MUTCD gripes

[kphoger]

From the 2003 edition......

How the recommended lane marking changes from dotted to broken between mainline and decelaration lanes at a highway exit.  IMO, there is no good reason the broken line portion shouldn't be dotted or solid instead.  Both dotted and solid imply 'this is a special lane, get out unless you intend to exit'; broken, OTOH, implies 'you're cool, this lane keeps going'.


This seems to have been corrected in the 2009 edition.

[vdeane]

I just don't like the dots.  I understand why they're there, but since NY has never used them until recently, they just look weird.

The Thruway can't even be bothered to do them properly.  They replaced the lane stripings with dots, but they end them like a lane striping would end.

[vtk]

From the 2003 edition......

How the recommended lane marking changes from dotted to broken between mainline and decelaration lanes at a highway exit.  IMO, there is no good reason the broken line portion shouldn't be dotted or solid instead.  Both dotted and solid imply 'this is a special lane, get out unless you intend to exit'; broken, OTOH, implies 'you're cool, this lane keeps going'.

I think this setup is meant for use with a short deceleration lane, with the assumption that drivers can see the gore before the line changes from dotted to broken. Otherwise, that is indeed flawed.  Actually, without the dotted line, this is how ODOT has done simple exits for as long as I can remember – but the deceleration lanes are so short, that diagram would actually be to scale!

I was going to gripe about MUTCD not being spefic enough that gore crosshatches must be chevron-shaped, but then I reread chapter 3B and near the end it gets quite specific on that point.  ODOT seems to always do simple diagonal crosshatches in exit and entrance gores, as if the neutral area is a paved shoulder.  (In the last few years, however, ODOT has been doing chevron crosshatches more often.  Was this a recent improvement with the MUTCD, or did ODOT just recently get with the program?)

[kphoger]

Simple diagonals are common in Texas, and they like to paint the exit number between the diagonal lines.  Please don't take that away, it's SO cool!

[kphoger]

I was looking at the concept for the Kellogg/Webb interchange, and thought I had found an error in signing at the Greenwich/Zelta exit.  This would be a two-lane exit, one of which would be an option lane.  The final exit sign (at theoretical gore), however, has a yellow EXIT ONLY tab spanning both lanes, with two arrows (type E11-1e).  Well, thought I, both of those lanes are NOT exit only.

Much to my dismay, the signing complies with 2009 MUTCD standards.  Apparently 2009 MUTCD has decided that, for minor interchanges not warranting arrow-per-lane signing, it is best to lie about one of the lanes.  Grrrrr......

See below:

[PurdueBill]

Evidently their reasoning is that if the sign in question is at the exit itself, possibly even past where the option lane splits, that it's reasonable to make sure people are aware that there are indeed two lanes exiting.

Steve's I-77 page has a couple examples pictured (exits 118 and 120) of something similar under the old paradigm of advance option lane signage with the same two-lane EXIT ONLY treatment at the gore.  The difference here, which is good compared to the MUTCD cartoon, is the pull-through which clarifies how many lanes go straight through as well as exit.

The combination of overhead with only one lane shown for the exit and the lane assignment signs and markings showing the option lane is just too confusing in my opinion, but what do I know.  They must have data to back up the new way, right?

[Kacie Jane]

Evidently their reasoning is that if the sign in question is at the exit itself, possibly even past where the option lane splits, that it's reasonable to make sure people are aware that there are indeed two lanes exiting.

This.  If you look carefully at where that final sign is posted, it is in fact after the optional lane has split into two.  Ergo, it is technically correct.

[J N Winkler]

The combination of overhead with only one lane shown for the exit and the lane assignment signs and markings showing the option lane is just too confusing in my opinion, but what do I know.  They must have data to back up the new way, right?

It is the approach recommended by the Upchurch study of 2003.  Of several different layouts tried, it was found to be correctly understood the most frequently.

The old arrangement had a tendency to fool drivers into making unnecessary lane changes to the left to avoid an option lane that they incorrectly thought was a dropped lane.  In comparison, the new arrangement will, at worst, fool exiting drivers into making an unnecessary lane change to the right to get into the dropped lane when they can use the option lane.  The population of drivers likely to be fooled by the new approach is thus smaller (exiting traffic is usually less than traffic staying on the freeway) and the result is usually more benign (lane changes to the right do not usually involve a possibility of cutting off traffic overtaking at high speed).

