The mechanics of a traffic jam

[mozirry]

I drive in the busiest part of I-270 in St. Louis every day, in and out.

I have the hardest time trying to figure out why the traffic will come to a standstill...speed up real fast...come to a standstill...speed up real fast...come to a standstill...for around 12 miles of highway, when there is no real accident or anything that should cause us to stop.


Now, the mechanics of a line... If we are all standing in a line shoulder to shoulder, if the person stops in front of you, it will cause a chain reaction, forcing the person behind you to stop, the person to behind you to stop, etc. etc.. Then, as the person infront of you leaves, you don't leave immediately, you take a few seconds to let your leader get some distance, and then you leave.

The line basically leaves slower then it forms, so you have more people stopping then leaving.

What I don't get is, why the chain reaction of going, and stopping, going , and stopping?

That means someone at the front of the line is stopping and going, for what reason we may never know.

----------
Anyway, here's my theory.

What if someone in the line just doesn't stop? If that one person never stops, then the people behind will follow and never stop.

The biggest problem with traffic jams is that once a vehicle comes to a standstill, it takes a much longer time to bring the car back up to speed. Now, a vehicle that slows its momentum (maybe to 1-3 mph) rather then halting its momentum, will bring itself back up to speed much faster than a completely stopped car.
-----------

In practice, I for the last few days have kept some distance between me and the car infront of me during a traffic jam, and driven at a steady pace. Basically, I drove around 15 mph with a huge gap infront of me while the rest of the cars in the other lanes would speed up quickly to 30 mph, and then come to a complete halt when they meet the car infront of them.

I quickly noticed that although people perceive the large gap as something to speed into, speeding into it and then coming to a complete halt will actually cause a chain reaction that will completely screw everyone else in the line.

So, as I peddle around around 15mph, I will slow my momentum as I come to the stopped car infront of me, and by the time the car infront of me leaves, I will move my speed back up to 15mph, so I continue to move my car and never actually come to a complete stop.

---------

The difference?

The line behind me is moving at a steady pace, never stopping, never losing time with a chain reaction of complete stops.

The other lines which continue to speed up to the stopped car infront of them continue the chain reaction down the line and actually end up losing time.

----------

How will this work in reality?

The best way to implement this is to, during the inspection year for tax and license purposes for each vehicle, mandate through state law that each vehicle is attached with a bomb that will explode the drivers side cabin if the vehicle ever comes below 1mph.

The bomb will only be activated via a sensing device as it enters a on-ramp and will then be de-activated when it exits the off-ramp, thus only putting this device to use on our main highways and not actually exploding drivers at red-lights, where it is illegal to stop moving.

 

[bobsmith1492]

Hm, you're probably right about the chain reaction thing. Hey, you should read the book "theory of constraints" - he learned a lot about his business by watching that same sort of thing happen.

Not so sure about the bomb idea though... What inspections are you talking about, anyway? Are you from California or something?

 

[mozirry]

They just passed the "St. Louis" clean air program that forces you to run your exhaust through inspections every time you have to renew your plates. I think it might just be a county thing.

Basically, it's just a big funnel for revenue without having to raise income taxes =)

 

[Gibsons]

Have you ever been driving on an interstate highway when traffic suddenly slows to a crawl? You inch along for many minutes while waiting to see the accident which must have caused the jam. At the same time you also curse the "rubberneckers" who are causing the whole problem. But then all the cars ahead of you take off at high speed. The jam is over, but no accident, no police cars, nothing. WHAT THE HECK WAS THAT?!! A traffic jam with no cause? In the rear-view mirror you see all the poor saps behind you still stuck in the jam. But why? If all those people could just speed up at the same time, the whole traffic jam would evaporate. Why don't they ever do that? What caused the mysterious slowdown in the first place?

from here

 

[CTho9305]

Originally posted by: Gibsons

Have you ever been driving on an interstate highway when traffic suddenly slows to a crawl? You inch along for many minutes while waiting to see the accident which must have caused the jam. At the same time you also curse the "rubberneckers" who are causing the whole problem. But then all the cars ahead of you take off at high speed. The jam is over, but no accident, no police cars, nothing. WHAT THE HECK WAS THAT?!! A traffic jam with no cause? In the rear-view mirror you see all the poor saps behind you still stuck in the jam. But why? If all those people could just speed up at the same time, the whole traffic jam would evaporate. Why don't they ever do that? What caused the mysterious slowdown in the first place?

from here

That article is really good... I read it a couple years ago when I was spending a lot of time in traffic. It's a lot less stressful to drive with a long gap than to drive like everybody else.

