I’ve spent a long while trying to make sense of the data NASA’s National Aviation Operational Monitoring Service Information Release. Making sense of the data is difficult at best. There’s been a lot of fuss made about this, statements that the findings were “so scary” that NASA tried to bury the data.

From my quick glance at it, I don’t think that’s the case at all. I think this study, while a great idea, is rather flawed in its execution. I think NASA realized this, and that’s why they’ve tried to bury it–mostly because it ended up being a waste of $11 million. The information was obtained in telephone interviews, but the people interviewing the pilots and recording the data didn’t speak the same language as the pilots. NASA basically contracted with a bunch of social scientists to conduct the survey, and reading some of the summaries makes it obvious that the folks asking the questions didn’t really understand the answers, particularly to the free-form sections. Furthermore, I think that it’s easy for the public to be a bit alarmist about the whole thing since they tend to be quite skittish about airplanes in general, and here are a bunch of pilots talking about their biggest safety concerns.

If we look at the safety record of the airline industry, we see the flying is safer than hurtling through the air at 500mph in an aluminum tube has any right to be. Are there still problems? You betcha, and a look at the pilot’s free-form responses to the question, “Was there anything not addressed in this survey that you think should be addressed in future surveys?” really reveals what pilots think are the biggest problems in airline safety: Radio frequency congestion and pilot fatigue. This isn’t news, and little is being done to combat either problem. Radio discipline among airline pilots sucks (and is only somewhat better among Air Force pilots I’ve flown with) and significantly contributes to the basic problem which is too many planes, too few controllers and too few frequencies. Pilot fatigue is the bigger issue. The current regulations on duty rest are stupidly complicated and don’t do anything to address the biggest problem: airlines failing to take into account circadian rhythm and/or actual sleep when calculating rest requirements. For instance, if you’re scheduled to fly 8 hours in a 24 hour period there must be a rest period of at least 9 hours in that 24 hour period (could be reduced to 8 hours if you get 11 hours compensatory rest in the next 24 hours). Now, that doesn’t mean that you get 8 hours of SLEEP, just rest. Rest starts 15 minutes after the engines are shut down at the gate, and ends when you show up at the airport for your next flight. So, 8 hours of rest is going to be 6 hours of sleep if you’re lucky and the hotel is close. But, there’s also nothing that says that rest can’t be from, say, 6am-2pm. That would obviously not result in a well-rested pilot if he’s used to sleeping normal hours. It’s entirely possible that a pilot could be asked to work nights 4 days in a week, then switch and work days. Pilots call it “flip-flopping” and it’s dangerous, but it’s legal and the airlines do schedule it. Pilots responses in Block D3 of the NAOMS survey reveals pretty plainly that they’re worried about it without having to do crazy machinations on the data.