BBO Discussion Forums: pet peeve thread - BBO Discussion Forums

kenberg, on 2014-July-18, 19:43, said:

That's the same here. Reminds me of a time many years ago when I was living in Milwaukee. I was in my twenties and learned that there was a weekly duplicate game at the downtown YMCA. After the first time I played there, I walked out to the intersection with some of the other players. There was not a car to be seen anywhere, so I started to cross the street.

I heard a woman call in a stern voice, "Young man, it says DON'T WALK!" I rejoined the group on the curb.

Here we have up to two seconds (depends on the intersection) where it's red for everyone, so you get the best of both worlds

Two seconds is nothing. I can't even count the number of times when I was at a red light and everybody in view was stopped for what seemed like almost a minute, and I had to wonder who on earth we were all waiting for.

Winstonm, on 2014-July-18, 08:03, said:

Which is why we have traffic scientists who do the thinking.

An example of the results of their thinking (and investigating and calculating):

kenberg, on 2014-July-18, 19:43, said:

This may work, but it doesn't work best, as traffic scientists have found out quite a while ago.

They have determined that the best way to solve these merging problems is by 'zipping': Everybody stays in their own lane until very close to the end. The only thing drivers do is gradually increase the distance to the car in front of them, so that another car will fit in between. Then, at the point where the lane ends, the two lanes merge like a zipper: Car from the left, car from the right, car from the left, car from the right... Everybody keeps driving, no need to apply the brakes.

This method maximizes the capacity of the road:
- the whole road is used until the very end. With the "blocking method", effectively the merging takes places a long distance before the place where the lane actually ends. This means that any traffic jam will start earlier.
- there is a smooth, continuous reduction in speed without braking. This avoids the formation of traffic jams. Traffic jams in heavy traffic are caused by disturbances in traffic: braking, accelerating. With the blocking method, the merging point is moving in a saw tooth pattern: the merging point is behind the blocking car that is moving forward, until the blocker merges and the merging point is at the location of the next blocker. This repeated shifting of the merging point leads to "propagating ripples" in traffic density which cause traffic jams.

So, in Europe, you will see traffic signs that tell you NOT to merge until instructed by a sign "zipping starts here". If a "blocker" (who uses the technique described by Ken) gets caught by the police he can count on a hefty fine. I think, though, that this is not the main reason why people "zip". People are aware that when we all "zip" correctly, we all will be home earlier.

Rik

Trinidad, on 2014-July-19, 01:22, said:

There are times that traffic engineers can do a lot.The situation that I described has changed much for the better within the past year but it took some construction. A mile or so past the merge there had been a traffic light. This limited capacity. It would allow X number of cars through at a time and the only issue was to get everyone properly in line. The line extended from the light back to and beyond the merge. The ad hoc solution we found worked fine and since there was no way to get more cars through the light, nothing could work better. The obvious solution was an overpass and they built it. Of course some of the cars want to get onto 32, not pass over it, and we accommodate them with fair ease. I did not expect the improvement to be nearly as effective as it was.

Perhaps there is a larger point from this. Sometimes the road is at or above capacity. Nothing can be done to move the cars more quickly until some physical change takes place. In these circumstances, as long as everyone comes to some sort of understanding, we are all fine. We just wait our turn. That was the case with the merge on 32 before the overpass, and it was the case when I was on the Brooklyn-Queens Expressway. A cooperative approach can go a long way in keeping everyone calm and keeping things fair. In other cases, below capacity, the total travel time of the herd may well be affected by just how we handle traffic decisions.

The traffic engineer can help with suggesting construction in the first case, and help with suggesting policy and regulation in the second. Some cases are a hybrid. And sometimes we are just screwed.

As I say, the traffic in the Washington D.C. area is the pits. A favorite story: I was near the downtown area and just could not figure how to get to where I wanted to go w/o getting caught up in very heavy traffic, so I asked a cop for suggestions. He thought a bit and then it went as follows:
You see that No Left turn sign at the corner?
Yep
Go down there and hang a left.

This worked fine.

Added: Everyone is entitled to a pet of his/her choice. I favor letting people in, but it's not in the Ten Commandments.

Thiros, on 2014-July-19, 00:56, said:

I live in a town with a large University. Some of the intersections around campus have an all-red cycle to facilitate pedestrian flow.

