Then you are senior enough to know what “ then-CEO now-chairman Gary Kelly” really meant was “I haven’t funded our technology team well enough to have resources to test a scenario like this”.
Or "we decided that the cost to plan for this is so high that it's not even worth testing. If it happens then we're fucked anyway and we'll just eat it."
You can drill the initial failure, but not really the cascading events. In something as large as a global airline you are dependant on 1000s of third parties actions and the weather. No simulated drill is going to be sufficient or realistic. The only way to really mitigate or plan for something like this is multiple layers of segregation so that events in one area have less or no impact on others. Then you could drill total failure in various segments.
Testing reveals the presence of bugs, never their absence. With hindsight you can always feel smugly superior in saying “you should have tested for this”, but there’s an infinitude of things you might need to test, and if you haven’t encountered a failure you didn’t test for, you’re probably just lucky.
I’m very certain Chaos Engineering is known in the airline industry