Get a sneak peek at whats next for Permanent Hazards on our April 7th Office Hours!

Post Reply

[NEW PAGE] Routing performance tradeoffs

Post by DwarfLord
The last few years have seen increasing awareness among experienced editors of the tradeoffs involved in adding junction nodes. However, our documentation has not kept up with this awareness and we are now struggling with how to codify these tradeoffs for each specific situation where we'd like to add junction nodes. Instead, I'd like to suggest writing them up in one central place for reference.

Here is some text that could either be put in a new article, or added to an existing article. Once it is in place, it could be referred to elsewhere, for example by discussions of adding junction nodes for speed limits, true elevation, railroads, etc.

Big question is, where would this go? I considered the Junction Style Guide but I'm not sure it fits there. Maybe it needs its own article?
Tradeoffs of creating new junction nodes wrote:Waze's routing performance depends on its accuracy measuring the time it takes each individual Wazer to complete a maneuver. By design, Waze can only measure this time over the space of the single segment leading up to the maneuver. The longer the segment, the higher the confidence in the timing measurement. This means that adding junction nodes, and thus shortening existing segments, can affect Waze's timing.

In most circumstances this effect is negligible and it is OK to add junction nodes when justified by best practice. In specific cases where the wait for one upcoming maneuver may be different than the wait for another, however, adding a junction node can degrade routing performance.

For example, consider a highway approaching a stoplight with a short left turn light and common backups for straight ahead, but typically free-flowing right turns. Before the intersection, the highway flares out to provide turn pockets for left turners to wait and for right turners to bypass the straight-ahead backups. Consequently, the delays to pass through the intersection differ profoundly depending on whether one is turning left, turning right, or continuing straight. Waze will measure those times over the space of the final segment approaching the junction. Provided this final segment is long enough to include the turn pockets and any additional distance over which wait times could be different, Waze will accurately capture the different delays.

Now consider that, for reasons of best practice that would normally be fine, one adds a new junction node on the approach in the midst of the turn pockets. This will force Waze to average together the delays associated with turning and going straight for all drivers approaching the new junction node. Only the delays after the new node will continue to be measured separately. The loss in timing quality will degrade routing performance.

Because of this effect, one must always evaluate potential timing impacts and consider alternatives before adding a junction node. In some cases it may still be necessary to add a junction node despite the drawbacks. In others, an alternative may be found, or it may be that the benefit of adding the junction node is too small to justify the potential routing degradation.
[EDITS: Updating thread title to reflect the way the thread conversation is going.]
DwarfLord
Wiki Master
Wiki Master
Posts: 2512
Has thanked: 1065 times
Been thanked: 1451 times

POSTER_ID:16850907

1

Send a message
Last edited by DwarfLord on Fri Jul 20, 2018 10:31 pm, edited 2 times in total.

Post by Inactive user -1697532064-
I like this idea, but wouldn't the single-word title Tradeoffs work? After all, the whole wiki is about mapping
Inactive user -1697532064-
Wiki Master
Wiki Master
Posts: 1308
Has thanked: 549 times
Been thanked: 703 times
Send a message
Galaxy S20 FE on Mint
Retired SM Ohio
Then you will know the truth, and the truth will set you free.
-John 8:32

Post by Inactive user -1697532064-
Since the article is about tradeoffs, why not mention how junction boxes solve this problem but come with their own? Those would be of the "tradeoffs in the optimal generation of routes" variety. The existing link to the JB page could be updated to go to this excellent section.

Also, you have the problem that as segments get shorter, the signal-to-noise ratio involved in GPS tracking gets lower, so that also degrades route timing, especially when placing wayfinder stubs. Might want to mention that.

Even though the info is found elsewhere, as Ethan and Alan pointed out, I agree that this page has its use. I like how it isn't overly technical, but maybe it could use a little opening statement as to its introductory nature, kind of like this Wikipedia article.

