LADOT Night time flash.. but not how you would expect.

[deathtopumpkins]

Do they function like what is described above?  Are the pedestrian signals active and will they temporarily kick the signal into normal operation for the pedestrian crossing?

Yes. Ped signal heads will display DW while vehicle signal heads display flashing Y. When a ped pushes the button, the vehicle signals change to steady Y, then R, and peds get W then FDW, then back to DW and flashing Y.

Here's an example installation: https://goo.gl/maps/9r3obsjyuqEME8gk6 (you'll have to move around to actually see all of the various signal heads)

This is only for older installations though. New ones use steady green in place of flashing yellow. So it shouldn't really be considered here.

[Roadrunner75]

Here's an example installation: https://goo.gl/maps/9r3obsjyuqEME8gk6 (you'll have to move around to actually see all of the various signal heads)

Wow - that's not only a very old signal, but also very expansive and confusing.  It looks like it has flashing reds to 3 separate side streets, as well as acting as a fire signal.  Do Fellsway West and Walnut get both flashing red normally and solid red during the pedestrian phase?  I see they have "No Turn on Red" signs.  Wheatland also looks like it might operate that way too.

[mrsman]

But very few expect a flashing yellow to suddenly change to red.

Agree - And even worse at HAWK signals where it's dark and suddenly comes to life.  Another problem with HAWKs - dark signals are supposed to be treated as a "Stop" as if the signal was out, but we created a signal that conflicts with that rule.

I have witnessed the change at both 1 am and 5:30 am at the signal closest to my home.  At 1 am the signal on the main street will go from green to yellow to red (briefly) and then flash yellow.  The side street is all red and then begins to flash when the main signal flashes.  At 5:30 am, the signal just changes from flashing yellow directly to green and the flashing red becomes a solid red.

Does the side street turn to solid red for a few seconds before the main road turns to green?  If someone on the side street approaches right when that changes, they see the flashing red and treat it as a stop and then might pull out right at the change into what has become a green for the main road.  Obviously they had to look for a break in traffic anyway though...

I believe it does.  I know that the few times when I'm up that early, I try to get through that corner by 5:27 so that I can get through with just normal yielding.  (From the perspective of the side street of course)  It is a very long light during normal times and I would have to wait for a while if it were steady red.

[mrsman]

Do they function like what is described above?  Are the pedestrian signals active and will they temporarily kick the signal into normal operation for the pedestrian crossing?

Yes. Ped signal heads will display DW while vehicle signal heads display flashing Y. When a ped pushes the button, the vehicle signals change to steady Y, then R, and peds get W then FDW, then back to DW and flashing Y.

Here's an example installation: https://goo.gl/maps/9r3obsjyuqEME8gk6 (you'll have to move around to actually see all of the various signal heads)

This is only for older installations though. New ones use steady green in place of flashing yellow. So it shouldn't really be considered here.

In my neck of the woods in Maryland, we have something similar at an intersection.  It is reminiscent of fire station signals, but there is no fire station at this location.

Regular operation: Flashing yellow main/ flashing red side and steady DW.  There is a stop sign on the side street as well.  The signal is 12-12-8.

If a pedestrian pushes the button, the flashing yellow will then start to flash a bit faster (and some additional warning lights about 500 ft ahead of the intersection also come on).  At some point, you then get steady yellow and steady red.  At the moment of steady red on the main street, the side street has a double steady red as well.  At that same time the walk signal comes on.

Both the main street and the side street have a steady red for the entire duration of the pedestrian phase: W, FDW, and even a few seconds of steady DW.  Then, the signal reverts to the regular operation: Flashing yellow main/ flashing red side and steady DW.

Here is the location:

https://www.google.com/maps/@39.039868,-77.0428572,3a,75y,314.44h,79.25t/data=!3m7!1e1!3m5!1shr9MGP5YGfLHDqcsN1knAw!2e0!6s%2F%2Fgeo1.ggpht.com%2Fcbk%3Fpanoid%3Dhr9MGP5YGfLHDqcsN1knAw%26output%3Dthumbnail%26cb_client%3Dmaps_sv.tactile.gps%26thumb%3D2%26w%3D203%26h%3D100%26yaw%3D101.49685%26pitch%3D0%26thumbfov%3D100!7i16384!8i8192


Not a huge fan of this.  I've crossed here several times and seen cars go on through, but it has become better with the additional warning lights.  Also, it is huge waste to leave on the steady red for the entire pedestrian phase (it would be better if it went from steady red to flashing red during FDW to take advantage of the congestion benefit inherent in HAWK signals.)  Plus, it is frustrating for side street traffic to get a steady red during the one time when main street gets their red.  It is really hard to find a break in traffic here to cross through or make a left turn.

