Hacker News Reader

Comment by WalterBright | Hacker News Reader

I won't believe cities are serious about reducing congestion until they fix the traffic lights to dynamically maximize flow.

No, it isn't done already. Too often, I am in a platoon of 20 cars that has to slam to a stop so one car can cross the intersection on the cross street, and then there's yet another minute of red with nobody in the intersection.

You'd think with cameras already installed on the lights that this would be a far, far simpler job than self-driving cars.

View on HN

In Paris the idea is to make the city more hostile to cars (any car, your private car, taxis, ubers) by reducing the number of roads they are allowed onto.

The city voted yesterday to close the main road through the city ("quais rive droite") after the equivalent road in the other way ("quais rive gauche") had been closed 3 years ago.

I think London is thinking of doing the same (plans to close Oxford Street?)

How it will end is a mystery to me; not everyone is comfortable using public transportation (esp. the elderly). In London the subway is kind of okay... in Paris it's horrible.

I was in Rome recently and they seem to do something similar. It makes for a lot more pleasant city when you have so much less traffic and the associated noise.

Yes, London is planning to close Oxford street, it just has to figure out how buses can be redirected in a smart way to allow shoppers to get there easily. Furthermore, one of the busiest intersections (bank) will be closed to all vehicles but buses from next year on. The fact that it will even be closed to black cabs is a strong signal.

In addition to that there are also plans to revert some of the 70's car-friendly areas in the city. Mainly roads that were designed to be one-way to ease traffic flow will be made less car-friendly. This is supposed to slow down traffic which should make it safer for cyclists and pedestrians. ("The one-way system encourages high traffic speeds, which is at odds with the City's aspiration for safer streets within the Square Mile.") [0]

I'm not aware that there is any project in London at the moment that will ease congestion, it's all about making life for pedestrians and cyclists easier. As a cyclist, I really welcome that trend. The new cycle-superhighways [1] are a great thing, IMO.

[0] Project 48: https://tfl.gov.uk/travel-information/improvements-and-proje...

[1] https://tfl.gov.uk/modes/cycling/routes-and-maps/cycle-super...

Congestion in an inner city can reach a point where fiddling with the lights won't accomplish anything. But in moderate traffic areas, big improvements should be possible and practical.

After all, once an adaptive system is created, it could be installed on most intersections.

I keep thinking of adaptive car fuel injection systems that Ford uses. It's so adaptable that they're pulled off of 90's Mustangs and installed on Mopar V8's from the early 70's, despite being a completely different engine with different firing order, etc., and people have reported great results.

I never understood the interior layout of subways in Paris... they have seats positioned right at the door, where people would stand normally. It seemed to me like everyone was always stepping on each others toes, and on each stop the whole train was shuffeling around for people to get in and out.

That's because those seats are supposed to not be used when there's too many people.

This will be hard to believe but it's true: for years and years, the Paris subway authority considered that people would not accept to "be transported laterally", which means seats HAD to be either facing the front or the back of the train, NOT the side.

It still goes on today; some of the newer trains now have a few seats facing the side, but that's as far as they'll go.

This thinking, based on absolutely zero evidence but deeply ingrained, wastes an incredible amount of space aboard subway trains.

> In London the subway is kind of okay... in Paris it's horrible.

Having commuted in both London and in Paris, I would say it is quite hard to make this kind of statement.

When comparing my commute on the Northern Line to my old one on Line 2 and 1 in Paris, I can definitely say it was better in Paris.

And I'm sure that someone who could compare commuting on Line 13 versus, say, the Victoria Line, would have a very different opinion.

There are also more static variants. I know that Yandex collaborated with Zelenograd [0] (city to the north of Moscow) to optimize timings of city traffic lights. Yandex has similar data to what Google does in other places, due to Yandex.Maps app, and they reported to increase average car speed at many intersections without dropping average at any other one.

On the example shown, average car speed at one particular intersection improved from 10 km/h in rush hours to 30 km/h, which is fairly significant.

For many cities which do not want to spend money to replace all traffic lights, such collaborations could be less-than-optimal, but still a great improvement.

[0]: https://yandex.ru/blog/company/38835 (Russian)

Using statistics is one way, but is suboptimal. An optimal system would use the cameras to adjust to each situation like a traffic cop would. Though I bet even a modest adaptive light system could do better than a cop, for the simple reason that lights could coordinate with their neighbors.

I think your original comment was spot-on, and far more interesting than the hackneyed discussions taking place in this thread. For a little context:

"In 1922 traffic towers were beginning to be controlled by automatic timers. The first company to add timers in traffic lights was Crouse Hinds. They built railroad signals and were the first company to place timers in traffic lights in Houston, which was their home city. The main advantage for the use of the timer was that it saved cities money by replacing traffic officers. The city of New York was able to reassign all but 500 of its 6,000 officers working on the traffic squad; this saved the city $12,500,000."

Crouse-Hinds was still getting sued in the '90s for liability in accidents at intersections. I don't know if these lawsuits are still happening, but I assume liability has something to do with why your suggestion of a more adaptive light system is not tried.

London does it. http://abstracts.aetransport.org/paper/index/id/2041/confid/... https://www.youtube.com/watch?v=w4oeMB0tYII

The video is a bit cheesy, but it gets the point across.

> I won't believe cities are serious about reducing congestion until they fix the traffic lights to dynamically maximize flow.

> No, it isn't done already.

Its obviously not done most places yet, but adaptive traffic control [0] is a thing.

[0] https://en.wikipedia.org/wiki/Adaptive_traffic_control

That link leads to this:

https://en.wikipedia.org/wiki/InSync_adaptive_traffic_contro...

which is exactly the kind of system I've thought of. I didn't know it existed. Thanks for the pointer! This is great news.

Most of these points become moot with driverless cars. They can be small two-person smart cars that can drive bumper-to-bumper and wheel-to-wheel. First they'll get their own roads and then inner cities will ban manual-driven cars.

Unfortunately, I wouldn't hold my breath. Even Bill Gates couldn't fix this problem (literally -- who remembers TrafOData?)

Apparently, cities don't want smooth traffic flow. It's bad for their public transit agenda.

Reply on news.ycombinator.com