Right-turn arrow with simultaneous opposing left-turn arrow

[tradephoric]

Being that the unusual traffic light configuration has remained in place for many, many years, without almost anyone on these boards even being aware of this intersection until this thread came about, I would say that the crash rate is probably fairly average for this type intersection without any considerable or unusual congestion.

Objection:  Conjecture. 

Is "probably fairly average"  a technical term for "I have no idea what the hell I'm talking about" ?  Can we at least agree that the crash rate of this intersection would be meaningful to this discussion?  I'm not saying rip out this intersection and start over but I would at least like to know the facts.  I better not push for the crash data or i might be accused of "cherry-picking" again.

[NWI_Irish96]

I used this intersection twice a day during rush hour plus several other times for six years and only remember one time there being a problem due to a crash. 

[tradephoric]

I used this intersection twice a day during rush hour plus several other times for six years and only remember one time there being a problem due to a crash. 

Anecdotal. 

[Revive 755]

Since IDOT is currently studying the I-290 corridor and its interchanges, it is very likely they have already gathered crash data for the Harlem interchange; it could already be on the study website somewhere.

EDIT:  There is some crash data on Page 64 of 93 of this report for the Harlem interchange; Austin Avenue is on the next page.

EDIT 2:  There is also a discussion on Page 34 of 55 of this report.

[paulthemapguy]

^At least your anecdotal evidence is based on the actual intersection in question (unlike Baloo’s anecdotal evidence).  Can anyone actually prove that this intersection functions well and doesn’t have a high crash rate based on something more than their own personal experiences or second hand accounts?  Until then a compelling argument hasn't been made. 

Being that the unusual traffic light configuration has remained in place for many, many years, without almost anyone on these boards even being aware of this intersection until this thread came about, I would say that the crash rate is probably fairly average for this type intersection without any considerable or unusual congestion.

As what happens a little too often, we try to make things a problem that have never been a problem, simply based on a photo or post.

I knew about that weird intersection...But people seem to be figuring it out decently well.  Having just taken the P.E. exam, I recognized that the Highway Capacity Manual's calculations for peak vehicle flow involve a factor fp (the "population factor") considering driver familiarity.  If the population using the intersection is highly familiar with the area, i.e. the overwhelming majority of drivers at a location are locals, it has a different effect on the traffic there.  People using this interchange are almost invariably locals, because the Eisenhower is largely avoided by those in the broader region, due to its infamous congestion problems.  Unless you're trying to reach the city itself, in which case it might be your only viable roadway option (protip: take the train!!).

Main point:  Just because you would get confused by the intersection doesn't mean the locals don't know it like it's second nature.  The locals know what's up even if we don't.

Also-

I used this intersection twice a day during rush hour plus several other times for six years and only remember one time there being a problem due to a crash. 

Anecdotal.

No shit sherlock.

[kphoger]

Unless you're trying to reach the city itself, in which case it might be your only viable roadway option (protip: take the train!!).

Unfortunately, the most affordable option for taking transit into the city involves exiting at Harlem, parking for free at the Whole Foods just north of Lake (supposedly, you can't park there if you're not a customer, but so many stores share the parking lot that it's impossible to enforce), and taking the Green Line.

[tradephoric]

Having just taken the P.E. exam, I recognized that the Highway Capacity Manual's calculations for peak vehicle flow involve a factor fp (the "population factor") considering driver familiarity.  If the population using the intersection is highly familiar with the area, i.e. the overwhelming majority of drivers at a location are locals, it has a different effect on the traffic there.  People using this interchange are almost invariably locals, because the Eisenhower is largely avoided by those in the broader region, due to its infamous congestion problems.  Unless you're trying to reach the city itself, in which case it might be your only viable roadway option (protip: take the train!!).

Main point:  Just because you would get confused by the intersection doesn't mean the locals don't know it like it's second nature.  The locals know what's up even if we don't.

Your main point is wrong.  Harlem Avenue had the highest crash rate of any intersection analyzed in the I-290 crash analysis.  And I don't care what you learned in kindergarten class today Sherlock. 

The top five crossroads with the highest crash rates are reviewed in further detail.  Harlem Avenue, Cicero Avenue, Austin Boulevard, 25th Avenue, and 1st Avenue all have three year crash totals that exceed 100 crashes and were analyzed in greater detail to determine any noteworthy crash factors.  More detailed information regarding the below referenced crossroad traffic operations can be found in the I-290 Phase I Study Existing Roadway Operations Technical Memorandum.

