3DI numbering scheme—how it is, and how it should be

[briantroutman]

We all know the numbering rules for Interstates, including at least a vague notion of the rules for 3DIs–first digit odd: spur, first digit even: loop/beltway.

And if you look at the examples in the MUTCD, they're all very simplified. A spur connects with its parent at one point, extends outward in one direction, and terminates at some destination. Loops always reconnect to their parent. "Link" 3DIs that connect two Interstates but do not return to their parent–like I-280 in NJ, for example–aren't specifically listed in the examples, but the consensus seems to be that a 3DI ending at an Interstate on both ends qualifies for an even first digit. Not that far away (in PA), though, we have I-380, which also connects Interstates at each end. Obviously the real world is infinitely more complicated than the tidy examples, and no matter how you want to interpret the rules, they're being broken in many places.

So–multi-part question–what is your interpretation of the existing rule (for odd/even numbering), and in either a pragmatic or perfect world, what do you think it should be?

It seems to me that the only really useful information that odd/even can communicate is whether or not the 3DI returns to the parent. If I'm driving south on I-95 to Florida, and in the middle of Virginia I see an exit for I-295, I can take it with the security of knowing I'll end up back on I-95 eventually. But if I'm driving north again on I-75 and see I-675 as I'm approaching Atlanta, just knowing that it connects to another Interstate (any Interstate) doesn't appear to help me much. If I knew what Interstate it connected to, then I probably have a map, so the entire odd/even scheme is already superfluous to me. The numbering scheme primarily helps the person navigating without a map.

[NE2]

https://www.aaroads.com/forum/index.php?board=20.0

It seems to me that the only really useful information that odd/even can communicate is whether or not the 3DI returns to the parent. If I'm driving south on I-95 to Florida, and in the middle of Virginia I see an exit for I-295, I can take it with the security of knowing I'll end up back on I-95 eventually.

So what do you do about a 3/4 loop, like, well, 295? Have a 264/295 overlap?

[briantroutman]

An even x64 from I-64 to I-64 and I-295 from I-95 to I-95–with the overlap in the NE corner–would seem to make sense. That way, I-64 motorists could easily identify their bypass, and I-95 motorists likewise.

[triplemultiplex]

I'll skip right to the 3di's that connect to another interstate, but not back to the parent.
I generally want to go with even numbers.  To me it's more important to get across the message that "this freeway is libel to dump you onto a non-freeway if you keep going" when using odd 3di's.  But I'm not entirely opposed to odd numbers, especially when you're running low on even 3di's.

Beltlines, even incomplete ones, should have one number (sorry, MSP) and should most certainly be an interstate regardless of tollness (I'm looking at you, Denver, Nashville, Phoenix, Orlando...).

Generally, I'd prefer to keep the mainline running through the city and use even spurs for beltlines/bypasses.

I have an odd preference for I-5xx airport spurs even though I-1xx's are slightly more common.  But I'm a huge proponent of interstate spurs into airport terminals because of the intermodal usefulness. (I've got dozens on my fictional maps.)

[Alps]

I don't agree. A beltway is even-numbered, but you would never use that to return to the parent. If you're going around randomly taking exits without knowing where they go, I have no sympathy for you getting lost. I do think the anomalies like I-540 should be corrected, though.

[agentsteel53]

A beltway is even-numbered, but you would never use that to return to the parent.

I might.  a beltway may function as a bypass of a city.  I've definitely taken 270 around the south side of Columbus, OH, as opposed to 70 straight through, for traffic reasons.

[Alps]

A beltway is even-numbered, but you would never use that to return to the parent.

I might.  a beltway may function as a bypass of a city.  I've definitely taken 270 around the south side of Columbus, OH, as opposed to 70 straight through, for traffic reasons.

But you need to know that it's useful for that purpose. I-295 around Jacksonville is unlikely to be better, and even if it is, do you head east or west?

[NE2]

I-295 around the east side of Jax is totally better (1.2 miles longer). Even more so when I-795 is built.

[agentsteel53]

But you need to know that it's useful for that purpose. I-295 around Jacksonville is unlikely to be better, and even if it is, do you head east or west?

do I take 405 or 5 through Los Angeles?  the numbering will not help you glean the traffic...

the fact that there are three options (one leg of 295, the other leg of 295, or just plain 95) as opposed to two (405 or 5) does not invalidate the use of an even number for a full circular beltway.

[Mr_Northside]

I actually agree with the a lot of the original post. 
All the technicalities of 3DI numbering aside, the simple notion of "is this route itself mostly a belt/loop/bypass, or mostly a spur" should be the primary litmus test for even or odd prefixes.

For instance, I've always thought that I-270 in Maryland should be an odd numbered 3DI; it's a functionally straight and direct route from Frederick to the DC Metro area with absolutely NO belt/bypass/loop qualities to it.

Of course, there are plenty of examples of 3DI's that, for whatever reason, don't neatly fall into either category.  (The specific example I'm thinking of is the relatively recently extended I-376 here in Pittsburgh, which is now a half belt (loop?) that then becomes a spur north.), in which case, just take an available number.

All this being said, I personally wouldn't advocate re-numbering a preexisting 3DI just for these reasons.  I doubt the average motorist knows/cares enough about 3DI rules for it to really matter (especially in this age of G.P.S.), and the renumbering benefits wouldn't outweigh any potential confusion (even short-term) and resigning costs.