 

[dkozloski]

Keeping things moving is also a great tactic on slick and icy roads. Coast very slowly towards the redlight at the intersection and it's much easier to speed up as the light turns green. Watch the idiots spinning and sliding as they hurtle toward the intersection. Keep a big gap behind the person ahead at the intersection so you can watch the rear view mirror as some moron tries to hit you in the rear before you can pull ahead or change lanes.

 

[jersiq]

These mechanics have fascinated me as well, and living in Houston, I have plenty opportunity to ponder it while I sit in my own traffic jams!

I read a similarly worded excerpt from James Gleicks "Faster", which gives a similar treatment of traffic jams. Most times, all that it would take is a person slowing down (10-15 mph) to let another person into his lane to create problems for the people behind them.

My friend and I have a reoccuring argument as to whether it would be better to just give up control of the vehicle to a computer handling all the traffic. It's a rough sketch, but I would propose that some sort of magnetic system that would allow cars to change lanes. All the cars acceleration and movements would be controlled by distributed processors, thus removing the boneheaded mistakes that people make on the roads. I argue that this would alleviate, if not eliminate "rush hour" by having the cars act more uniformly in entering, exiting, and maintaing course on a highway. It would also give you better estimates for drive times that the current "passive" RFID readings that are used in Houston today. (I don't know if other cities use this)

A little pipe dreamish on my part, but hey I have plenty of time in traffic to think.

 

[dkozloski]

jersiq, after reading about the mishap the Germans had with the MAGLEV train I want to be making my own decisions. I also remember reading about the cockamamie scheme that was originally used to control the BART sytem in San Fransisco. That was really scary. It was commands sent into the blind with no verification and hoping they got to the target.

 

[CTho9305]

My friend and I have a reoccuring argument as to whether it would be better to just give up control of the vehicle to a computer handling all the traffic. It's a rough sketch, but I would propose that some sort of magnetic system that would allow cars to change lanes. All the cars acceleration and movements would be controlled by distributed processors, thus removing the boneheaded mistakes that people make on the roads. I argue that this would alleviate, if not eliminate "rush hour" by having the cars act more uniformly in entering, exiting, and maintaing course on a highway. It would also give you better estimates for drive times that the current "passive" RFID readings that are used in Houston today. (I don't know if other cities use this)

With a single controller, all cars could accelerate at the same time, so waves probably wouldn't happen - waves arise largely from the delay in acceleration of people stuck in the wave.

 

[CycloWizard]

I've long thought about traffic as a fluid mechanics problem as well, though usually I think about the 2-D problem of multiple lanes. Ideally, a velocity gradient would exist across the lanes with 'partial slip' at the extreme right side (exit lanes) and 'complete slip' at the far left. Thus, traffic flow would behave in an orderly fashion like laminar flow, with the maximum velocity achieved in the far left lane and a minimum velocity in the far right lane with a parabolic velocity profile across the lanes. Instead, when people drive like idiots (as they like to do here in St. Louis), they create interactions between the flow lamina, inducing unstable/chaotic conditions and the equivalent of 'turbulent' flow. In turbulent flow, the velocity field is approximately uniform across the entire cross-section (all lanes) and fluid particles (i.e. cars) move randomly between lanes.

I've also thought of this in conjunction with mass transfer, where incoming and exiting cars are diffusing species, but no time for that right now.