The layout Kphoger was examining is here:

http://www.wichita.gov/NR/rdonlyres/163329A7-8B99-41BE-A45F-ACE98B312C58/0/20111206WebbConcept.pdf

The signing is pretty plain-vanilla MUTCD 2009, although there is a small glitch--"FREEWAY ENDS" signs on almost the same overhead sign bridge as "FREEWAY ENDS 1 1/2 MILE."

[Revive 755]

See below:

This arrangement is flawed as it assumes the ground mounted signs will not be obscured by traffic in the right lane.  It also assumes perfect placement of the final overhead sign near the gore.  For new construction, the sign structure could end up being shifted farther ahead of the gore due to a conflict with an unknown or poorly located utlity.  For replacement of an existing sign, it is assuming the sign crew gets the panel lined up far enough to the right and does not accidentally put it somewhat over the farthest right through lane.

As for the alternate overhead design shown on Page 195 of the 2009 MUTCD, it could get rather wasteful if the facility has more than three through lanes.  IMHO the white arrow for optional lanes was better, or a some other arrangement should be tried.

[vdeane]

The way you avoid people changing lanes is pull-through signs with arrows for all the through lanes.  But the idiots in the FHWA banned them.  As much as I like arrow-per-lane over diagramic signs, the pull-through and exit signs with down arrows are even better.

[SignBridge]

I agree with Deanej. I've said from the start that I think the FHWA addressed a problem that I just don't see. The conventional overhead sign showing one down arrow enclosed by the "exit only" wording with the second arrow not so designated was perfectly clear to me and I just don't understand how any driver could find this confusing.

On the other hand a recent episode at a civic association meeting was an eye opener re: the mentality of half the people out there. When you hear some of your neighbors seriously telling local police and traffic officials to put stop signs and speed bumps on thru county roads in the area to combat speeding, you get a realistic picture of the mentality of many people. A very sad commentary.

Deanej also has a point about the alignment of the overhead sign at the theoretical gore. At two locations in my area NYSDOT's contractor made errors in this regard. In one case, the sign for an approaching optional lane exit has the left arrow over the lane-line separating the dropped lane from the second lane, making it almost look like both arrows are over the dropped lane. In another case the 1/2 mile sign for another lane-drop has the single arrow positioned over the lane-line instead over centered over the exiting lane. Very sloppy.  

[roadfro]

Apparently 2009 MUTCD has decided that, for minor interchanges not warranting arrow-per-lane signing, it is best to lie about one of the lanes.  Grrrrr......

See below:

This is one of the major gripes I had with the revisions to signing option lanes at multilane exits. In my opinion, this new scheme would've worked just fine if the arrow over the option lane was white on green (like it used to be) and everything else new remained the same. No white-on-green down arrow on advance signage, no "lie" about the exit only status. This arrangement would've also allowed the exit direction sign to remain at the theoretical gore like it used to be instead of being moved downstream of the theoretical gore (keeping sight lines of the arrows above proper lanes from a distance).

[SignBridge]

Roadfro, I don't understand what you're saying. The E/D sign shown in the diagram is at the theoretical gore, not downstream of it, like you seem to be saying.

[roadfro]

^ Hmm... must've been interpreting incorrectly. To correct myself: older arrangements appear to have had the exit direction sign slightly more upstream from the theoretical gore, more towards the beginning of the ramp taper. Using everything else I mentioned, keeping this placement would've allowed the arrows to line up above the proper lanes from a distance on straight sections.

[J N Winkler]

Hmm... must've been interpreting incorrectly. To correct myself: older arrangements appear to have had the exit direction sign slightly more upstream from the theoretical gore, more towards the beginning of the ramp taper. Using everything else I mentioned, keeping this placement would've allowed the arrows to line up above the proper lanes from a distance on straight sections.

