What is "perfect" signal timing?

[tradephoric]

LOL I was expecting that!  To improve intersection safety, retractable bollards should extend up from the roadway when a signal turns red.  That way if anyone blows through a red light, they will crash into the sturdy posts before they have a chance of t-boning anybody in the intersection.  No more need for roundabouts!  Here is a crash test to simulate how safe an intersection would become when someone tries to blow through a red light!

I'm hoping this wasn't a serious recommendation...

I do think retractable bollards to stop red light runners would be a terrible idea.  But I also think it's a terrible idea to have immovable fixed objects in the central islands of roundabouts (especially at suburban roundabouts where approaching speeds are high).   In the roundabout thread I pointed out that there have been 3 fatalities at the 96th and Westfield roundabout in Carmel.  In each instance the driver struck a concrete wall in the middle of the central island.  Just like the crash video of the truck striking those bollards, it can be a vicious crash when a driver blows through the middle of a roundabout and strikes a concrete wall.  Many people defended the fixed objects in the middle of roundabouts, arguing that it prevented drivers from blowing through the roundabout completely and potentially hitting a vehicle on the other side.

This roundabout features a rather massive concrete "planter" inside, which gets crashed into somewhat often by drunk/impaired drivers headed out of D.C.

I don't think it is poorly designed, and it improves Maryland traffic safety by "intercepting" some of those impaired motorists before they crash into something else, or a pedestrian or bicyclist.

If you drive like an idiot you live (or die) with the consequences.  These are the same people that would have blown the stop sign and taken out another car when it was a stop controlled intersection -- I'm much happier with them being stopped by an inanimate object rather than another vehicle.

You never know...there could've been someone traveling on the other side of the roundabout, and the boulder saved the idiot driver from plowing thru the median into their car, killing them instead.

If you think about it, fixed objects in the central islands of roundabouts only protects the circulating vehicles on the far side of the roundabout.  What about the circulating vehicles on the near side of the roundabout who are in between the concrete wall and the vehicle blowing through the roundabout?  They are still in danger of being t-boned.  With the retractable bollards, all vehicles in the intersection would be protected as the bollards would prevent the red light runner from ever entering the intersection.  Retractable bollards at signalized intersections would protect 100% of drivers already in the intersection, while fixed objects in the central islands of roundabouts only protects 50% of the drivers (the drivers circulating on the far side of the roundabout when somebody blows through it).  So signalized intersections with retractable bollards would be twice as safe as roundabouts with fixed objects in the central islands

[thenetwork]

After driving down the "free" frontage roads which parallel the Toll Roads in Dallas this past week (S. Rayburn, G. Bush,...), it seemed to me they were perfectly timed to make you stop at every light, in order to make you "consider" how much faster you could get to your destination had you paid $$$ to use the Tollway instead.

[mrsman]

Here's another question for those who are familiar with Detroit-area corridors like Woodward and Telegraph that feature great two-way progression:

If Woodward intersects a relatively minor street at a traffic signal, are the traffic signals tied in to a pre-arranged timing sequence in order to maintain progression (i.e. traffic on Woodward would always get a red at the pre-ordained time in the cycle, even if there is no traffic on the side street) or are the signals demand sensitive (resting on green favoring Woodward, only turning red if both at the preordained time in the cycle and if there is demand from the side street.)

I've wondered if demand-based triggering, which is generally a good thing, would have adverse effects on a carefully engineered traffic progression plan.

In NYC, the vast majority of signals are set by timer.  The one-way progression on the avenues is set into the timing, but they will face red lights even if there is no cross traffic at the wrong side of the cycle.

[thenetwork]

As far as the true N/S or E/W main roads in Metro Detroit (Telegraph, n-Mile Road,...)  All the lights are on a timed cycle, so the side streets -- and in the case of Telegraph Road with the lights at a U-Turn/Michigan Left intersection -- the lights will change whether or not someone is waiting for a green light off of Telegraph.

With the "spoke" roads in the Detroit area that are moreso angled (Grand River, Dix, Woodward, Gratiot,...) than true N/S or E/W "grid" roads, I'm sure they are timed for the main artery as best as possible.  It's when those streets intersect the main grid roads as to how the timings are set.

Most of my Detroit travels were usually south of 8-Mile and west of the Southfield Freeway, so it was pretty much all grid-based roads.  If you averaged 45 MPH on any grid road, you were pretty much good to go in any direction, as far as hitting green lights goes.

[Brandon]

Here's another question for those who are familiar with Detroit-area corridors like Woodward and Telegraph that feature great two-way progression:

If Woodward intersects a relatively minor street at a traffic signal, are the traffic signals tied in to a pre-arranged timing sequence in order to maintain progression (i.e. traffic on Woodward would always get a red at the pre-ordained time in the cycle, even if there is no traffic on the side street) or are the signals demand sensitive (resting on green favoring Woodward, only turning red if both at the preordained time in the cycle and if there is demand from the side street.)

They typically are timed for signal progression in order to maximize throughput on the main road (i.e. Woodward, Telegraph, etc).  After a certain hour, many of them may flash yellow for the main street and flash red for the side street (like between 10 pm and 5 am).

I've wondered if demand-based triggering, which is generally a good thing, would have adverse effects on a carefully engineered traffic progression plan.

If you ever want to see demand-based triggering all the time, 24/7, go to Chicago's suburbs.  Every signal there is triggered, and it makes it a right royal pain in the ass to keep going at the speed limit on a major thoroughfare like North Avenue or IL-59.

[kphoger]

If you ever want to see demand-based triggering all the time, 24/7, go to Chicago's suburbs.  Every signal there is triggered, and it makes it a right royal pain in the ass to keep going at the speed limit on a major thoroughfare like North Avenue or IL-59.

