I think you will find it much worse in areas where freeways curve hard. Thus the ramp is actually a shorter route.

I have seen waze route down a ramp as recent as 3-4 months ago, so I personally continue to put restrictions on these ramps when possible.

I also still see it at a left/right split up to a road have the user go left and turn right; another thing I often turn restrict.

I certainly wish I didn't "have" to do these things but otherwise routing is goofy sometimes.

jondrush wrote:There is really no need for the split road over the highway, but it should be OK. Make sure you restrict the straight across movement from off-ramp to on-ramp so Waze doesn't route people off the highway, as it sometimes does.

In this case, he didn't split it-- the segments have existed for several years.

On the ramp-to-ramp "turn" restriction-- does that really still happen? I find those particular artificial restrictions to be jarring -- when I'm making scans across areas and see them I have to zoom in and check each segment to make sure that they're correct. The faux restrictions have been absent from central Ohio Interstate (and other) ramps for at least a year (if not longer) and I've never seen a misroute (or any URs regarding same).

Perhaps it only happens in places where there are perennial/predictable traffic jams (with average speeds slower than the ramps) on the adjacent freeways? If that's the case, I'd rather see the restrictions added as-needed rather than across the board.


Jeff

jondrush wrote:I think you will find it much worse in areas where freeways curve hard. Thus the ramp is actually a shorter route.

Are you actually running with "favor shorter distance over shorter time" set? If that's the case, then in the scenario you describe, taking the ramp is arguably the correct behavior.

In the default setting (favor shorter time), I would think that the average speed on the ramp would be dramatically lower and there would be the penalty of an intersection. After considering all that, if it is indeed still shorter time-wise to take the ramp, then -- again -- taking the ramp is arguably the correct behavior.

Do you have any [recent] examples of this happening?

bz2012 wrote: The first place I saw such complaints was in the middle of nowhere in Wyoming!
And, they were RECENT complaints, at the time, much less than a year ago.
I still run across 'recent' reports by users as I hunt for 'old ballons' to pop.

Average speeds on the segments, turn restrictions, hidden restrictions/connectivity, device GPS tracking, etc. all checked out and supported the proper routing? If this is still happening, it'd be really good to have folks look closely at them. Every time we've run across strange things like that in our area, someone was able to find something obscure causing the issue.

bz2012 wrote:At that time, I was informed by another CM that such intersections should ALWAYS be set so 'straight through' traffic was forbidden (if it was possible to do so without effecting other traffic).
The wiki still indicates this:
http://www.waze.com/wiki/index.php/Limi ... utor_Lanes

Hmm, I hadn't seen that. The wiki seems a bit inconsistent, though: this section (first "Con" of a ramp) says that there is a penalty for a ramp transition.

I've [re-]posted the question here to get some better visibility for it.