I would prefer a standard signal setup, by far.  Provide cross street green and provide a normal expectation for main street drivers.

[roadfro]

There's a flaw in the first premise. If you watch the video, you can see that the "walk" signal is on while the adjacent flashing yellow is on, which implies the side street seeing red would also be seeing the upraised hand for pedestrians.

No, I assumed that the walk signals were working.  A pedestrian walking up would see the "Don't Walk" signal but also possibly the flashing yellow signal on the main street.  Since they would assume it's on a regular night-time flash, they might assume the walk signal is not functioning (i.e. will not change to "walk") and just treat it as such and go ahead and cross the street.  The average pedestrian is not going to know that the pedestrian signal is supposed to be dark during the regular flash mode.  If they walked up and saw the signal functioning normally, they would probably be more likely to assume they can push the button (if equipped) and wait for the walk signal.

I agree with your interpretation. Most pedestrians, particularly those who don't drive, are not going to be intimately familiar with the operations of intersections in flashing modes (hell, most drivers aren't). The number of drivers that I've seen interpret a solid red arrow lasting slightly too long as meaning "clearly broken", and then turning through a gap, is an indication that people are impatient as shit.

I would agree that most pedestrians and drivers don't know the intricacies of how signals are supposed to operate, so the situation Roadrunner describes could be plausible. But I think a pedestrian is equally likely (or perhaps more likely) to think "the main road is flashing, but the don't walk is still lit, so maybe I should press the button".

I am very skeptical and not a fan of the arrangement.  The pedestrians should wait until it is safe to cross and wait for their WALK signal.

But they don't, so why do we insist that we design an operation that requires them to do so?

LA could have gone with the standard red/yellow flash pattern, which doesn't require the pedestrian to press a button to cross and wait for a walk signal...

[roadfro]

Obviously, this is being implemented to improve night-time pedestrian collisions. The flashing yellow orbs are meant to draw the attention of drivers more than a standard green orb, especially since the former ostensibly requires you to stop and give way to pedestrians, whereas the latter does not (short of intentionally hitting a pedestrian). These intersections would still have active crosswalk signals, but the flashing yellow is probably trying to reduce instances of drivers hitting pedestrians who are crossing against solid "don't walk" signals. Yes, the pedestrian is in "the wrong", but they're still (potentially) dead, and not helping LADOT's Vision Zero goal.

These are not MUTCD-compliant, but I've always thought that was kind of a dumb metric since a lot of road improvements over the last fifty years likely weren't MUTCD-compliant either. Things are "not compliant" until they are proven to be worthy of compliance. Clearly LA has the money to study this phasing; if it's successful, I'm sure they could petition the FHWA to modify the relevant MUTCD section.

Your rationale makes sense. But it would be more preferable to use some kind of operation that is compliant unless there's a compelling reason not to, which I don't see here. The greater problem is getting pedestrians to actually press a button and cross legally with the light.

I don't know if getting pedestrians to "press a button" should be the goal here. These Vision Zero programs generally put the burden on drivers to not hit pedestrians, rather than keeping pedestrians from getting hit by cars. This is largely because, unlike driving, you don't need a licence to walk. The onus is on drivers to operate their vehicle responsibly, and never hit a pedestrian if at all possible. Pedestrians are stupid, so forcing them to jump through hoops just to cross a street is not exactly a wise path forward. You might say pressing a button isn't much of a hoop; maybe, maybe not. But the operation of intersections, which clearly favours cars, does tend to force pedestrians to get impatient and just do their own thing. Especially at night. Especially in cities. Especially when drunk.

Does it make sense for there to be a goal of "getting pedestrians to actually press a button"? Sure, that way everyone is following their signals and no one is put in an actionable position (assuming there are no incidents). But in reality, intersections are not friendly for pedestrians outside of timed corridors; most actuated intersections have tiny windows of time to hit those buttons to get the walk sign to come on. Late at night, you come up to cross Broadway, but you arrived two seconds after the walk sign failed to activate. Are you going to wait an entire cycle to cross? No, you just go, cautiously. Are there pedestrians who do this not cautiously at all? Definitely, but there are just as many drivers screaming straight through those intersections under the correct premise that they have the right of way to proceed straight-on (again, short of running someone down intentionally). By replacing the green with a flashing yellow at night, we provide a gentle reminder to drivers that "hey, a lot of people jaywalk around this area at night; don't expect them to cross on a walk sign; drive with caution".

