Traffic flow

Cecil's Columns/Staff Reports
1

Cecile just sent out a “classic” form 06/17/83, concerning back-ups on freeways.

While I certainly appreciate the answer given (which was informative), I think that there are some other factor(s) at work, besides psychology.

I am not educated in anything like “fluid dynamics”, its my intuition that such might be applied to traffic-flow as well.

As a “for instance”, note how after there is a stoppage, it takes some time to get back up to speed. When drivers stop their cars, they tend to get fairly close to the car ahead of them. And when the traffic starts to move again, drivers wait until there is more space between their cars and the one in front of them. One might hypothesize that when the traffic starts up, that all cars could start off all at once, but in fact it takes some time, between when the front cars take off and when cars farther back can start to move.

Following that thought, one can also intuitively understand (if one visualizes it) how one slow-moving car (such as one just entering the stream), greatly effects the speed of cars behind (not just psychological, but also mechanical/mathematical?). If the car just behind the slow car (going, say 45) had been going 60, and the car behind that had been going 60, on can see how the traffic would more or less collapse like an accordion, reducing distances and speeds, far behind the initial slow-down. The farther back that the traffic stays at something like the maximum density, then the farther back would the collapse take place, and the slower would the cars be able to go as the distance increases from the initial incident, until, at some point, the traffic stops altogether. The the process/phenomenon of “starting up again” takes place.

Maybe some new study could be given to this, investigating research done since 1983 – surely there has been research done by highway departments and such, that verify these observations and hypotheses. Yes?

2

LINK

3

The main cause of freeway stoppages is volume of traffic. The most amount of traffic a single lane can carry is about 2000 car per hour, at about 35 - 40 mph. What happens is you get another on ramp adding cars to the road when it already is at capacity. There is simply no way to increase the number of cars getting through. To stop this completely stopping the traffic, most places now limit the number of cars that can enter at each ramp through lights.

I recently was living in Seattle. Rush hour (two inaccurate words) now lasts from about 6 a.m. to 9:30 a.m, and 3:30 p.m. to 7 p.m. And the hours between 9:30 and 3:30 are damn near capacity, in both directions.

anyhow, if you have 5 lanes, and need to get 50,000 cars through to work, then you need 5 hours to do it in (2000 per lane x 5 lanes x 5 hours). If they all want to be at work between 8 and 9, then u r screwed. Period.

Another interesting concept is pent up demand. The roads get so bad its easier to drive 15 miles through the back roads. So even if you add another lane, when it opens, all the other alternate routes just move a fraction of their traffic over and the system is still overloaded. In most of the big cities its an unwinable battle.

The best thing to do is have gas go back to $4 a gallon. It actually got some people on bikes and transit. The higher the price of gas, the less people use, the less money goes to foreign countries. People survived at $4/gal and the money could go towards fixing bridges, roads and some of it towards the debt.

On a separate note, having low taxes and running a deficit is the same as taking out another credit card when you are maxed out on the one you have now. Sooner or later you have to pay the bills. Personally I am going to move to Canada in 10 years time when it is time to pay the piper.

4

I think the point is that you can have 2,400 cars per lane at 65 MPH (just to pick numbers Cecil used) safely, but people are slowing down at 1,500 cars per lane. Slowing down early. This causes people behind them to slow down, and people behind them to slow down. That’s why the psychological effect.

Certainly, if you can support 2,400 cars per lane, and suddenly have 3,000 cars per lane, then there’s a problem. But Cecil’s article suggests that we’re seeing the problem at 2,000 cars per lane - well before the 2,400 number. So the traffic symptoms exists before the overloading problem occurs.

I agree in principle, but this means that you trust our government to spend extra money where it needs to be spent, or even earmarked. I haven’t seen that ability out of our government. As soon as you can provide us with a government which we can all trust to spend money wisely, I will fully support this idea.