Harlem Avenue — Rank 1
The Harlem Avenue crossing of I-290 consists of two 11'wide through lanes in each direction that interchange with I-290 via a full access, single point intersection controlled by one signal.  Of the 163 crashes on Harlem Avenue, the predominant crash types were rear-end (37%), turning (32%), and sideswipe (23%) with the majority of crashes (83%) having occurred during congested periods.  Harlem Avenue's Average Daily Traffic (ADT) is between 36,000 and 40,000 with peak hour operations functioning at an overall LOS of E. The left and right turn lane demand exceeds the available single turn lane storage capacities by over 200%.

The excess demand causes turn queues to spill back into the through lane, resulting in a blockage to through traffic.  This imposed blockage contributes to congestion and increases the probability of rear end crashes.

Furthermore, vehicles attempting to maneuver around a stopped turning vehicle also increase exposure to sideswipe crashes. I-290 on-ramp metering may also be contributing factor for rear-end and turning crashes. During peak hours, the stop requirement on the on-ramp may cause the ramp to back up into the intersection resulting in turning vehicles not being able to clear the intersection before a yellow or red signal.

The 11' through lane widths are narrower than the 12' lanes that are preferred where truck traffic exists. Narrower though lanes contribute to sideswipe crashes as there is less room between lanes for trucks to maneuver. Also, the sustained 4% grades of the approaches to the I-290 intersection are not desirable for operations. When releasing the clutch to accelerate, stopped trucks and vehicles on the grade may roll backward, increasing the chance of a collision, and steeper grades may also factor into the 7% of crashes having occurred in ice or snow conditions.

http://eisenhowerexpressway.com/pdfs/crash_analysis_report.pdf

[jeffandnicole]

Being that the unusual traffic light configuration has remained in place for many, many years, without almost anyone on these boards even being aware of this intersection until this thread came about, I would say that the crash rate is probably fairly average for this type intersection without any considerable or unusual congestion.

Objection:  Conjecture. 

Overruled.  This ain't a court.

[tradephoric]

Good thing.  You would make a probably fairly average lawyer. 

[paulthemapguy]

Unless you're trying to reach the city itself, in which case it might be your only viable roadway option (protip: take the train!!).

Unfortunately, the most affordable option for taking transit into the city involves exiting at Harlem, parking for free at the Whole Foods just north of Lake (supposedly, you can't park there if you're not a customer, but so many stores share the parking lot that it's impossible to enforce), and taking the Green Line.

The totally-legal-and-stuff way to do it is to do what I do all the time:  Exit at Des Plaines Ave, park at the park-n-ride lot, and get on the blue line at the line's west end.  Pretty painless, few congestion issues once you're off 290, and parking isn't insanely priced.

[tradephoric]

Paulthemapguy, you going to ignore the fact that Harlem Ave was the highest crash prone intersection in the I-290 study?  What ever happened to your theory that the locals get it?  Oh yeah, your theory was bunk based on no actual facts.

[paulthemapguy]

most facts are learned by going outside

you should drive through the intersection sometime

oh yeah, actually interacting with the physical thing you're talking about is merely "anecdotal" and therefore meaningless

We know when things are anecdotal.  You don't need to point it out.  The overwhelming majority of learning comes from interacting with the thing you're learning about, not reading about it from a distance.

Yeah the intersection sucks and it's dangerous.  That's real.  But ROW is nonexistent for reconfiguring any alignments, so that leaves us with reconfiguring the intersection.  Got any suggestions for something we can implement going forward, as opposed to reading reports on occurrences looking back?

[tradephoric]

It's hard to know where you stand paulthemapguy. 

I hate this interchange.  KILL IT WITH FIRE! 

I knew about that weird intersection...But people seem to be figuring it out decently well....  Just because you would get confused by the intersection doesn't mean the locals don't know it like it's second nature.  The locals know what's up even if we don't.

Yeah the intersection sucks and it's dangerous. 

So which one is it?  Is the intersection dangerous or do the locals know how to drive it and it's functioning decently well?   Make up your mind.

[paulthemapguy]

It's a bit more prone to crashes but it isn't the disaster you're making it out to be.

[tradephoric]

It's a bit more prone to crashes but it isn't the disaster you're making it out to be.

If you believe I think Harlem Avenue is a "disaster"  simply because it was the highest crash prone intersection in the I-290 study then sobeit.   But consider this.  Suppose there is a room full of people and Harlem is the oldest person in the room.  Someone may hear the phrase "oldest person"  and automatically assume Harlem must be really old.  It turns out it's a room full of kids and Harlem is only 12 years old.  It's not my fault someone wrongfully assumed Harlem is 80 just because they heard the phrase "oldest person" .   You always have to consider the facts.

