altruistic" methods whose results are doubtful. In reality, what we have here is a simplified model of capitalism versus socialism and I think the experimental results ar e fairly conclusive when it comes to these two systems. I therefore suggest you put the atruistic traffic on the left and the self-interest traffic on the right on your animated web page. Actually, there are a few experimentally proven methods of reducing traffic jams. I travel a certain section of the 405 freeway, sometimes several times a day, and I noticed a rather dramatic reduction in traffic jamming when they (are you waiting?) RAIS ED THE SPEED LIMIT! Of course, the speed limit was only increased to 65mph from 55, but the difference was and is quite noticeable. Dare I suggest that raising the speed limit to 75mph would really have a salutory effect? Now, I realize that raising sp eed limits isn't politically correct and won't give you that warm fuzzy altruistic socialist feeling, but it might actually work. This particular rant aside, Science Hobbyist is my very favorite web site and is an example of the internet at its best. I have spent countless hours reading practically everything here and following the endless trail of links leading away to an incredib le array of knowledge and information previously unavailable to most people. THANKS, BILL!
Michael S. Foster <michael.foster @ mailexcite.com>
Los Angeles, CA USA - Thursday, July 09, 1998 at 01:11:28 (PDT)
In regards to your commentary on Traffic Waves, if everyone would take note: traffic jams would clear a lot faster in all cases if people would just slow down and leave at least 4 car lengths of space between their own car and the one in front of them and crawl. As a result you loss the "back wave" so to speak because there are no stopped cars. Also, there is plenty of room for those impatient fools who can't wait in line to cut in front of you with none, or very few implications. Not to mention it lowers your stress level by at least 85% because you still feel like you're getting somewhere since you're still moving. Come on guys lets kill the whole driveway phenomenon and turn our highways back into highways!
Dawn J. V
Shawnee, Ks USA - Friday, February 20, 1998 at 15:35:25 (PST)