[national highway 1]

I don't agree. A beltway is even-numbered, but you would never use that to return to the parent. If you're going around randomly taking exits without knowing where they go, I have no sympathy for you getting lost. I do think the anomalies like I-540 should be corrected, though.

And I-520.

In what way does I-605 CA function as a loop off I-5?

[Kacie Jane]

In what way does I-605 CA function as a loop off I-5?

Without looking at a map to remind myself of exactly which freeway 605 is, I'm guessing it's in the way that California has an extreme shortage of 3di numbers.

[agentsteel53]

In what way does I-605 CA function as a loop off I-5?

it falls in quite nicely with the "ends at another interstate" rule... on both ends, actually.  it connects 405 to 210.

[PHLBOS]

All this being said, I personally wouldn't advocate re-numbering a preexisting 3DI just for these reasons.  I doubt the average motorist knows/cares enough about 3DI rules for it to really matter (especially in this age of G.P.S.), and the renumbering benefits wouldn't outweigh any potential confusion (even short-term) and resigning costs.

FWIW, that didn't stop the Feds from mandating many states (TX & MN being the sole exceptions) to renumber all their suffixed 2dis to 3dis decades ago.  So, in theory, similar could happen again.

In the case of the recent I-376 extension (onto PA 60 & the lower leg of I-279); another 3di number could've been selected for the highway portion of PA 60 and lower I-279 could've been left alone.  Since the upper-end of PA 60 ends on I-80 and the lower ends on I-79, it could've been easily redesignated as I-x80 or I-x79; thereby retaining its status as a north-south road.

[hbelkins]

I wondered why they used 376 and not 279.

[vdeane]

Probably to avoid direction-swapping.

[hbelkins]

Probably to avoid direction-swapping.

That hasn't been a problem in a lot of other instances.When the I-265 loop is completed in Kentucky and Indiana, it will have three directions (E/W, N/S and W/E).

[Brandon]

Probably to avoid direction-swapping.

That hasn't been a problem in a lot of other instances.When the I-265 loop is completed in Kentucky and Indiana, it will have three directions (E/W, N/S and W/E).

It might have been due to the distance and the fact that it wouldn't really form a loop or even a 3/4 loop like I-265.

IMHO, the I-376 extension might've been better as either I-379 or I-580.  But, this is Pennsylvania which chose to use I-476 north to Scranton from Philadelphia.

[PHLBOS]

IMHO, the I-376 extension might've been better as either I-379 or I-580.

Agreed.

But, this is Pennsylvania which chose to use I-476 north to Scranton from Philadelphia Chester.

FTFY 

[cpzilliacus]

For instance, I've always thought that I-270 in Maryland should be an odd numbered 3DI; it's a functionally straight and direct route from Frederick to the DC Metro area with absolutely NO belt/bypass/loop qualities to it.

I agree.  Since Maryland no longer has an I-170, its I-270 really ought to be I-170.  And that's still a lower-numbered 3DI than its short spur, I-370.

[cpzilliacus]

We all know the numbering rules for Interstates, including at least a vague notion of the rules for 3DIs–first digit odd: spur, first digit even: loop/beltway.

Should 4DIs be allowed in states that "need" more 3DIs?

[hbelkins]

Why not? Kentucky has four-digit state routes, and Virginia has four-digit secondary state routes.

[WillWeaverRVA]

Why not? Kentucky has four-digit state routes, and Virginia has four-digit secondary state routes.

Well, we actually have FIVE-digit secondary state routes, they just haven't made it to shields yet.

[Brandon]

We all know the numbering rules for Interstates, including at least a vague notion of the rules for 3DIs–first digit odd: spur, first digit even: loop/beltway.

Should 4DIs be allowed in states that "need" more 3DIs?

There is, effectively, already a 4di: I-H201, if you consider the 'H' as part of the number.

[Mr_Northside]

All this being said, I personally wouldn't advocate re-numbering a preexisting 3DI just for these reasons.  I doubt the average motorist knows/cares enough about 3DI rules for it to really matter (especially in this age of G.P.S.), and the renumbering benefits wouldn't outweigh any potential confusion (even short-term) and resigning costs.

FWIW, that didn't stop the Feds from mandating many states (TX & MN being the sole exceptions) to renumber all their suffixed 2dis to 3dis decades ago.  So, in theory, similar could happen again.

I suppose it is possible.  Though I'm not betting on it.  And I'd be against it (not that the Feds would call me up and ask my opinion).  Like I said, personally I think it's just as well to let sleeping dogs[questionable/incorrect first 3DI digits] lie.

In the case of the recent I-376 extension (onto PA 60 & the lower leg of I-279); another 3di number could've been selected for the highway portion of PA 60 and lower I-279 could've been left alone.

Though, that would have defeated part of the purpose of what they were trying to do. 
Part of it was business reasons; the belief that an I-designation will help development, and that Pittsburgh International Airport needed "to be served" by an interstate.  Just how important/beneficial it actually is, I think, can be debated....
But the rest of the reason was to unify a continuous highway that was a hodgepodge of numbers under one route designation, so they wouldn't want one number to just end, and other to just begin.

Should 4DIs be allowed in states that "need" more 3DIs?

I'd have no beef with that, in situations where there is no other option.