Algorithm of quick boarding

Surely, many experienced this feeling, standing "at the door" of an airplane in the queue for landing of accommodating passengers, - I want to sit down, buckle up and take off as soon as possible. But there are a lot of people in front of you, and the queue barely crawls. And it seems that everyone was ready for planting, but this is not done quickly. Have you ever wondered if there is a more efficient way to quickly get everyone in their seats and finally take off?

Millions of other people have probably already pondered this issue. Finally, one of them wrote a program to find the answer. Jason Steffen , an astrophysicist from Northwestern University, a few years ago modeled various methods of boarding to find out what was slowing her down. At the same time, he realized how airlines can speed up this process.

Most foreign airlines land from tail to nose, which, at first glance, is reasonable. But Steffen claims that this is one of the worst ways to plant. He believes that "traffic jams" appear because of the loading of things on the upper shelves above the seats. Even if the queue was perfectly lined up, only one or two people could pack their bags at a time. At this moment, those who are waiting for their turn block neighboring baggage shelves for several more rows.
')
In his model, Steffen began with the assumption that people would automatically go until they hit the person in the back of the front or reach their row. Spaces are filling up quickly, but the queue is moving slowly, as simulated passengers are waiting for the opportunity to put their luggage on the shelf. Even if the first person in the queue sits in the last two rows, only a few people will be able to throw bags onto the luggage rack in time, and everyone else will have to wait. “Many people think that all that needs to be done is to move the queue inside the aircraft,” says Steffen, “But the queue will not move faster.” The landing time is almost the same as when landing from tail to nose.

Steffen found out that people need to be guided by someone so that there is free space between the accommodated passengers and no congestion. So he began experimenting with queue management, changing the number of seats on which every next passenger sits. He started the simulation again and again, and each time the order of landing of two random passengers changed. If the plane was filled faster, the order found was kept as promising. If it is slower, it was chosen new. With each iteration, the program was getting closer and closer to the optimal order of landing.

In the end, the scheme was formed. The first person in line should sit in the last row, at the window seat. The next one is also at the window, but two rows closer. The queue should be seated in such a way that a row is passed after each passenger sitting at the window, right up to the nose of the plane. The same on the opposite side. Then fill the space at the windows in the rows between the already occupied. The scheme is repeated for medium spots on each side, and then for extreme ones.

According to Steffen’s calculations, his method makes landing at least five times faster than traditional tail-to-nose filling (efficiency varies depending on the size of the aircraft). And it works not only as a model. A few years ago, the Steffen method was tested on the web show This Vs. That is where he surpassed five other methods .

Airlines have already tried many tricks to try to speed up the landing, and Steffen is not the first scientist with a promising solution . But today, according to Steffen, only Virgin America has contacted him, and nothing came of it anyway. He admits that the problem with his model is that it does not correspond to human nature. For example, it’s hard to imagine that people traveling together are divided just to get everyone in as soon as possible (especially if there are small children in a group). Add to this the language barriers that are late and the general difficulty of joining large groups to follow complex routes, as it becomes immediately clear why the idealized model of Steffen has not yet been translated into reality. We need special control over passengers, which airlines simply do not have.

In addition to different variations of landing from tail to nose, landing from the windows to the extreme places (United Airlines) and free landing (Southwest Airlines) are also common. Steffen says that the Southwest method is better than landing from tail to nose, but its main drawback is that passengers choose their seats themselves and, accordingly, accumulate crowds that lead to the same traffic jams. Jason says that landing in a truly random way will work fine.

Given all the above, it seems that we still have to put up with inefficient and slow bursts for landing. Maybe now it will be easier for you to transfer expectations at airports. Steffen does not give up and is now working to accelerate the movement of people through checkpoints.