 

[Genx87]

Heh as a person who has a long commute and sits there pondering the reasons why we go from 0 to 45 back to 0. I call it the slinky effect. Somebody stops and the chain reaction goes through the entire line of traffic. As to the resons why? I think there are probably people who drive too fast and are forced to slow down, this send the slinky wave back but eventually with the compression of the cars it causes cars to come to a complete stop down the line.

I have tried what you did by driving a steady slower pace to even it out, sometimes it works, other times you do get people who speed in front of you and fill it in.

Another theory of mine is how the roadways can mimic a free market.
As you can see, I have ample amount of time to think about these things

 

[Qriz]

I think you guys are thinking about this too much. "Traffic congestion is a road condition characterized by slower speeds, longer trip times, and increased queueing." Though traffic patterns are obviously very complicated, the reason for any sort of traffic jam is simple: there is a reduced flow of traffic.

As far as I'm concerned, this could be caused by either too much traffic OR a disturbance that wouldn't normally be there, like construction or an accident. When there's too much traffic for the roads (obviously a big problem in cities,) there are too many cars trying to get to their destinations. Individual situations arise (caused by road rage, confusion) that back up the flow even further. Obviously, accidents and construction are both a distraction and drivers are sometimes instructed to go slow when driving through both constuction and accidents.

That's pretty much all there is too it. Individual jams have their own specific characteristics and currently, there isn't some magical way to solve the problem (computer-driven cars...that's a BIG change.) Forget the velocity gradients (although it's worth mentioning) and stick to common sense. You can't always come up with a complicated explanation for a simple situation.

 

[CycloWizard]

Originally posted by: Qriz
I think you guys are thinking about this too much. "Traffic congestion is a road condition characterized by slower speeds, longer trip times, and increased queueing." Though traffic patterns are obviously very complicated, the reason for any sort of traffic jam is simple: there is a reduced flow of traffic.

As far as I'm concerned, this could be caused by either too much traffic OR a disturbance that wouldn't normally be there, like construction or an accident. When there's too much traffic for the roads (obviously a big problem in cities,) there are too many cars trying to get to their destinations. Individual situations arise (caused by road rage, confusion) that back up the flow even further. Obviously, accidents and construction are both a distraction and drivers are sometimes instructed to go slow when driving through both constuction and accidents.

That's pretty much all there is too it. Individual jams have their own specific characteristics and currently, there isn't some magical way to solve the problem (computer-driven cars...that's a BIG change.) Forget the velocity gradients (although it's worth mentioning) and stick to common sense. You can't always come up with a complicated explanation for a simple situation.

Your entire premise is that 'cars slow down when something bad happens or the road is too crowded'. We already knew that. We are interested in the mechanics of said slowdowns and how they can be mitigated. Those people studing traffic engineering (which is a very real field) would beg to differ with you when you say the flow of traffic is some trivial problem. We are looking for solutions to the problem while you're just saying 'Yep, there's a problem. Oh well.'

 

[Qriz]

Originally posted by: CycloWizard
Your entire premise is that 'cars slow down when something bad happens or the road is too crowded'. We already knew that. We are interested in the mechanics of said slowdowns and how they can be mitigated. Those people studing traffic engineering (which is a very real field) would beg to differ with you when you say the flow of traffic is some trivial problem. We are looking for solutions to the problem while you're just saying 'Yep, there's a problem. Oh well.'

*Sigh.* People on forums get mad when you make reference to one of their posts while disagreeing with them, even if not directly (no matter how ambiguous the disagreement may be, and in this case it most certainly was.)

The fact of the matter is that we KNOW what causes traffic jams (as we've all touched on by now, I know.) We all have our nice metaphors, lengthly explanations:

Originally posted by: CycloWizard
I've long thought about traffic as a fluid mechanics problem as well, though usually I think about the 2-D problem of multiple lanes. Ideally, a velocity gradient would exist across the lanes with 'partial slip' at the extreme right side (exit lanes) and 'complete slip' at the far left. Thus, traffic flow would behave in an orderly fashion like laminar flow, with the maximum velocity achieved in the far left lane and a minimum velocity in the far right lane with a parabolic velocity profile across the lanes. Instead, when people drive like idiots (as they like to do here in St. Louis), they create interactions between the flow lamina, inducing unstable/chaotic conditions and the equivalent of 'turbulent' flow. In turbulent flow, the velocity field is approximately uniform across the entire cross-section (all lanes) and fluid particles (i.e. cars) move randomly between lanes.