5

It’s kind of interesting how small a change there has to be to slow traffic down. Have you ever been in a situation where there is a traffic jam, where there perhaps normaly wouldn’t at that time of day, and after a while you pass a wrecked (or even just a stopped) car on the side of the road, and after that it flows smoothly…? It seems that every, or at least many enough drivers, slow down slightly to look at the car at the side of the road, and this small change(decrease) in speed is enough to cause a traffic jam. Kinda weird. :slight_smile:

6

Can’t find the article but the Washington post had one a while back on beltway traffic. At certain congestion levels the traffic flows pretty good, at a slightly heavier level traffic slows down a LOT. However increase congestion one more time and traffic starts to flow much better as there are less attempts to perform lane switching to try to get ahead of the flow. It showed that the “fast” drivers constantly moving from lane to lane are actually part of the cause of the heavier congestion as sane drivers constantly brake to avoid hitting them.

7

There are other factors making slowdowns happen. For instance, on ramps and off ramps typically overlap in the merge zones, so people trying to speed up to enter the freeway are trying to mesh with people trying to slow down (well, most of them slow down) to exit the freeway. The jockeying associated with changing lanes and changing speeds bleeds from the outer lane into the inner lanes, causing slow downs. A consistent swarm of vehicles all moving in the same direction might be able to maintain a set spacing and average velocity, but the entrance/exit merging affects that.

Then there’s always the dummy in the “fast” lane driving slower than the peak traffic speed. And the speed demons weaving through traffic. Neither of those do traffic flow any good.

8

The ‘clog’ of traffic from an earlier incident, perhaps involving one or two lanes being closed, doesn’t clear the instant the blockage is removed. It takes time. The wrecked car on the side of the road might have been in the middle of the road an hour earlier, until police had finished gathering evidence. A ‘stopped’ vehicle might have had all sorts of problems you don’t know about - even just a breakdown in the fast lane could have caused a big tailback.

9

the road is like a pipe. It will only allow so much water through. more than this is backs up. studies have shown that the optimum speed for getting the maximum amount of traffic past a point is about 40 mph. As speeds pick up drivers spread out more for safety reasons. This is both perceived safety and actual safety. The closest you can follow another vehicle and have any chance to react to braking or something is about a second. At 60 mph this is about 88 ft. Safety groups say you should actually be about 2 seconds, but that works out to 176 feet and about 1800 cars per lane per hour. The maximum capacity that u can ever get out of a lane in practice is about 2400 vph. The problem is once the system breaks down to stop and go traffic, capacity drops down to about 12-1400 vph. I haven’t done a lot of this stuff in the last 20 years (i build highways i dont design em)

10

Usually every traffic jam on Interstate highways I see is caused by a merging situation. There’s not much that can be done for incoming traffic, but for exiting traffic I am fairly convinced if they could design two lanes to be “Through traffic” lanes and stick to it, there’d be much less backups. When you’ve been driving for an hour and suddenly your lane turns into an exit, that’s bound to cause problems when traffic is dense.

11

Seattle has a non-typical traffic problem, and shouldn’t be used as an example. The issues here are cultural, not scientific. For example, in Seattle, a driver will come to a complete stop in the middle of the road and block 100 cars behind him to allow someone else to merge, or to allow a pedestrian to cross against the light. Seattleites actually talk about this tendency proudly, as an example of how much friendlier people are in this area, ignoring how rude it is to the drivers being blocked, or how badly it screws up traffic patterns.

Seattle drivers are also the worst for blocking the passing lane. I once drove from Detroit to Seattle via route 80/84, and I kept a scratchpad next to me the entire trip. Every time I saw a driver in the left lane who was not passing anyone, I put down a check mark. I got all the way to Oregon and only had 12 total, never more than 4 per state. In Oregon I saw 25, but 18 of them had Washington plates. I turned north from Portland into Washington and after 1 hour I had lost count around 60… and I still had 2 hours to go before I got to Seattle, where the problem is worst of all.

I’ve driven through rush hour in Chicago, Detroit, Cincinatti, Albany, Boston, Dallas, Tampa, Las Angeles, San Diego, and probably a bunch more I can’t remember and none of them have Seattle’s sloppy driving standards. </rant>