"Down" with the zipper merge?

[AmpedUP]

Early merging before a travel lane is lost creates multiple instances of merging, which will always be superior to a single choke point with a single merge. Ever hear of parallel processing? 'Same thing here. In fact, if there was enough cooperation at merges, they could theoretically happen at full speed...something that is simply impossible with a single choke point, er, "zipper merge." That said, proper merging only works if there is significant cooperation...and that's been lost with present day drivers.

 

[HankLloydRight]

Early merging before a travel lane is lost creates multiple instances of merging, which will always be superior to a single choke point with a single merge. Ever hear of parallel processing? 'Same thing here. In fact, if there was enough cooperation at merges, they could theoretically happen at full speed...something that is simply impossible with a single choke point, er, "zipper merge." That said, proper merging only works if there is significant cooperation...and that's been lost with present day drivers.

Science disagrees with you. Your "parallel processing" analogy significantly slows down the lane people are merging into. Essentially, lane closures should work no differently than the high speed merge on-ramp onto a highway, where people all merge at the end of the on-ramp, and not in a parallel fashion into whatever break they can slip into ahead of the merge point.

 

[Al Sherman]

The zipper merge is the best way, and should be standard.

 

[doug]

Zipper is the Pareto efficient solution.

 

[HankLloydRight]

Zipper is the Pareto efficient solution.

Awesome. Learn something new every day. thanks.

But I'll probably forget its name in a week.

:/

 

[Reykjavik]

Early merging before a travel lane is lost creates multiple instances of merging, which will always be superior to a single choke point with a single merge. Ever hear of parallel processing? 'Same thing here. In fact, if there was enough cooperation at merges, they could theoretically happen at full speed...something that is simply impossible with a single choke point, er, "zipper merge." That said, proper merging only works if there is significant cooperation...and that's been lost with present day drivers.

People don't have to ignore the fact that they will soon be merging until the last instant, as they approach they assess how much space they need to have between cars when they are in the one lane, adjust speeds to make the space available, and prepare to merge. They just keep using the whole road while they do it, instead of doing the exact same thing earlier.

And you can only merge at a constant speed if there is enough room to do so. If the two lane road has more traffic on it than one lane can handle, cars will need to slow down. Sometimes the fact is the road is not big enough to handle the traffic, and there will be a slow down until there aren't as many people coming.

 

[Kipernicus]

Good points, Reykjavik.

I'm with the zipper crowd: Traffic: Why We Drive the Way We Do (and What It Says About Us)

Final thought: WWAPD (What Would Auto-Pilot Do)?

 

[DPDsModelS]

Agreed.

However, the site talks simply about two lanes merging to one and uses the high traffic construction zone example. It is difficult to argue with this logic (IMHO).

However, does the same logic follow for a highway off-ramp or turning lane?

Theory says you're always supposed to merge at the end or as "late" as possible.

see:
Zipper Merge
Questionably Late | Before you get mad at people who “cutâ€￾ in front of...
City of Saskatoon (via noodls) / ZIPPER MERGE AT HIGH-TRAFFIC CONSTRUCTION ZONES