The year so far has made it clear that loss of control as a killer accident category should be firmly on the airline industry's agenda as a problem to be solved. It is not reducing when almost all other accident categories are.
But work on the issue is under way. The US National Transportation Safety Board has been carrying out a study on such accidents that were caused by pilot spatial disorientation, including those involving Crossair (January 2000), Gulf Air (August 2000), Flash Airlines (January 2004) and Adam Airlines (January 2007), to determine what remedial actions might work.
Boeing, meanwhile, has conducted a simple but cleverly designed test to determine if one of the theoretical solutions - upset recovery training for pilots - is a viable solution. The test revealed that pilots got measurably better at recovering quickly from extreme attitudes following training, but that doesn't necessarily make upset recovery training a complete answer to disorientation-induced loss of control. Why? These pilots were given a set of four predetermined situations from which to recover, and although they did not know what attitude/speed/trim condition to expect, they knew it would be unusual, and were not disoriented.
Potential solutions advanced by the NTSB's senior human performance investigator Dr William Bramble include an education for pilots about the physiological and psychological causes of disorientation, including the fact that distractions from instrument flying are frequently a precursor to disorientation. Distractions during a turn at night or in instrument meteorological conditions were common to all the disorientation cases in the accidents mentioned. A more radical solution is installing a pilot-activated automatic recovery system that has been tested by the US Air Force.
A solution not highly rated by the NTSB is changing the type of artificial horizon display, on the basis that pilots could misinterpret any of them if the feeling of disorientation is sufficiently powerful.
One effective solution is a good first officer and a company culture that allows the co-pilot to intervene effectively when the situation demands it. In all the fatal cases mentioned here, the captain was the pilot flying, and the co-pilot had a clearer picture of what was happening but failed to intervene until too late. Bramble's study cites other compelling cases in which co-pilots saved the day. That is an immediate solution available, via good crew resource management training, to airlines at no additional cost.
First officer intervention could prevent some accidents