What frosts me is the guys who come up from behind at high speed in a side lane (usually one that's ending or has stopped or very slow traffic up ahead) and then expect the folks in the long line at the merge or end of the stopped traffic to let 'em in - and people do. Most folks would never consider jumping a line when on foot at a bank or store or whatever, but put 'em behind the wheel of a car and it's a different story.

kenberg, on 2014-July-19, 05:51, said:

The solution I gave above has nothing to do with construction. It only has to do with drivers' behavior. Changing the drivers' behavior (away from the blocking method you described as a solution and towards "zipping") considerably improves the flow of traffic and reduces traffic jams.

Rik

blackshoe, on 2014-July-19, 12:55, said:

Yes, and I see that as sort of the same thing I am talking about. The drivers trying to merge from a parking lot have other options than pulling directly into the lane in which they want to travel - they could turn right and then turn around - or they could turn right and circle the block.

Similarly, the driver in your example has the option of merging earlier into traffic or risk getting shut out totally.

blackshoe, on 2014-July-19, 12:55, said:

I had a coworker who said he had no qualms about waiting until the last minute to merge into the exit lane. I tried making this analogy to him, but it had no effect.

barmar, on 2014-July-20, 02:25, said:

I console myself with the fact that most people don't do this. Out here, there are a great many places and times when traffic gets pretty impossible. I deal with it by trying to plan my travel times, by taking the metro, and if worse comes to worse then just sitting back and listening to the cds. Or, sometimes, by not going. I once got off the road, called my wife, and said "I can't stand it, I'm having dinner here, I'll be home when things calm down." I try not to do that often. I think that the sheer awfulness of it all has, in some weird way, brought most of us to an acceptance and then to cooperation. There are still a few nuts, there always will be. On the capitol beltway, the speed limit is 55 for the most part. Much of the time traffic moves much slower, but at other times it is open and typical speeds are 70 to 75. So I happily move with the traffic. Then out of nowhere some guy comes streaking past me at maybe 15 mph, or more, faster than I and everyone else are going. Weaving through us all. Either he has a passenger who just had a heart attack and they are on their way to the hospital, or he is a total nut. Or both.

When someone rudely pushes his way in front of others, I have learned to accept it. I don't do it, most don't do it, maybe he can someday decide that he won't do it.

Interestingly, if most did it it would kind of solve the problem. The back-log would just move from the right lane to the send-to-right lane, where all the people waiting until the last minute to merge would get stuck behind each other.

Traffic jam is an interesting phenomena from a statistical point of view. Utterly unpredictable. Somewhat akin to a super-cooled liquid, maybe?.

helene_t, on 2014-July-21, 02:59, said:

Having studied this as part of my chaos theory course, it's surprising how little deviation from completely predictable behaviour it takes to cause one in simulations.

Dangling modifiers! (not actually a pet peeve but I figured it is in the spirit of this thread to point it out)

barmar, on 2014-July-21, 02:52, said:

There are interesting variations. In the merge I was describing, it goes like this: As we drive along 32, say in late afternoon, it is 100% predictable that we will need to slow down and merge. We all know what is coming. This has the following affect on me. I am usually in the left lane of the two lanes, and generally we are passing, not flying by but passing, the cars to our right. At some point, about a mile or a bit more before the merge, I start to think that passing more people would be tacky. The psychology changes from "I am just choosing to drive in the fast lane" to "I'm now barging ahead of others as we approach the merge". And so I move to the right lane. Interestingly, so do most others. And then, often, there will be a "guardian" who stays in the left lane but slows down to stay even with the car to his right, blocking those who wish to skip ahead. Mostly we have all come to accept lining up.


I think it is the predictability, the inevitability, of it all that makes it work. Other times and places, where the jam just appears out of nowhere, things become far more chaotic.

The big issue was/is to get the rwo lane, meaning one lane in each direction, road moving at a decent clip. The overpass did wonders for this. Without it, the situation was hopeless, no matter how we did our lining up. It's still a pain, but a much lesser one.

Here in Rochester, the interchange where Interstate 490 and Interstate 590 meet was, back before I moved here, a very dangerous PITA. They called it "The Can of Worms". They rebuilt the interchange, also before I moved here, and it's much better now, though still sometimes a PITA. And old timers still call it "The Can of Worms".

gwnn, on 2014-July-21, 04:08, said:

Tihs does not really qualify as a dangling participle but has some of the same amusing ambiguity: Advice on a gym wall "Do not exercise until the heart is pounding"

According to the book traffic it is actually better for traffic flow for people to late merge.

kenberg, on 2014-July-21, 19:27, said:

Yes, negations are often ambigious in English. In most other languages I know of it would be clear that it means "Do not (exercise until the heart is pounding)". If they meant "(Do not exercise) until the heart is pounding" they would have said "Do not (exercise before the heart is pounding)"