My PhD is in chemistry, not traffic control, but please bear with me. I have been seeing for the last few years in your column and other places references to ‘scientific studies’ on the last-second ‘zipper’ method of merging traffic. I have tried to find original studies on it, and as far as I can tell, all these alleged studies are mere computer simulations. They have no real-world data.
Chemists use a lot of thinking and computer models to predict what will happen in a chemical system. But we never publish something as fact without actually doing lab experiments to test our hypotheses. The zipper people have apparently never gotten roads, cars, and stopwatches involved. They put what they think should happen into a computer, and that is what it predicts. Garbage in, garbage out.
Two things that definitely did not go into their calculations are HACCP (Hazard Analysis Critical Control Point) and human nature. The fact that traffic generally flows 40% faster in two lanes than in one is irrelevant. What matters in any process is the Critical Control Points. In the case of the zipper, they are compressing the critical control point to 50 feet, and the time to about 2 seconds. Human nature ensures that the maximum number of cars will be concentrated at that point. Any rational margin of safety by experienced and cooperative drivers will dictate that the speed will be very slow.
This could be actually tested by some state highway patrol. Take a spot and time of day where a backup is occurring, and implement the American Automobile Association recommendation of enforcing a gradual merging ½ mile before the actual last-chance spot. I saw this being done years ago on I80 in Indiana. They put up a sign, ‘Illegal to be in left lane past this point’, and parked a squad car under it. Merging occurred at 50mph and there was no backup, as opposed to the walking speed and half-hour backup of an unregulated merge.
I was in Florida in April, and I found another tactic which was possibly accidental, but nearly as effective. They gave no warning. The first sign they had was, ‘Left Lane Closed’, and it was closed in about 100 feet. There was no chance for aggressive drivers to pass everybody, no reason for right lane drivers to be resentful. They zippered smoothly. The traffic was heavy, but slow (Florida 1) but there was no backup.
Anybody but a lane-crasher knows all this without a study.
Hopefully, someone will see this and do some actual scientific experimentation that leads to better policies that save gas, time, and grinding of gears and teeth.
I’ve spent enough time in traffic to see th3 benefit. The idea is that the faster cars can go on a roadway, th3 greater th3 capacity. Slow cars down and it can handle less traffic. Using two lanes and a quick merge provides increased capacity. Same as driving 70 versus 55.
The Minnesota state patrol just had an article again on this. I’m sure Mn dot has some evidence since it was like pulling teeth to get them to change. Still I see some foolish truckers blocking lanes and forcing cars into a dead stop single lane a mile back. I think it is more math or physics than chemistry though but works fir me.
First of all just like your post in 2018 directed to Ray if he does read this forum I don’t think he ever respondes here. Sorry, I did not wade through your wall of text but the people who make lane merging a problem are not going read something or change their driving.
I live in Florida and drive a highway merge lane at least once a week exactly like this (common here). At least half the time the people in the right lane apparently don’t read the sign and make no effort to merge at all. No speed up, no slow down, no turn signal, with a car right beside them. Friction occurs with one or both running on the shoulder. I’ve not seen an accident but plenty of close calls.
You can’t fix clueless.
Zipper merging is ruined by those who zoom ahead of the slowing traffic in one lane. If instead they stayed alongside the traffic in the other lane, a zipper merge could actually be performed safely at an appropriate location, everyone continuing to move at an appropriate speed.
You have the benefit of modeling “things” that are not affected by emotion. They are expected to react the same regardless of which side of the test tube they woke up on that morning 
The universal truth on modeling- all models are wrong but some are useful.
The issue here is in constructing a model that accurately reflects human reactions under numerous, very different scenarios. Even the same human may react differently from one instance to another. So far too many variables to take into account to rely solely on modeling.
What makes you think they haven’t tried to empirically test their theories? Because you haven’t found any published works? Think how complex the test matrix would need to be in order to establish results with statistical significance…good luck.
I’m having trouble understanding how this works. How is this different than just moving the last chance merge point 1/2 mile back? Maybe people behave differently just because there appears to be a police presence that might enforce some penalty on them?
As someone that has lived in the Volunteer State all my life, people say our color is orange due to the Vols, but for those of us that have always lived here, we know it is because of the orange barrels… lol
What I almost always see is people wait until the last second to try to merge and ends up having to all but stop, or stop, and then they pull out making the flowing traffic to all but stop and therefor causing both lanes to now become stop and go traffic… It only takes one or two vehicles to screw it up…
Living near Tulsa,OK it seems as if I-44 has had construction for the last 100 years . Sure sometimes it might be a little difficult to handle the lane changes . Some times it is more problem than others but as long as reach the other side that is all I care about.
Another fun spot in OK - I-35 southbound, south of OKC. Long-time 1 lane area, folks would start merging over a couple of miles before the actual restriction, with semis blocking the open lane. Seems a bit too far to me. And those stories of ‘it works fine with earlier merging’ likely have lower traffic density than many merges around Dallas.
That guy in Peoria really didn’t understand the concept either. I think the trick is to merge without slowing down much. Keep it moving. Like how the entrance ramps are suppose to work until someone stops at the bottom.
Yup!
Luckily, I’ve only encountered those clueless people a couple of times, but they are a major safety hazard.
If states wanted zipper merging to work, they would have LEOs directing traffic. I have even seen vehicles passing in the grass then push their way back in.
If this picture is showing how to do it correctly.
What is the correct thing for the white car next to the bus to do?
Speed up and cut the bus off ?
or
Slow down and the red truck will pull in behind the bus and then cut the red truck off?
As my elderly father said in the '90s, “This is now the F-you society”.
Common courtesy disappeared several decades ago.
Assuming they are all going about the same speed, the white car should slow and merge after the bus since the merge is to the left. The red truck should follow the white car into the left lane.
I think the red truck being that close to the back of the bus should slow down and fall in behind the bus and then give enough room for the white SUV to move over in front of him/her…
But we all know that either the white SUV and or the red truck will speed up and cut off the bus, OR the white SUV will slow down enough to fall in behind the bus and tick the red truck off cause he is being made to slow down, then the red truck will move behind the bus and tail gate it making the white SUV have to fall back even further and possibly ticking the other cars off now behind the red truck speed up so the white SUV has to stop and or cut someone off trying to move over, either way it will slow traffic down much more than if done correctly…
Well I didn’t look at it that close but I would say the most important concept is not to slow traffic. Therefore, the bus is likely to move slower so give it room to continue on without slowing down. So give it an extra car length.
Another thing that I have tried in heavy traffic is to allow five car lengths between cars. This allows everyone to continue smoothly without braking or slowing. Argue all you want but it works pretty well. There will always be the jerks that try to cut in and take up the empty space so just let them.
I thought about that too, and decided not to post it. Is your analysis an advertisement for self driving cars that won’t let egos get in the way of safe driving? Everyone but you and me, of course.
Zipper merging is a theoretical concept that in over 55 years of driving and over 1 million miles I’ve never seen implemented correctly.