One of the findings of the Upchurch study was that under the old non-Lunenfeld & Alexander arrangement that was permitted under MUTCD 2003 (though not recommended by it), a significant proportion of motorists did not realize that they had the option of not exiting from the option lane.  This arrangement called for the exit direction sign to be placed at or upstream of the theoretical gore point.  The new arrangement attempts to fix the problem by not having multiple downward-pointing arrows assigning the same lane and by positioning the exit direction sign at or downstream of the theoretical gore point, so that the exit direction sign does not appear either to "lie" about lane count or to imply that the option lane is a dropped lane.

From a practical perspective, my concern about the change from the old to the new arrangement is the potential to place a burden on agencies either to move sign structures or to restripe (and, depending on the state, regrade and repave) gore areas to ensure that signs are placed correctly with regard to the theoretical gore point.  The only positioning that is acceptable under both the old and new arrangements is right at the theoretical gore point.

Agencies which followed the vanilla MUTCD recommendation, which was introduced at some point in the late 1970's--after Lunenfeld & Alexander's report came out--and stayed in the manual through the 2003 edition, are not affected because they can easily upgrade to arrow-per-lane diagrammatics.  These are not universally required for exits with option lanes but can be used at service interchanges which have option lanes, regardless of MUTCD interchange classification.

The problem is that these agencies are very few in number, because the old non-Lunenfeld & Alexander arrangement was very economical.  It exploited the fact that a one-lane drop with an option lane was not classified as a multilane exit, and so did not require a pull-through with downward-pointing arrows.  In comparison, the Lunenfeld & Alexander approach (as diagrammed in the MUTCD) did require such a pull-through sign in this context.  This means that agencies which already had a full-width sign bridge to accommodate the vanilla MUTCD arrangement can just reuse it for an arrow-per-lane diagrammatic, whereas agencies which used cantilevers with the traditional non-L&A approach may now have to do some gore-point shuffling.

[vtk]

It's simple: option lanes exist, so advance signage should indicate their existence.  Arrow-per-lane signs are not economical or aesthetically pleasing, so the older style signage should be allowed.  If drivers don't understand the signage style that has been around for decades, then driver's education should include this in a unit on reading guide signs.

Seriously though, how are people confused by this? Shouldn't the meaning of a white downward-pointing arrow on an exit sign be obvious after driving past a few of them, and seeing what the road actually does downstream?  This is how I learned to read guide signs, and as far as I can tell, anybody who misses this concept isn't paying attention.

[SignBridge]

JNW, I'm curious where you're getting your info on the placement rule for the E/D sign. I've checked the standard for the overhead exit direction sign in the 1988, 2003 and 2009 Manuals, and they all read the same: the Exit Direction sign should be installed overhead over the exiting lane in the vicinity  of the theoretical gore.  So the installing agency always had some leeway in this regard. Any slight deviation upstream or downstream apparently was always at the engineers' discretion and judgement. Most of the Manual diagrams show them right at the gore point, but the written rule is what actually applies.

And as I've said earlier, I am in complete agreement with VTK re: understanding conventional down-arrow signing. 

[J N Winkler]

JNW, I'm curious where you're getting your info on the placement rule for the E/D sign. I've checked the standard for the overhead exit direction sign in the 1988, 2003 and 2009 Manuals, and they all read the same: the Exit Direction sign should be installed overhead over the exiting lane in the vicinity of the theoretical gore.  So the installing agency always had some leeway in this regard. Any slight deviation upstream or downstream apparently was always at the engineers' discretion and judgement. Most of the Manual diagrams show them right at the gore point, but the written rule is what actually applies.

It comes from the Upchurch study.  Part of the solution to the option-lane problem it recommends is to position the exit direction sign downstream of the theoretical gore point, where it is evident that the option lane has already divided into two.

And as I've said earlier, I am in complete agreement with VTK re: understanding conventional down-arrow signing.

I liked the old non-L&A arrangement just fine, but:

*  Many agencies failed to implement it consistently, even within their own jurisdictions, which makes it very difficult for drivers to "learn by doing" as Vtk suggests.

*  It performed poorly in comparison with other alternatives in the Upchurch study--the option corresponding to the least frequency of driver error is that now diagrammed in the 2009 MUTCD.