Thank you for explaining how highways are supposed to work. Although a simpler explanation would be

"There are multiple lanes in a highway. The leftmost lane tends to have the fastest-moving traffic, while the rightmost lane tends to have the slowest traffic due to exit ramps being placed on the right side. Any middle lanes tend to move at speeds in-between the outer lanes. The closer it is to the left the lane, the higher speed traffic tends to move in a given lane. Sometimes, these tendencies are not demonstrated. This is mostly caused by drivers who, for whatever reason, do not follow them."

We probably all know this. And I think the word "trivial" was misused above. Never did I say that the problem was trivial (i.e of little importance.) I'll assume you meant to use it in place of something like "simple" as in "unsolvable in an obvious (simple) way."

I'm not trying to say it's an unsolvable problem (did I?) I DID say "You can't always come up with a complicated explanation for a simple situation." What we have, as far as I'm concerned, is a relatively simple situation: too many cars on the road, accidents/construction causing jams and "rubbernecking" is also worth mentioning. We have a simple situation that's going to require an extremely complicated resolution. And likely, it's not going to be a single, revolutionary action. That isn't what traffic engineers (of course it's a real field! I respect all civil engineers!) are looking for. It'll be a continuing series of regulations in the interest of making our driving easier. Well, it has been, hasn't it? Roads have been around for a while.

And my guess is our musings aren't touching on something (worth mentioning) that they haven't. No offense to the creator meant here- I know that is not specifically what the topic was for.

SO. As Thoreau says: "Simplify, simplify." Which I hardly demonstrated in my enourmous post.

 

[imported_Seer]

Traffic engineering is a very complex problem with, yes, complex solutions. Cyclo was merely responding to your post. Yes, he did so rather heatedly, but there's no reason to pretend to be a victim.

 

[SickBeast]

The causes of the "waves" are twofold:

1. Cars weaving in and out of lanes.

2. Cars entering and exiting the freeway.

The cause of a jam is usually an accident or adverse weather, but that's just common sense.

The real solution is effective public transit. Endless roads lead to urban sprawl with "bridge bottlenecks".

IMO, most of Europe is an example of effective and responsible urban design.

Do cars pollute the most when they're idling? I wonder if there is a way to reduce that beyond simply applying a filtration system. The hybrid is a nice thought, but battery production is not very environmentally conscious as far as I know.

 

[CycloWizard]

Originally posted by: Qriz
Thank you for explaining how highways are supposed to work. Although a simpler explanation would be

"There are multiple lanes in a highway. The leftmost lane tends to have the fastest-moving traffic, while the rightmost lane tends to have the slowest traffic due to exit ramps being placed on the right side. Any middle lanes tend to move at speeds in-between the outer lanes. The closer it is to the left the lane, the higher speed traffic tends to move in a given lane. Sometimes, these tendencies are not demonstrated. This is mostly caused by drivers who, for whatever reason, do not follow them."

We probably all know this. And I think the word "trivial" was misused above. Never did I say that the problem was trivial (i.e of little importance.) I'll assume you meant to use it in place of something like "simple" as in "unsolvable in an obvious (simple) way."

I'm not trying to say it's an unsolvable problem (did I?) I DID say "You can't always come up with a complicated explanation for a simple situation." What we have, as far as I'm concerned, is a relatively simple situation: too many cars on the road, accidents/construction causing jams and "rubbernecking" is also worth mentioning. We have a simple situation that's going to require an extremely complicated resolution. And likely, it's not going to be a single, revolutionary action. That isn't what traffic engineers (of course it's a real field! I respect all civil engineers!) are looking for. It'll be a continuing series of regulations in the interest of making our driving easier. Well, it has been, hasn't it? Roads have been around for a while.