[mrsman]

I hope the flashing yellow arrow talk isn't referring to any left arrows for the offramp traffic.  Every turning movement should see green for at least one phase.  The flow rates at these approaches is too great for FYA options; the queues need to be cleared as quickly and effectively as possible.  That's why the setup is what it is, as questionable as it is.  My vote is to remove the green right-arrows from the signal phase for the offramps and reserve them for the phase with Harlem's left-turning traffic.  If we want to try a FYA, I think it would be cool to try a right-turn FYA that is active with the offramp's steady green left arrow.  So:

New proposition:
Phase 1:  Left turns off of Harlem, protected. Steady-burn green right arrow for offramp approaches
Phase 2:  Harlem thru-traffic
Phase 3: Offramp approaches have steady-burn green left arrow, and flashing yellow right arrow

Sorry for chiming into the conversation late.  But I believe that the best approach would be for phase 3 to have steady green left arrows and red balls for the right turn.  This would make the intersection similar to a lead-lead with a doghouse.  At an intersection like that, you know that the left turners have the right of way and you can turn right when the left turners don't conflict with you.  Probably much easier to enforce than making sure people stick to their own lane.

Two intersections that aare somewaht similar with these issues:

Linden Dr / Wilshire Blvd in Beverly Hills, CA.  The signaling is similar to my solution above.  Traffic on Linden must turn onto Wilshire in both directions.  Linden always has a red ball and its "green phase" is green left arrow with red ball in a doghouse.  right turners yield to left turners and a diagonal pedestrian crosswalk.  Left turners have no conflict with the crosswalk.


Union Turnpike / Queens Blvd in Jamaica, NY.  Left turners from Q to UT have a left arrow when right tuners also have a right arrow.  There is some channelization to make this better, but there is still conflict.  To avoid conflict right turners must turn into the curbside lane, but then half a block ahead there are parked cars in that lane, so then you have to merge into traffic.  NYC has a no turn on red law and I beleive the right turn arrow signal was meant to work mainly with the pedestrian crossing without much thought to traffic on UT or the left turning traffic from Q to UT.  IMO, the green right arrow should be replaced with a FYA unless there is no vehicular conflicts.  (I believe the only part of the signal when there are no vehicluar conflicts [when Q thru traffic has right of way] has the pedestrian conflict.) 

See: https://www.google.com/maps/@40.7142755,-73.8304349,3a,75y,98.6h,61.34t/data=!3m6!1e1!3m4!1swPSKRtXLlIHYGwwNEFUpEA!2e0!7i13312!8i6656!6m1!1e1

[jakeroot]

Found a setup that is basically identical to the OP in Kingston, WA.

Two one-way streets meeting a major one-way street, but both one-way streets have green arrows at the same time for the turn onto the major one-way street. Only way this can safely work is if both cars turn into the near lane. Which they should, of course, but I still don't think this is permitted.

http://bit.ly/2U8CPnz (spin camera around to see the other signal)

[Roadsguy]

Found a setup that is basically identical to the OP in Kingston, WA.

Two one-way streets meeting a major one-way street, but both one-way streets have green arrows at the same time for the turn onto the major one-way street. Only way this can safely work is if both cars turn into the near lane. Which they should, of course, but I still don't think this is permitted.

http://bit.ly/2U8CPnz (spin camera around to see the other signal)

I wouldn't have much issue with this if they were ordinary green signals, but a green arrow implies a protected turn, with any opposing traffic only making a turn on red in which they must yield. That's what makes this dangerous.

[jakeroot]

Found a setup that is basically identical to the OP in Kingston, WA.

Two one-way streets meeting a major one-way street, but both one-way streets have green arrows at the same time for the turn onto the major one-way street. Only way this can safely work is if both cars turn into the near lane. Which they should, of course, but I still don't think this is permitted.

http://bit.ly/2U8CPnz (spin camera around to see the other signal)

I wouldn't have much issue with this if they were ordinary green signals, but a green arrow implies a protected turn, with any opposing traffic only making a turn on red in which they must yield. That's what makes this dangerous.

Absolutely. Regular green orbs would be totally fine and absolutely normal. Maybe even changing one of the directions to a flashing yellow arrow. Considering that both movements permit turns on red, plus the high likelihood of the movements being low-trafficked anyway, means that issues rarely ever occur, but I'd rather they swap out the green arrows anyway, just to prevent the opportunity.

[Amtrakprod]

Found a setup that is basically identical to the OP in Kingston, WA.