Upchurch presented his findings to GMITC (the NCUTCD subcommittee which handles guide signs) around 2005 and apparently the experts present were sufficiently convinced that they ratified his suggested layout, which is now in the 2009 edition of the MUTCD.

Now, do I agree with it?  If I am pressed to give a categorical answer, I would have to say No, but the reality is that I am willing to acquiesce in a change of this standard because I do not believe any of the alternatives are unqualifiedly successful.  I have been following the technical literature on this issue for years and I have never seen an effective solution that did not require considerably more sign panel area than any of the only somewhat effective solutions currently in use.  I also used to collect traffic signing manuals before sign design sheets became my main preoccupation.  While it has some large gaps, the sun never sets on my collection, and I have yet to see a manual which lays out a method for dealing with the option-lane problem that is both economical and effective.  Collecting sign design sheets now gives me a chance to examine solutions that have been tried but not written into standards documents and I haven't seen any out-and-out winners there either.  I like the arrow-block diagrammatics used for system interchanges in TxDOT's Houston district, but those don't work well for service interchanges where text legend has to be provided.  I have seen efforts in Kansas and Alabama which are wasteful when they are not out-and-out weird.  I like MnDOT's artful use of vertical ruled lines (notably in the main "Unweave the Weave" contract), but that uses a lot of sign panel area.  And so on and so forth.

In the US we have traditionally preferred downward-pointing arrows because these provide positive guidance at no more than moderate cost, and even so we have fallen into a quagmire.  Classic L&A (as diagrammed in the MUTCD through 2003) is expensive.  Classic non-L&A is cheap, but confuses.  The Upchurch layout is somewhat cheaper than classic non-L&A, and may confuse somewhat less.  Arrow-per-lane diagrammatics would confuse even less still, at some cost in sign panel area, but the current MUTCD writes us into a high cost basis because we are not (as far as I can tell) allowed to use arrow-per-lane diagrammatics without arrows for the through lanes, which is what we need to replace the classic non-L&A approach.  (WisDOT has done this for some US 41 contracts, and IMO this practice should be legalized for the entire country.) 

Don't forget either that the preferred approach, whatever it is, has to be able to accommodate exits on curves.  In addition to Upchurch's presentation to GMITC, which I think is still available on the Web somewhere, Gene Hawkins' group at TTI did a study some years ago on the problems involved in signing option lanes at exits on curves.  This is well worth reading to get an idea of the motivations underlying current arrangements.

[SignBridge]

(Chuckle!) Okay, J.N..........Thanks for your comprehensive reply. Who would have thought the placement of an exit-direction sign would become such an interesting subject?

[roadfro]

Any chance any of these studies/reports (Upchurch, Lunenfeld & Alexander, etc.) are online or in an electronic form available from a library? I'd be very interested to read some of these and look at some of the alternatives.

[J N Winkler]

For the Upchurch study and the GMITC reaction to it, I suggest Googling {NCHRP Project 20-7 (155)} (no braces).  The relevant hits, including the 72-page final report, will be in the top ten.  (I prefer to paste links, but since Google now provides redirect URLs instead of direct links and I don't yet have a batch script to parse the redirect URLs, that is very awkward.)

Lunenfeld and Alexander's study (Signing treatments for interchange lane drops, June 1976) is unfortunately not online, to the best of my knowledge.  I found and photographed a copy in the government documents section at the Wichita State University library.  Similar facilities at other university libraries may have it; otherwise, I would expect most state DOT libraries to have a copy.

[kphoger]

From a practical perspective, my concern about the change from the old to the new arrangement is the potential to place a burden on agencies either to move sign structures or to restripe (and, depending on the state, regrade and repave) gore areas to ensure that signs are placed correctly with regard to the theoretical gore point.

Actually, I tend to think we should focus on striping more.  Signage can be imperfect as long as striping is accurate.  I am a big fan of the way Texas treated the I-35 South / I-410 South split on the north side of San Antonio.