And my guess is our musings aren't touching on something (worth mentioning) that they haven't. No offense to the creator meant here- I know that is not specifically what the topic was for.

SO. As Thoreau says: "Simplify, simplify." Which I hardly demonstrated in my enourmous post.

Thoreau was obviously not an engineer. You are also obviously not an engineer. If you were, you'd know how drawing analogies as the ones I drew previously have led to knowledge catalysis and vast improvements in the understanding of seemingly unrelated fields. The only words in your post that seem relevant are "as far as I'm concerned". You're obviously not very concerned, so you decided to try to belittle those who are. Unfortunately for you, in doing so, you succeeded in demonstrating your own ignorance of the power of scientific analogy. Just a couple examples:

1. Adolf Fick postulated his laws of diffusive mass transfer by analogy with Fourier's laws of heat conduction.
2. Findings from Social Network Analysis (SNA) - the study of how people are linked to other people - led to the development of a revolutionary computer search algorithm. (Article)
3. The differential equations governing mass, heat, and momentum transfer (Navier-Stokes Equations) are easily derived from one another by analogy. They were originally developed as an extension of the Navier Equations - the equivalent equations for the theory of elasticity.

So no, I don't think it's correct to say that keeping things as simple as possible by ignoring possible parallels between fields of study is the correct approach to every problem. To suggest that it is correct is merely ignorance of the histories of science and engineering.

 

[makken]

Originally posted by: mozirry
----------
Anyway, here's my theory.

What if someone in the line just doesn't stop? If that one person never stops, then the people behind will follow and never stop.

The biggest problem with traffic jams is that once a vehicle comes to a standstill, it takes a much longer time to bring the car back up to speed. Now, a vehicle that slows its momentum (maybe to 1-3 mph) rather then halting its momentum, will bring itself back up to speed much faster than a completely stopped car.
-----------

In practice, I for the last few days have kept some distance between me and the car infront of me during a traffic jam, and driven at a steady pace. Basically, I drove around 15 mph with a huge gap infront of me while the rest of the cars in the other lanes would speed up quickly to 30 mph, and then come to a complete halt when they meet the car infront of them.

I quickly noticed that although people perceive the large gap as something to speed into, speeding into it and then coming to a complete halt will actually cause a chain reaction that will completely screw everyone else in the line.

So, as I peddle around around 15mph, I will slow my momentum as I come to the stopped car infront of me, and by the time the car infront of me leaves, I will move my speed back up to 15mph, so I continue to move my car and never actually come to a complete stop.

---------

The difference?

The line behind me is moving at a steady pace, never stopping, never losing time with a chain reaction of complete stops.

The other lines which continue to speed up to the stopped car infront of them continue the chain reaction down the line and actually end up losing time.

Interesting theory, but here's how I see it working in real life:

The car in front of you speeds up to say 40mph, and has to come to a complete stop every few seconds. You on the other hand, leave a large margin so you can always maintain 15mph regardless of what the person in front of you is doing.

Now, when the car in front of you starts moving from a complete stop, the gap between you and him starts widening. This is an open invitation for a car in the next lane to come in and fill that gap. This will cause you to have to slow down to maintain the same gap.

I do in fact follow your method (keeping a gap and going at a steady pace), but i do it for the sake of saving my brakes and clutch, not time. Whenever a gap opens up in front of me, it is almost immedately filled by cars in the lane next to mine, causing me to coast for a little bit to reduce speed.

 

[imported_inspire]

Does anyone know of a good place to look to learn more about modeling traffic?

 

[SickBeast]

Originally posted by: inspire
Does anyone know of a good place to look to learn more about modeling traffic?

A school of Urban Design or perhaps Architecture.

 

[imported_inspire]

Ah, except I've got my hands full with grad school right now, and I'm not interested enough to go to class to learn about it - just wondering really. Any other ideas? Books, Journals, Websites?

