Originally Posted by
FMS82
just from the top of my head;
- AA587 in 2001
- The recent Atlas 767 nose over
- UPS1354
all very flyable aircraft flown into the ground in benign weather. We can debate training standards I guess, but even in decent outfits, low performers slip through the cracks... If we include a bit broader scope (not just us majors) the list becomes long (Asiana in SFO comes to mind most prominently, the fairly recent Emirates botched go-around, Turkish in AMS, those PIA cowboys landing on their engines and going round again...)
Thanks for your examples. I had to think about it for a few days, and now I come to the conclusion that these accidents actually prove my point.
AA587 was 20 years ago! And it was a failure in proper training. No serious training organisation will nowadays advocate what these pilots did. On the contrary, worldwide, UPRT training has been introduced, to teach pilots how to deal with such situations. A similar accident has not happened again and it is hard to imagine it will happen again in any seriously trainig airline.
UPS1354 was a combination of huge fatigue and non precision approach. Such an approach could not be flown by a fully automated airliner with the necessary low failure rate. So to go fully automated you would need an approach with lateral plus vertical guidance, which then would also have been flown successfully by the crew of UPS1354. It is kinda an argument that bites it's own tail. Plus, fatigue awareness is on the rise worldwide, amongst pilots, amongst unions and also amongst employers.
And also the Atlas 767 nose over proves my point: the FO had a history of training problems, he probably would not have made it into a flight deck with proper selection and training standards.
And against that background, I can only see safety stay at an acceptable level by moving the pilots out of the loop, where they can not do the damage they have done in the past two decades.
There are nowadays extremely few accidents in properly regulated air traffic systems. We still can improve even amongst them with regard to selection and training. UPRT has been introduced to avoid loss of control accidents, which is good.
Additionally, you completely ignore the thousands of decision pilots take every day to prevent things from escalating.I feel your believe in the reliability and the decision making capability of a complex machine in a complex environment is naive.
On a side note, a single pilot cruise airliner, I would also not exclude anymore. It has the advantage that even on two man red eyes one can sleep, and on longer sectors you do not share rest amongst three guys but amongst two guys, prolonging rest time. So it is not all bad.
But a fully autonomous or single pilot airliner won't happen in the next 30 years.