At the end of the day, if this pilot program has received approval from the various regional departments, it's not like California doesn't do their own thing all the time anyway. I'm more than interested to see how this works out.

If the overall goal is "vision zero" and zero fatalities, then the burden should be on everyone to take steps to avoid fatalities. I absolutely agree that there is a lot of onus that (rightfully) needs to be put on drivers to slow down and follow traffic laws, especially those regarding pedestrians (and other non-motorized road users). But pedestrians also have to take some responsibility in not doing things that would put themselves in a position that unnecessarily increases their exposure risk.

So yeah, if there's a button that activates a crossing warning, they should use it--give themselves the best chance to be seen by drivers and to have drivers recognize their presence. And engineers and planners should also be thinking about design strategies that promote safety and better accommodation for all road users--including decreasing delay for pedestrians, among other things. I'm just not sure that LA's operation is the best given other operational strategies available (i.e. standard red/yellow flash).

[deathtopumpkins]

Wow - that's not only a very old signal, but also very expansive and confusing.  It looks like it has flashing reds to 3 separate side streets, as well as acting as a fire signal.  Do Fellsway West and Walnut get both flashing red normally and solid red during the pedestrian phase?  I see they have "No Turn on Red" signs.  Wheatland also looks like it might operate that way too.

Yup, all three side streets rest on flashing red, and turn solid red during the ped or fire phases. Compliance (with NTOR or otherwise) is.....poor. Most people seem to just treat the side streets as a stop sign regardless of what the signal displays.

[jakeroot]

I am very skeptical and not a fan of the arrangement.  The pedestrians should wait until it is safe to cross and wait for their WALK signal.

But they don't, so why do we insist that we design an operation that requires them to do so?

LA could have gone with the standard red/yellow flash pattern, which doesn't require the pedestrian to press a button to cross and wait for a walk signal...

Actually, I would have been fine with this. But I'm sure they were worried about pedestrians trying to cross that wide of a road without protection if they so sought it.

There is another option that I'm not sure anyone has considered yet: Keep the intersections in flashing mode. When a pedestrian arrives, they are free to cross when safe, or they can hit the button. Hitting the button activates that signal into a solid green indication for a few seconds, before going yellow and red. The other street would see a walk sign, but could remain flashing red (as is the norm at many half-signals with stop signs). The other road would see a red light, and the ped signal could either display a solid "don't walk" or remain blank.

Half signals are kind of an iffy topic, but they're not new concepts, so I don't see why this isn't at least possibly viable.

Obviously, this is being implemented to improve night-time pedestrian collisions. The flashing yellow orbs are meant to draw the attention of drivers more than a standard green orb, especially since the former ostensibly requires you to stop and give way to pedestrians, whereas the latter does not (short of intentionally hitting a pedestrian). These intersections would still have active crosswalk signals, but the flashing yellow is probably trying to reduce instances of drivers hitting pedestrians who are crossing against solid "don't walk" signals. Yes, the pedestrian is in "the wrong", but they're still (potentially) dead, and not helping LADOT's Vision Zero goal.

These are not MUTCD-compliant, but I've always thought that was kind of a dumb metric since a lot of road improvements over the last fifty years likely weren't MUTCD-compliant either. Things are "not compliant" until they are proven to be worthy of compliance. Clearly LA has the money to study this phasing; if it's successful, I'm sure they could petition the FHWA to modify the relevant MUTCD section.

Your rationale makes sense. But it would be more preferable to use some kind of operation that is compliant unless there's a compelling reason not to, which I don't see here. The greater problem is getting pedestrians to actually press a button and cross legally with the light.

I don't know if getting pedestrians to "press a button" should be the goal here. These Vision Zero programs generally put the burden on drivers to not hit pedestrians, rather than keeping pedestrians from getting hit by cars. This is largely because, unlike driving, you don't need a licence to walk. The onus is on drivers to operate their vehicle responsibly, and never hit a pedestrian if at all possible. Pedestrians are stupid, so forcing them to jump through hoops just to cross a street is not exactly a wise path forward. You might say pressing a button isn't much of a hoop; maybe, maybe not. But the operation of intersections, which clearly favours cars, does tend to force pedestrians to get impatient and just do their own thing. Especially at night. Especially in cities. Especially when drunk.