 

[Qriz]

 

[Qriz]

Originally posted by: CycloWizard
Thoreau was obviously not an engineer. You are also obviously not an engineer. If you were, you'd know how drawing analogies as the ones I drew previously have led to knowledge catalysis and vast improvements in the understanding of seemingly unrelated fields. The only words in your post that seem relevant are "as far as I'm concerned". You're obviously not very concerned, so you decided to try to belittle those who are. Unfortunately for you, in doing so, you succeeded in demonstrating your own ignorance of the power of scientific analogy. Just a couple examples:

1. Adolf Fick postulated his laws of diffusive mass transfer by analogy with Fourier's laws of heat conduction.
2. Findings from Social Network Analysis (SNA) - the study of how people are linked to other people - led to the development of a revolutionary computer search algorithm. (Article)
3. The differential equations governing mass, heat, and momentum transfer (Navier-Stokes Equations) are easily derived from one another by analogy. They were originally developed as an extension of the Navier Equations - the equivalent equations for the theory of elasticity.

So no, I don't think it's correct to say that keeping things as simple as possible by ignoring possible parallels between fields of study is the correct approach to every problem. To suggest that it is correct is merely ignorance of the histories of science and engineering.

Well now that you've decided to rant about people "ignoring possible parallels between fields of study" I'll have to respond, and clarify.

I don't remember expressing this opinion. I never said connections between subjects are irrelevant or useless, and I don't know where you got that. Because obviously, as you mentioned, it is useful. I didn't say "we need to keep things as simple as possible." I quoted Thoreau urging simplification, because in your case it seems you tried to make it as complicated as you could.

However, the special little "analogies" that you talked about don't contribute anything to the conversation. Representing the movement of a projectile with a parabola is helpful, because it's easier to manage and represent. You can take your known factors and use them to calculate others, and this can be said for the examples you cited. Applying fluid mechanics to traffic patterns is completely useless to us. We're dealing with the real world here. You don't see drivers shouting at each other "Hey! You're supposed to go 1.2719 times faster than me! You're 5 thousandths off!"

No one is piloting that projectile in its parabolic motion, trying to keep it on track. There isn't a little person inside a computer that runs that algorithm and messes up sometimes. There are specific, measurable factors in those situations that don't vary.

You can't solve the traffic problem by simply describing how it's SUPPOSED to move. Even in good road conditions traffic won't move the way you described it. They aren't going to pass regulations that ask drivers to understand and take actions that conform with a complicated explanation like yours, even if that's what traffic engineers used to solve the problem. They're going to make it a bit simpler.

Yes, I notice that we're in the "Highly Technical" section of a technology website. But the tropic is "the mechanics of a traffic jam." You'll notice that the observations thus far have been fairly simple, because that's all they need to be. For example the creator and bobsmith mentioned the "chain reaction effect." Jersiq observed that "Most times, all that it would take is a person slowing down (10-15 mph) to let another person into his lane to create problems for the people behind them." We're looking at the conceptual physics.

Individual traffic jams and their causes are extremely unique. You'll notice that they aren't solving traffic problems with velocity gradients. They're updating traffic management. They're broadcasting road conditions. They're making public transportation cheaper and more convenient. They're building new roads and enlarging old ones. THESE solutions can be applied. YOUR attempts to impress us cannot. I hope you contributed more in your other 5000 posts than you did here.

 

[CycloWizard]

Originally posted by: Qriz
Well now that you've decided to rant about people "ignoring possible parallels between fields of study" I'll have to respond, and clarify.

I don't remember expressing this opinion. I never said connections between subjects are irrelevant or useless, and I don't know where you got that. Because obviously, as you mentioned, it is useful. I didn't say "we need to keep things as simple as possible." I quoted Thoreau urging simplification, because in your case it seems you tried to make it as complicated as you could.

No, I could make it much more complicated. You seem to have missed the boat though, perhaps because you're new here. There are a lot more people here familiar with fluid mechanics than there are traffic engineers. In fact, I don't think we have any traffic engineers. Thus, such analogies add substantially to the understanding of the issues involved in traffic flow patterns.

