But… why? What actually happened?
There doesn't exist an automated way for the crew to log in the system that the flight is canceled.
The crew of the cancelled flight calls in to a call center in a central office. There's a person there on the other end. The crew tells them that SW1234 or whatever was canceled. The person manually enters into SkySolver that the flight was canceled. SkySolver runs the algorithms or whatever and reschedules/reshuffles planes and crews to ensure that all passengers get to where they're going. SkySolver automatically sends out the corrections to all the crews and agents what the new crew/aircraft/flight assignments are.
This works great if the number of canceled flights is less than the capacity of the call center. But if the number of canceled flights exceeds the capacity of the call center by 1 aircraft, SkySolver does not know that the canceled flight did not arrive at its destination, and is unable to perform its next leg. It does not know that there's still an aircraft taking ramp space, so when/if the next flight arrives they have nowhere to park. So your 1 cancelled flight turns into 3 cancelled flights. But we're already over the capacity of the call center; so those 3 cancelled flights turns into 9 cancelled flights. 27 cancelled flights. 81 cancelled flights. 243 cancelled flights.
All cancelled flights.
Essentially the system never know if a flight was successful or not, it just assume it went well, in the case of a delay or a change some employee has to input data manually into the system. For a few flights that's ok, but for a exceptional event like this the problem snowballs and now the system doesn't know were planes, crew and passengers are.
Quoting the top comment by 4Sammich here for those who refuse to touch Reddit:
> I have friends in CS and the hotel assignment side too. There were 2 specific problems, the software for scheduling is woefully antiquated by at least 20 years. No app/internet options, all manual entry and it has settings that you DO NOT CHANGE for fear of crashing it. Those settings create the automated flow as a crewmember is moving about their day, it doesn’t know you flew the leg DAL-MCO it just assumes it and moves your piece forward.
> In the event of a disruption you call scheduling and they manually adjust you. It does work, it just works for an airline 1/3 the size of SWA.
> So the storm came and it impacted ground ops so bad that many many crews were now “unaccounted” for and the system in place couldn’t keep up. Then it happened for several more days. By Xmas evening the CS department had essentially reached the inability to do anything but simple, one off assignments. And to make matters worse, the phone system was updated not too long ago and it was not working well.
> Last nite they did a web form and had planned to get the system up as much as possible with what communication they could muster, however it was too much to keep up on and ultimately the method for tracking crews failed again.
> This 100% is at the feet of all management who refused to invest in technology updates because it is the southwest way to be stuck in 1993. Heck, they still do 35 min turns on a -700 and 45 on an -800 frequently with only 2 man gates. But the good news is HDQ has a pickle ball court now.
> Edit: I just realized I never added the 2nd issue. Staffing. When the weather hit all those stations at once the ramp crews had to work in shifts to not become injured due to the cold. That slowed down the turns and backed up the planes. Many many ramp staff quit because of the management harassment (Denver) and just over it. So many rampers are new and making around 17/hr. Once they lost so much staff the crew scheduling software inputs couldn’t keep up because CS is also woefully understaffed and it became what we have today.
> This 100% is at the feet of all management who refused to invest in technology updates because it is the southwest way to be stuck in 1993.
https://www.reddit.com/r/SouthwestAirlines/comments/zxg6op/t...
The unnecessary, axe-to-grind editorializing undermines some of the credibility here, imo. There's no way this event hinged on a simple, binary "invest in tech" or "don't invest in tech," and that execs thumbed their nose at the latter.