And yes, gas station PLRs are a common example where this tradeoff comes into play. It would be good to mention those, although I would say that definitely if you have turn restrictions involved, we should add the PLR for a corner gas station. In those cases, the benefit to drivers far outweighs the cost in routing accuracy or of adding a JB.
Inactive user -1697532064-
Wiki Master
Wiki Master
Posts: 1308
Has thanked: 549 times
Been thanked: 703 times
Send a message
Galaxy S20 FE on Mint
Retired SM Ohio
Then you will know the truth, and the truth will set you free.
-John 8:32

Post by Inactive user -1697532064-
But the info you need has already been written. I think all you need is a couple sentences like "for a more technical treatment of route timing, see [[Routing Server]]" or whatever, coupled with "junction boxes solve the routing problems created by short segments, but they come with their own tradeoffs. See [[Junction box#When to use|when to use junction boxes]].
Inactive user -1697532064-
Wiki Master
Wiki Master
Posts: 1308
Has thanked: 549 times
Been thanked: 703 times
Send a message
Galaxy S20 FE on Mint
Retired SM Ohio
Then you will know the truth, and the truth will set you free.
-John 8:32

Post by abc1357
I'm not sure I like Performance tradeoffs. When I first saw the title before opening up the topic, I thought it may be related to performance of WME in Chrome vs. Firefox, or the performance of the various scripts on different computer systems.

Since this is about the tradeoffs of junctioning and segment lengths, why not title this article as what it is Segmenting and junctioning considerations ?
abc1357
State Manager
State Manager
Posts: 865
Answers: 2
Has thanked: 313 times
Been thanked: 236 times
Send a message

Post by dfortney
I agree we need a wiki page on this, and having a separate page referenced by multiple pages is the most efficient approach. I recommend we add parking lot roads and driveways to the list of editor actions that cause the shortening of segments and deserve consideration of routing tradeoffs. One situation deserves special mention: the U.S. has many major highways through towns and cities that devolve into a prolonged series of strip malls. If every parking lot entrance got its own road, the major road gets chopped into bite-sized pieces.

DwarfLord eloquently lays out how junctions along turn pockets can inhibit Waze from collecting accurate differential turn/transit times. In addition, there’s another way in which short segments can foul-up route timings — short segments provide fewer opportunities for GPS data collection and timings, and any inaccuracies have a larger percentage impact vis-a-vis longer segments. I don’t know the measurement accuracy of GPS on consumer mobile devices, and what impact poor cell signals or multi-tasking apps have on data accuracy, but if those kinds of factors result in several seconds of variation, and if that kind of inaccuracy is repeated multiple items along a potential route, resulting routing decisions could be woefully suboptimal.
dfortney
Posts: 281
Has thanked: 129 times
Been thanked: 104 times
Send a message


Post by DwarfLord
Thanks for the suggestions!

I must apologize, I thought I had submitted this in the USA-only forum. My last understanding is that the global wiki is effectively no longer maintained. I will seek an administrator to move it.

Nothing about that negates your thoughts, which are very helpful!
DwarfLord
Wiki Master
Wiki Master
Posts: 2512
Has thanked: 1065 times
Been thanked: 1451 times
Send a message

Post by DwarfLord
ehepner1977 wrote:It is in the US Wiki forum already.
Actually a very kind and responsive administrator (thanks vince1612!) moved it this morning, it did start out in the global forum.
DwarfLord
Wiki Master
Wiki Master
Posts: 2512
Has thanked: 1065 times
Been thanked: 1451 times
Send a message

Post by DwarfLord
There does seem to be a leaning towards a separate article, and I lean that way too. What I can't sort out is how to title it! I've been stuck.

Today I thought, perhaps we should set our sights a bit broader and leave room for the article to expand. Instead of making it specifically about Junction Nodes, why not title the article...

Performance tradeoffs in mapping

...starting out with just two sections, Principles and Junction nodes & segment lengths.

Later, we could add sections on other performance tradeoffs, like (1) road typing and routing and whether or not to use the "one type higher/lower" feature or (2) whether to set dead-end private roads to PR type. I'm sure there are zillions more examples.

Right now I am NOT advocating for any other content to the article, I'm just suggesting that calling the article "Perfomance tradeoffs in mapping" would provide a home for other concerns that are broader than the "when to use and when not to use" binary guidance generally followed by the rest of the wiki.

How does that sound? If it sounds OK I will draft a beginning "Performance tradeoffs in mapping" article for review.
DwarfLord
Wiki Master
Wiki Master
Posts: 2512
Has thanked: 1065 times
Been thanked: 1451 times
Send a message