Merge Late, Save Gas and Time

You know the moment. You are trucking down the highway and you see the orange sign ahead: construction ahead, lane closes in 2 miles. What do you do? Well, you want to be polite, and you know the lane is about to close, so you move over right away. That's the best thing to do right?

Wrong. Believe it or not, our gut instinct to move over right away, and our consequent opinion of those who zoom by us only to merge later as rude jerks, is very inefficient and actually causes more traffic jams and slows our progress more than if we all did what the "jerks" do and merged late.

This can be demonstrated in several ways. Intuitively, consider why we have multi-lane roads in the first place. It is in large part so faster traffic has more chances to pass slower traffic. In one lane, we all go only as fast as the slowest car in front of us. With two lanes we pass and go as fast as we like. The principle holds with going from 2 lanes to 3, or 3 to 4, etc. More lanes is faster on average. But if we all merge early, we are essentially eliminating a lane, thus slowing our progress.

Another way to get behind this idea is go to the extreme. If the sign said "lane ends 100 miles", you wouldn't move over. Well why not? If moving over 2 miles in advance is good, wouldn't moving over 100 miles in advance be even better? Of course not.

The biggest problem with early merging is that it often crosses the density threshold that causes traffic jams in the first place. Have you ever wondered how traffic comes to a stop even in the absence of an accident or some other event that forces that first car to stop? Why would the first car that stops do so? They do so when the congestion of traffic reaches the point where the car in front of you is so close that when it hits its brakes, you have to hit yours, and the car behind you has to hit his, etc. So it becomes akin to "the wave", creating a permanent halt at that point until traffic lessens. But now think about what happens when the next person hits their brake. That makes a second slow down point, right? As this continues with constant density, and hundreds of braking events, eventually, that last car barely moving hits its brake, and comes to a stop, and the jam is made.

When we merge early, we do a lot of braking to do it. This pushes us towards that jam threshold. If we merged more smoothly, the odds of hitting that threshold is lowered, and reduces the time it exists if it is reached. So when you see the "lane ends" sign ahead, don't panic, and don't feel like you need to brake and swerve over right away. Take your time, merge smoothly and as late as you can. Of course, you don't want to be a daredevil and merge at the last possible minute, and obviously once the jam occurs and the mergers are stopped, there's no point in zooming past them (then you WOULD be a jerk). But if we could all merge later, we save time, evergy, gas, and a lot of aggravation, with no more effort than we put forth before.

There was actually a study done a few years ago that supported my contention, but I have been unable to find it. If anyone can, I wuld appreciate the link.