Two one-way streets meeting a major one-way street, but both one-way streets have green arrows at the same time for the turn onto the major one-way street. Only way this can safely work is if both cars turn into the near lane. Which they should, of course, but I still don't think this is permitted.

http://bit.ly/2U8CPnz (spin camera around to see the other signal)

I wouldn't have much issue with this if they were ordinary green signals, but a green arrow implies a protected turn, with any opposing traffic only making a turn on red in which they must yield. That's what makes this dangerous.

Absolutely. Regular green orbs would be totally fine and absolutely normal. Maybe even changing one of the directions to a flashing yellow arrow. Considering that both movements permit turns on red, plus the high likelihood of the movements being low-trafficked anyway, means that issues rarely ever occur, but I'd rather they swap out the green arrows anyway, just to prevent the opportunity.

They should give FYA to the least traveled road.

[UCFKnights]

Another simple solution would be to install the flexible plastic bollards between the 2 lanes to force vehicles into the correct lane, eliminating the conflict. i've seen that before (don't ask me where, it was a long time ago)

[jakeroot]

They should give FYA to the least traveled road.

That would be fine too.

Another simple solution would be to install the flexible plastic bollards between the 2 lanes to force vehicles into the correct lane, eliminating the conflict. i've seen that before (don't ask me where, it was a long time ago)

Given that there's no cross-traffic, that seems like a reasonable option too. I may even suggest it to WSDOT!

[jeffandnicole]

Another simple solution would be to install the flexible plastic bollards between the 2 lanes to force vehicles into the correct lane, eliminating the conflict. i've seen that before (don't ask me where, it was a long time ago)

Assuming there's even an issue right now (in other words, it's been working fine until this was posted, now we're in search of a solution where a problem hasn't existed), the plastic bollards could only come out as far as it wouldn't interfere with cross traffic.

Also, they would limit large vehicles from turning properly, so they may not be a good solution anyway.

[UCFKnights]

Another simple solution would be to install the flexible plastic bollards between the 2 lanes to force vehicles into the correct lane, eliminating the conflict. i've seen that before (don't ask me where, it was a long time ago)

Assuming there's even an issue right now (in other words, it's been working fine until this was posted, now we're in search of a solution where a problem hasn't existed), the plastic bollards could only come out as far as it wouldn't interfere with cross traffic.

Also, they would limit large vehicles from turning properly, so they may not be a good solution anyway.

Just because it hasn't caused an accident doesn't mean its not an issue, there is still conflicting movements with green signals. There is no cross traffic permitted, so the bollards would help prevent that illegal manuever from happening, and they are flexible and typically pop right back up if a truck needs to go over them to make a turn, and they're cheap

[jeffandnicole]

Another simple solution would be to install the flexible plastic bollards between the 2 lanes to force vehicles into the correct lane, eliminating the conflict. i've seen that before (don't ask me where, it was a long time ago)

Assuming there's even an issue right now (in other words, it's been working fine until this was posted, now we're in search of a solution where a problem hasn't existed), the plastic bollards could only come out as far as it wouldn't interfere with cross traffic.

Also, they would limit large vehicles from turning properly, so they may not be a good solution anyway.

Just because it hasn't caused an accident doesn't mean its not an issue, there is still conflicting movements with green signals. There is no cross traffic permitted, so the bollards would help prevent that illegal manuever from happening, and they are flexible and typically pop right back up if a truck needs to go over them to make a turn, and they're cheap

Just because this was found out yesterday doesn't mean there has been or will be issues.

One of the fallacies in planning is finding out about something that has existed for years, and feeling the need to make corrections to it, even when corrections aren't needed.  Looking at Street View, this particular layout has been in existence since 2013.  There's no reason to split the lanes, especially for those coming up the ramp.

Not only that, but the opposing arrows here are completely lawful, as nothing is stated within the law that the arrows provide a protected movement.  Per https://app.leg.wa.gov/rcw/default.aspx?cite=46.61.055 ,

Traffic control signal legend.

Whenever traffic is controlled by traffic control signals exhibiting different colored lights, or colored lighted arrows, successively one at a time or in combination, only the colors green, red and yellow shall be used, except for special pedestrian signals carrying a word or legend, and said lights shall indicate and apply to drivers of vehicles and pedestrians as follows:

(1) Green indication

(b) Vehicle operators facing a green arrow signal, shown alone or in combination with another indication, may enter the intersection control area only to make the movement indicated by such arrow, or such other movement as is permitted by other indications shown at the same time. Vehicle operators shall stop to allow other vehicles lawfully within the intersection control area to complete their movements. Vehicle operators shall also stop for pedestrians who are lawfully within the intersection control area as required by RCW 46.61.235(1).

So again, there's no issue with the issue as is, and introducing a hazard in the middle of the roadway, especially for the straight thru movement, is usually not advisable.