MLRC Forum - Print Page

MLRC Forum
https://www.mlrc.ca/cgi-bin/yabb/YaBB.pl
General >> Rally >> Traffic Light experiment
https://www.mlrc.ca/cgi-bin/yabb/YaBB.pl?num=1299336046

Message started by Robert Roaldi on 03/05/11 at 09:40:46

Title: Traffic Light experiment
Post by Robert Roaldi on 03/05/11 at 09:40:46

Want to see what happens when you shut down traffic lights?

http://www.wimp.com/trafficlights/

Title: Re: Traffic Light experiment
Post by nhibbert on 03/05/11 at 18:54:38

That's interesting. I'm going to send that link to a couple of traffic engineers I know.

Title: Re: Traffic Light experiment
Post by suprf1y on 03/05/11 at 23:49:29

I've been bitching about lights for 30 years, and in that time it's only gotten worse. Lot's of traffic lights makes for bad traffic, and bad drivers.

The place I go on vacation has few lights, NO stop signs, and traffic moves all the time. Have a look next time you're out. Would it be a tragedy to replace stop signs (and lights) with either nothing, or a yield sign?

Title: Re: Traffic Light experiment
Post by Napping on 03/07/11 at 13:49:09

excessive stop signs even worse. in 90% places yield sign would suffice.

Title: Re: Traffic Light experiment
Post by gully on 03/07/11 at 19:06:17

Peterborough is perhaps Canadas capital for all way stops

IF people understand them they are no problem its just the little old lady who cannot makeup her mind if its her turn

I have one intersection I use frequently nearly all ways two cars in each direction. Basically you have a roll up and go most of the time

There are some traffic lights which would be better served by Allways IMO

Pete g

Title: Re: Traffic Light experiment
Post by Da_Moose on 03/07/11 at 22:03:14

Even better would be round-a-bouts, never have to stop unless you're path is blocked.

Plus they're great fun when your bored!! :D

Title: Re: Traffic Light experiment
Post by Dave Cotie on 03/08/11 at 09:28:33

Da_Moose wrote on 03/07/11 at 22:03:14:

Even better would be round-a-bouts, never have to stop unless you're path is blocked.

Plus they're great fun when your bored!! :D

The City of North Bay constructed a round-a-bout 3 years ago.

I have to admit I was a vocal doubter before it was built. Now I am a fan. I would say that about 75% of the time I flow right through or just have a slight slowing as I go through. They take a lot of room but they sure do work.

A traffic engineer explained to me that signal lights have a 0.33 service ratio (only 1/3 of the time will you get through without stopping - on average). Round-a-bouts are 0.67 - double that of lights.

Title: Re: Traffic Light experiment
Post by Jeannie on 03/08/11 at 10:44:57

Napping wrote on 03/07/11 at 13:49:09:

excessive stop signs even worse. in 90% places yield sign would suffice.

Much of Ottawa-area traffic is unclear on the distinction between yield and merge. Maybe it's better where you live.

We have a roundabout in Almonte, and it works well. It can be fun if traffic is light -- gives your commute a happy moment.

Jeannie

Title: Re: Traffic Light experiment
Post by GuyB on 03/08/11 at 10:49:15

[quote author=66746D6D78010 link=1299336046/0#4 date=1299542777]Peterborough is perhaps Canadas capital for all way stops

I will add a few places to the competition: North Sydney, NS and Sorel, QC.

In Quebec, drivers attack 4 way stop sign intersections. I would suggest that it looks a lot more like what is in the video! Not bad actually!

Title: Re: Traffic Light experiment
Post by Robert Roaldi on 03/09/11 at 07:55:13

I wonder if anyone has statistics on accidents during power black-outs. That's almost like eliminating traffic lights, and might be a good proxy for conducting a more thorough experiment.

Title: Re: Traffic Light experiment
Post by Jeff_Hagan on 03/09/11 at 09:28:20

Heh... I heard an expression a while back that stuck with me: "everyone's a lawyer and everyone's a traffic engineer." ;)

Robert Roaldi wrote on 03/09/11 at 07:55:13:

I wonder if anyone has statistics on accidents during power black-outs. That's almost like eliminating traffic lights, and might be a good proxy for conducting a more thorough experiment.

I don't have any firm data, but I have some anecdotal evidence of collision rates sharply increasing during blackouts.

However, it's not a good comparison:

- even with signals, it looks like the street in the video was fairly low-speed and low-volume: the video talked about 2000 vehicles per hour (apparently for the intersection, though the video wasn't clear on what it was referring to); a busy urban intersection in Southern Ontario will experience many times that volume, and be carrying traffic at much higher speeds.

- I looked at the the area where the signals were removed in Google Maps: most of it is a jumble of narrow streets, sharp intersection angles, and weird multi-leg and offset intersections... all the normal hallmarks of a town core that was laid out before the age of the car. In general, this is a particularily difficult sort of area to signalize anyhow. North American-style wide streets on a grid arrangement were (usually) designed with signalization in mind and can normally handle being signalized much more efficiently.

- kinda building on that point, it's not like they council in the video would've just decided to get rid of signals without any sort of review or analysis to see what happened. It's likely this decision was the result of lots of traffic operations analysis, safety analysis, and probably a fair bit of simulation. It's also quite possible that if this review had found that this location wasn't a good candidate for removal of signals, they would've shelved the idea and that video would never have been made. There are a lot of tools in the traffic engineering toolbox, and no one tool is appropriate for all situations.

- a big factor affecting safety is driver familiarity and expectation. It's one thing to decide to remove signals from an intersection after a huge public notification campaign so the residents know what to expect. It's a whole different thing to just "flip the switch" and turn the lights off without warning.

Disclaimer: while I am a traffic engineer, this post is not meant as formal engineering advice or an official opinion. It should not be relied upon for any matter.

Title: Re: Traffic Light experiment
Post by Robert Roaldi on 03/09/11 at 10:42:56

Thanks Jeff. I happen to live near a semi-major intersection in an Ottawa suburb that was recently replaced by a roundabout. The two cross-roads were 4-lanes wide (2 each way) with traffic lights. Outside rush hours, there is not much traffic at this intersection, it is not near any big box malls, for instance. But the north-south road is a highway 174 interchange.

See here: http://www.orleansroundabout.ca/en/view.php?key=1&lang=en.

Check out the simulation on that web site.

See my comments about it near the end of this thread: http://www.mco.org/phpbb/viewtopic.php?f=17&t=12173

Title: Re: Traffic Light experiment
Post by Dave Cotie on 03/09/11 at 12:58:50

Jeff_Hagan wrote on 03/09/11 at 09:28:20:

Heh... I heard an expression a while back that stuck with me: "everyone's a lawyer and everyone's a traffic engineer." ;)

Robert Roaldi wrote on 03/09/11 at 07:55:13:

I wonder if anyone has statistics on accidents during power black-outs. That's almost like eliminating traffic lights, and might be a good proxy for conducting a more thorough experiment.

I don't have any firm data, but I have some anecdotal evidence of collision rates sharply increasing during blackouts.

.[/i]

The last time I was in Toronto, I witnessed a set of lights out on Hwy 7. - 70 km/hr limit. I did as required. I approached the intersection and treated it as a 4-way stop. I was almost rear-ended. I also witnessed people blowing through the intersection at 90-100 km/hr and not even lifting off the gas!