Does it make sense for there to be a goal of "getting pedestrians to actually press a button"? Sure, that way everyone is following their signals and no one is put in an actionable position (assuming there are no incidents). But in reality, intersections are not friendly for pedestrians outside of timed corridors; most actuated intersections have tiny windows of time to hit those buttons to get the walk sign to come on. Late at night, you come up to cross Broadway, but you arrived two seconds after the walk sign failed to activate. Are you going to wait an entire cycle to cross? No, you just go, cautiously. Are there pedestrians who do this not cautiously at all? Definitely, but there are just as many drivers screaming straight through those intersections under the correct premise that they have the right of way to proceed straight-on (again, short of running someone down intentionally). By replacing the green with a flashing yellow at night, we provide a gentle reminder to drivers that "hey, a lot of people jaywalk around this area at night; don't expect them to cross on a walk sign; drive with caution".

At the end of the day, if this pilot program has received approval from the various regional departments, it's not like California doesn't do their own thing all the time anyway. I'm more than interested to see how this works out.

If the overall goal is "vision zero" and zero fatalities, then the burden should be on everyone to take steps to avoid fatalities. I absolutely agree that there is a lot of onus that (rightfully) needs to be put on drivers to slow down and follow traffic laws, especially those regarding pedestrians (and other non-motorized road users). But pedestrians also have to take some responsibility in not doing things that would put themselves in a position that unnecessarily increases their exposure risk.

So yeah, if there's a button that activates a crossing warning, they should use it--give themselves the best chance to be seen by drivers and to have drivers recognize their presence. And engineers and planners should also be thinking about design strategies that promote safety and better accommodation for all road users--including decreasing delay for pedestrians, among other things. I'm just not sure that LA's operation is the best given other operational strategies available (i.e. standard red/yellow flash).

Pedestrians absolutely have to take responsibility; if they don't, certainly it puts them in a very risk-averse situation. But that's kind of my point: those who aren't more personally responsible put themselves at a very high risk of serious injury or death, primarily due to their exposure to vehicles. After all, if there were no vehicles, there wouldn't be any pedestrian deaths (apart from being run down by an errant horse or a trolley). Now, obviously, there are vehicles, so that's a moot point. But the responsibility tends to fall onto drivers because, ultimately, they are the ones who determine a pedestrian's right to continue living. At any point, I could spontaneously drive onto a sidewalk and kill multiple people, all on a whim. We saw as much in Virginia in 2017 when that Challenger drove into a crowd, killing one person and injuring a couple dozen others. On the flip side, as a pedestrian, there's very little that I can do to harm a driver, so the number of responsibilities is likewise much lower (basically, just don't surprise drivers, and even that is primarily to prevent pedestrians from killing themselves).

With this in mind, I would agree that "engineers and planners should also be thinking about design strategies that promote safety and better accommodation for all road users". That's very well put, and I think Vision Zero, despite the negative press from road geeks, has resulted in a higher level of innovation than without Vision Zero. Things like the phasing in LA along Broadway (the topic of this video), or the HAWK signal, or leading pedestrian intervals, or segregated cycle paths, etc. We all have our opinions on each of these, but the fact that we're even thinking about them says a lot. Los Angeles might not have figured out the perfect phasing for Broadway, but they've at least attempted to shoe-horn some different strategies into a working concept that may or may not prove successful. But it beats doing nothing, especially if the "do nothing alternative" may result in continued pedestrian-vehicle conflicts along Broadway.

As much as I hate to admit it, the best thing for the future of roadway safety will be the advancement of self-driving technologies. After all, that 80 year old driver who has "experienced it all" and is, arguably, extremely good at driving, will eventually go blind or deaf, or become too feeble to operate a vehicle. That same day, a brand new 15-year-old will hit the road with none of those skills. We have to constantly teach and reteach these responsibilities to drivers and pedestrians. Largely, we've been successful, but we eventually have to accept that drivers will only ever be "pretty good" at driving. We will never have a population full of Michael Schumacher's. Even with exceptional driver training, the world's safest countries still have road deaths. Therefore, the best we can really do is "try this" or "try that", until the point where we don't even need a signal because everything is self-driving.