Gate Agents - Aircraft Boarding Timing Data

Every simulation of aircraft boarding starts its clock at the wrong moment. Most aircraft-boarding comparisons I found start the clock when the first passenger enters the aircraft. But the boarding order does not appear there by itself. Someone has to create it at the gate. I saw a simulation asking why airlines do not use the fastest boarding method. My background is Engineering and Maintenance, but I wanted to explore the question. Then I remembered relocating from Greece to Cyprus with my family, three suitcases and everything else in a container somewhere at sea. If somebody had tried to organise me into a perfect sequence that day, I might have become the subject of the next unruly-passenger report. And now I wanted to model the passenger frustration! So I built a simulator that starts the clock earlier: at the announcement to prepare for boarding. Within each comparison, the three methods receive the same 180 simulated passengers: same seats, bags, families and starting stress. Only the boarding method changes. Strict Steffen is genuinely the fastest at filling the cabin: Strict Steffen: 11:39 Random: 17:24 But from the first announcement to the last passenger seated: Random: 20:44 Back-to-front: 23:04 Strict Steffen: 25:01 (remember that I assumed 4 sec per passenger for the gate preparation, which is not the reality. More than 20 sec could be more realistic). The fastest method inside the aircraft finished the whole journey more than four minutes last. Why? In this model, building its exact queue takes almost 13 minutes. Now the important part: this result depends heavily on my assumptions. I assumed one passenger is called every four seconds and that all 180 must be correctly positioned before boarding starts. I could not find a published operational figure for this. Reduce the calling time enough and Strict Steffen wins again (2.5 sec per passenger). There is a slider in the simulator, so you can test it yourself. I also model a cost when people travelling together are separated. Strict Steffen separates 54 of the 180 passengers in this run. But there is evidence pointing the other way too. One experiment that measured both speed and passenger satisfaction found that the fastest method was the least liked. My model does not yet include the value passengers may place on an orderly process. So this is a hypothesis with its assumptions written down, not a finding and not an operational recommendation. Try it: flybycode.com/boarding Code, data, sources and limitations: https://lnkd.in/dJ92Qv4s If you work at a gate and have real timing data, I want to hear from you. That is the number I am missing. Visualisation inspired by Adam Jacobs.