Currently, the largest threat to aviation safety is the Loss of Control In Flight (LOC-I), and many suggest that this has to do with pilot training. The figure below shows the number of fatalities versus the cause, indicating the significance of LOC-I to aviation safety. LOC-I is the largest column to the left based on data collected and analyzed by the Commercial Aviation Safety Team (CAST) over the past 10 years.
Hindsight is a wonderful thing. If only we could always turn it into foresight.
In terms of safety, aviation enjoys a sky that is predominately clear and blue, as a result of continuous improvements to aircraft technology and training techniques. As technical improvements have mitigated threats in areas like Controlled Flight into Terrain and Traffic Collision Avoidance, the relative number of incidents due to Loss of Control In Flight, or LOC-I, have however not changed, making this the current number-one commercial aviation safety threat. As reported by Lambregts et al in 2008, LOC-I incidents can be split into several categories, as shown in the figure below.
Note that the above figure does not include the Colgan Air 3407 (NTSB animation below) or the Turkish Airlines 1951 (Amsterdam) incidents, which also would be categorized as stall-related.
Hindsight in Loss-of-Control Incidents
In retrospect, much can be learned from replaying these incidents through flight data recorder analyses, and putting oneself in the position of the flight crew. Here are two recent examples: A Q400 reducing power to land, approaching a stall, with airspeed continuing to drop. At the cusp of critical angle-of-attack, with the airplane’s protection systems announcing imminent danger, yet the stick shaker warnings are bypassed, and the perhaps startled crew continues to bring the aircraft into a fully-developed stall. The wings teeter in the opposite direction that the pilot applies control. The crew appears to attempt to maintain altitude, thereby further increasing angle-of-attack, yet never recovering.
Secondly, a 737-800 is on approach, while the crew is unaware of a malfunctioning radio altimeter (causing the engines to maintain their “retard” position), and entering the glide slope “hot and high”. As the airspeed is bled off too far, the aircraft enters a stall, and the crew is unable to recover in time.
Other examples include a CRJ on a ferry flight that stalled at FL410. The crew made several mistakes including improper stall recovery techniques. This is similar to an MD-83 that stalled at FL330 in the Caribbean. In both of these accidents there were secondary stalls which complicated the situation. Unfortunately, both aircraft and all occupants were lost.
Some modern airplanes have complex autoflight systems. This has resulted in conflicts between the pilot and the autoflight system. In a few cases the autoflight systems has trimmed the nose up responding to pilot inappropriate yoke input. When the airplane executed a go-around, the nose up pitch caused by the engines and the nose up trim resulted in a stall.
As an industry concentrating on improving training, we ask ourselves, what could possibly be done to prevent the recurrence of such incidents? How can we teach proper avoidance strategies, and if necessary, recovery techniques. What constitutes effective upset prevention and recovery training?
In June 2009, the Flight Simulation Group of the Royal Aeronautical Society held a two-day conference dedicated to the topic of LOC-I. Ironically, on the footstep of the meeting came the grim news of Air France 447 that had been lost over the mid-Atlantic. As we viewed statistics, videos, evidence, and even personal accounts, what became clear was the complexity of finding not only in containing a common causal thread, but more importantly, defining an effective solution.