I drove north on Naper Blvd/Naperville Rd for five miles or more late at night one time, along with basically one other driver.  I eased away from the lights when they changed green, kept it under the speed limit.  He gunned it off the line and drove about 5 to 10 mph over the limit.  Our progress was exactly the same, because he kept getting stuck at red lights triggered by cross traffic, and I kept rolling up right around the time they were changing to green again.

I get frustrated sitting at red lights that I feel are unnecessary, so I learned not to try and rush in the Chicago suburbs late at night.  There's no point.  But I'd still rather sit at a red light triggered by a single car that ended up turning right anyway, than sit at a red light with no cross traffic at all.  I hate timed lights in general, even if I understand they might actually benefit my travel overall.

[Joe The Dragon]

Here's another question for those who are familiar with Detroit-area corridors like Woodward and Telegraph that feature great two-way progression:

If Woodward intersects a relatively minor street at a traffic signal, are the traffic signals tied in to a pre-arranged timing sequence in order to maintain progression (i.e. traffic on Woodward would always get a red at the pre-ordained time in the cycle, even if there is no traffic on the side street) or are the signals demand sensitive (resting on green favoring Woodward, only turning red if both at the preordained time in the cycle and if there is demand from the side street.)

I've wondered if demand-based triggering, which is generally a good thing, would have adverse effects on a carefully engineered traffic progression plan.

In NYC, the vast majority of signals are set by timer.  The one-way progression on the avenues is set into the timing, but they will face red lights even if there is no cross traffic at the wrong side of the cycle.

what does cop / fire traffic light preemption due to that?

train traffic light preemption?

[tradephoric]

Take a look down Big Beaver in Troy, Michigan.  The only traffic signals that stop both directions of travel along this stretch are at major mile roads.  The turnarounds along the wide-medians help eliminate the need for half-mile signals that ruin progression.



Compare that to a standard undivided roadway.  Now seemingly every church, school, and strip mall along the stretch of road warrants its own traffic signal.  You end up with a string of traffic signals between major mile roads that stop both directions of travel and kills signal progression.  Instead of coordinating 2 lights along a 2 mile stretch, you have to coordinate 10 lights.  When there are so many traffic signals spaced closely together, they will often be timed to turn green and switch back to red simultaneously.  That's a tell-tale sign the road isn't going to have good 2-way progression. 

[jeffandnicole]

In the first picture, I see traffic bunched up and tailgating.  In the 2nd picture, traffic appears to be progressing nicely.

Then again, in the 2nd shot, that's in a downtown area where you want to keep speeds slow anyway.  https://goo.gl/maps/huWty3jtSVE2  That said, I don't see any issues with that 2nd picture. The first picture appears to have some congestion issues.

[kphoger]

* kphoger cringes.

[tradephoric]

In the first picture, I see traffic bunched up and tailgating.  In the 2nd picture, traffic appears to be progressing nicely.

Then again, in the 2nd shot, that's in a downtown area where you want to keep speeds slow anyway.  https://goo.gl/maps/huWty3jtSVE2  That said, I don't see any issues with that 2nd picture. The first picture appears to have some congestion issues.

The 2nd picture probably isn't a great comparison (since it isn't a typical suburban corridor) but the theory still applies.  There was a video already posted driving a 5-mile section of Big Beaver in both directions without getting stuck at a red light (apart from when the driver turns around to demonstrate the dual progression).  I'll point out that at 3:55 the driver just makes it through a yellow light which is annoying because there is no legitimate reason for that crossover signal to be out of step with the surrounding signals.  Unfortunately that signal was malfunctioning (it had been out of step like that for days), which goes to show that even a corridor that is capable of great signal progression doesn't always achieve it. 

[tradephoric]

Traveling 40 miles down Woodward Avenue without hitting a red light.  During the nearly hour long drive, the driver makes it through 124 consecutive green lights.

https://www.youtube.com/watch?v=mb2R2fPB1nE&feature=youtu.be

[Hurricane Rex]

https://jalopnik.com/oregon-admits-violating-the-constitutional-rights-of-re-1821086560#js_discussion-region

Thought this might be useful to some showing how old our signal timing formula is.

If you want to rant about him getting fined unconstitutionally, I've started a thread over in Northwest about that with Oregon specific red light (speeding) camera usage.

[epzik8]

See Boulton Street between Route 24 and Tollgate Road in Bel Air, Maryland. When a light went up at Boulton and Gateway in 2012 very close to the one at the northern entrance of the Harford Mall/main entrance of the Harford Mall Annex (there's a side entrance on Gateway), the Town of Bel Air did a great job reconfiguring the lights at Route 24, the Mall/Annex and Tollgate to accomodate the light at Gateway and preserve the overall flow of traffic on that segment of Boulton, paying special attention to the Mall/Annex light.

Also relatively close by to me is Route 146 in the north Towson area between Pot Spring Road and the Interstate 695 Beltway interchange. That one varies between good and not so good. One of the determining factors is rush hour. It is not uncommon in the evening for traffic southbound on 146 to have a green light at Pot Spring only to have to stop within or prior to the Pot Spring intersection because they have a red light at the next intersection, Seminary Avenue, which is very close by. The following pair of lights, Charmuth Lane and Hampton Lane/ramp to the I-695 outer loop, has the same situation.

[UCFKnights]

So on minor sidestreets intersecting with a major throughway, what is preferred as the "perfect" signal timing? I've noticed some areas have the side streets stay green longer then necessary, while others all try to "perfect" the timing of the red lights, while the green lights are just ASAP based on the sensors on the side streets, I assume figuring that some intersections are going to be dumping different amount of cars on the road and to get through the queue takes different amounts of time anyways...