However, the special little "analogies" that you talked about don't contribute anything to the conversation. Representing the movement of a projectile with a parabola is helpful, because it's easier to manage and represent. You can take your known factors and use them to calculate others, and this can be said for the examples you cited. Applying fluid mechanics to traffic patterns is completely useless to us. We're dealing with the real world here. You don't see drivers shouting at each other "Hey! You're supposed to go 1.2719 times faster than me! You're 5 thousandths off!"

The 'special little analogies' are simply linear constitutive relations that hold true for many, many engineering problems. In fact, they are not actually 'laws', but the name 'law' is still applied because they hold in so many cases and conditions that it took a good deal of digging for people to realize that they were not universally applicable in all situations. These laws do very much hold up 'in the real world here', despite your reckless disregard for their applicability. You fail to appreciate that there are real, physical reasons for using a parabola to describe the movement of a projectile and the velocity field of laminar flows. Yet, despite your disbelief, entire books have been written detailing these subjects. I've spent the last 8 years of my life learning all about them. It's been quite enlightening. So, instead of telling me how useless the last 8 years of my life have been because we're talking about 'the real world here', maybe you should invest your time in learning about the subject a little more before you try to denigrate someone else's views.

No one is piloting that projectile in its parabolic motion, trying to keep it on track. There isn't a little person inside a computer that runs that algorithm and messes up sometimes. There are specific, measurable factors in those situations that don't vary.

You can't solve the traffic problem by simply describing how it's SUPPOSED to move. Even in good road conditions traffic won't move the way you described it. They aren't going to pass regulations that ask drivers to understand and take actions that conform with a complicated explanation like yours, even if that's what traffic engineers used to solve the problem. They're going to make it a bit simpler.

Models like those I suggested are how traffic engineers design roads. They make assumptions as to how chaotic traffic will be, how fast the speed limit is relative to the safe exiting speed, and so on. My explanation is actually a gross oversimplification, not a complication. Of course, you're obviously not an engineer, so I wouldn't expect you to understand the utility of engineering assumptions, basic principles of design, or simple modeling techniques.

Yes, I notice that we're in the "Highly Technical" section of a technology website. But the tropic is "the mechanics of a traffic jam." You'll notice that the observations thus far have been fairly simple, because that's all they need to be. For example the creator and bobsmith mentioned the "chain reaction effect." Jersiq observed that "Most times, all that it would take is a person slowing down (10-15 mph) to let another person into his lane to create problems for the people behind them." We're looking at the conceptual physics.

Individual traffic jams and their causes are extremely unique. You'll notice that they aren't solving traffic problems with velocity gradients. They're updating traffic management. They're broadcasting road conditions. They're making public transportation cheaper and more convenient. They're building new roads and enlarging old ones. THESE solutions can be applied. YOUR attempts to impress us cannot. I hope you contributed more in your other 5000 posts than you did here.

I'm not trying to impress anyone. I simply offered an additional viewpoint to those who were interested in extending their view from the horrendously oversimplified 1-D model to something a little more advanced - a 2-D model. This is the natural progression of modeling. I'm sorry that you feel such e-penis envy that you had to demonize me for trying to add something that might be of interest to some of the members here. Next time you want less information instead of more, please let me know ahead of time so I can accommodate your quest for less knowledge and diminished understanding.

 

[jagec]

Originally posted by: makken

Now, when the car in front of you starts moving from a complete stop, the gap between you and him starts widening. This is an open invitation for a car in the next lane to come in and fill that gap. This will cause you to have to slow down to maintain the same gap.

In my experience, most people don't accelerate fast enough from a dead stop to make a sizeable hole, and if traffic is that packed, the people in the next lane won't be going at a very different speed anyway. Will people change lanes into the gap? Sometimes, but usually not enough of them to force you to slow down further. Also, while going a perfectly steady speed is ideal, by simply reducing the magnitude of your acceleration or deceleration (that is, reducing rather than eliminating the wave), you can usually keep the lane changes down.