Advance signage:  http://g.co/maps/m7yq8
Pavement arrows & shields:  http://g.co/maps/vrnvz
Further pavement markings:  http://g.co/maps/7h49z
Diagram (take 'em or leave 'em, I'm neutral):  http://g.co/maps/h29h6
And here's what is, to me, the most important part:  a solid white line between exit-only lane and option lane.  The final BGS uses a black-on-gold exit only arrow next to a white-on-green option arrow–as I would prefer.  But, with all the pavement marking leading up to the exit, and the combination of solid line and dual arrows, I think motorists would understand the configuration even if that last BGS weren't there.
http://g.co/maps/j99e9

This is in contrast to the way Wichita treated the I-135 / US-54 interchange when they redid the ramps.  Here, they used a confusing combination of broken and dotted lines; even where signage should be obvious and straightforward to anyone (including the exit shown in the satellite image below), I still see drivers move into the exit-only lane, then back into the other lane, often without signalling–and I see it probably every other day.  This is dangerous, as I am often on their left at the time, and get pushed onto the shoulder to avoid an accident.  However, it clearly shows that drivers think they must get far right for both movements.  I haven't seen drivers make this kind of mistake at the San Antonio example, though I've only driven it three times.
http://g.co/maps/mbyvx

[roadfro]

Actually, I tend to think we should focus on striping more.  Signage can be imperfect as long as striping is accurate.  I am a big fan of the way Texas treated the I-35 South / I-410 South split on the north side of San Antonio.

Advance signage:  http://g.co/maps/m7yq8
Pavement arrows & shields:  http://g.co/maps/vrnvz
Further pavement markings:  http://g.co/maps/7h49z
Diagram (take 'em or leave 'em, I'm neutral):  http://g.co/maps/h29h6
And here's what is, to me, the most important part:  a solid white line between exit-only lane and option lane.  The final BGS uses a black-on-gold exit only arrow next to a white-on-green option arrow–as I would prefer.  But, with all the pavement marking leading up to the exit, and the combination of solid line and dual arrows, I think motorists would understand the configuration even if that last BGS weren't there.
http://g.co/maps/j99e9

What you've shown here is a pretty decent application of the 2009 MUTCD pavement marking practice as it relates to intermediate exits with a dropped lane and an option lane. I didn't look too hard for the post-mounted regulatory signs, but the pavement marking arrows were there in plentiful order (possibly more so than recommended). The exit direction sign uses one exit only arrow and one standard arrow (instead of two exit only arrows), which departs from the MUTCD guideline but intuitively makes more sense to me.

Although I would note that this can be considered a system interchange. The diagrammatic arrow sign, as well as the final exit direction sign, would be an "arrow per lane" sign per current guidelines--for an intermediate interchange, the diagrammatic arrow sign could've had the single exit only down arrow and been in compliance.

[mjb2002]

The fact that counties are still using upper-case letters for the word "Town Limit"/"City Limit" on their signs, knowing full well that is not allowed anymore. I am in the process of creating a change.org petition that would address this issue.

Also, the fact that Advanced Street Name warning plaques must be yellow instead of fluorescent yellow-green.

The fact that the updated version of the manual does not have 6 inch lettering on Street Name signs as the new standard - that deadline passed on January 10, more than 4 months before revisions were made.

This used to be a gripe of mine, but not anymore since mixed-case is now mandated: the sign every street mandate that was implemented in 1988.

[Mergingtraffic]

The lack of drop down arrows on the BYS "This Lane Ends" or "Right Lane Ends" signs. 

If we're going to split hairs and add three dots on the "road narrows" signs, why can't we add down arrows for the BYS signs!?!?


Sometimes, the way the signs are mounted, it's hard to tell which lane they're actually talking about.  A drop down arrow, as seen in the link below, makes it that much easier to understand traveling at highway speeds.

http://www.google.com/maps?q=old+lyme,+ct&hl=en&ll=41.372123,-72.199101&spn=0.012254,0.01929&sll=41.312291,-73.088537&sspn=0.02466,0.038581&hnear=Old+Lyme,+New+London,+Connecticut&t=m&z=16&layer=c&cbll=41.372194,-72.199007&panoid=w3JFzbFQuTE37BvjTFZs2g&cbp=12,228.85,,0,0.7