04 June 2011

Should We Judge The Pilots of AF447 Yet?

Like many people I have read a lot of the reports on AF447's activity before the plane went into the ocean.

The behaviour of the pilot in command is not fully logical. It is easy to judge that the guy made a mistake and was doing something out and out wrong by pulling the plane into a nose up attitude.

However after reading the report from the French Investigators I am not so sure.

It is clear that the the instruments were not providing consistent and reliable information. Even in the best of weathers the Airbus is inherently a plane that needs software to fly it. The A330 is a fine plane. I have confidence in it and have flown and will continue to fly it. In the dark in a storm with little or no spatial awareness possible and confronted with conflicting and non-existing information - the pilots were making safe decisions, conservative decisions. Or at least that is what they thought.

The BEA Report available in English and in French was correctly released to curb speculation. I don't think it gives any definitive answer as to what happened. I opined earlier that this was another example of the pilot away from the cockpit. That was wrong. It was set correctly and he did the hand over correctly.

This one will be tricky to unravel.


1 comment:

Professor Sabena said...

I received this comment - I think its appropriate to consider these points.

Prof Sabena:

The two most important questions about AF447 aren't being asked: First, why did they fly into an ITCZ thunderstorm when at least three other a/c in the same area diverted west of the buildup? (Possible answer, Air France restrictions, meant to reduce fuel consumption, on diversions.)

Second, when the pitot tubes iced up and the AHRS's went off the air, the interim report says that the system didn't report angle of attack to the pilots, although it was available to the flight data recorder. When air speed is not available, AOA becomes the only way to manage attitude and thrust to avoid stalling the airfoil. (It appears that in the several other incidents, where pitot icing led to loss of air speed indication, a visible horizon provided attitude reference. Because of the time of day and the violence of the thunderstorm it can be assumed that there was no visible horizon for the crew of AF447.) This is a software system design issue covered by the topic of 'graceful degradation'. In addition, the interim report talks about the trimmable horizontal stabilizer going to full up trim. This is not something done by the pilot flying. It is another software system design issue that complicated the pilot's problem. All of these software issues are the responsibility of the airframer, Airbus.

It is always easy to blame the dead guys. The ultimate responsibility of the crew is to arrive safely and they didn't. The responsibility of the bureaucrats and designers is less obvious, but is at least as important. This is why NTSB procedures include, in addition to reps of the airline and manufacturer, an ALPA rep on the accident board. I don't see any such rep on the BEA board.

Jeff O'Byrne