I am pleased to see that the very comprehensive final report addresses several human factors issues that contributed to the accident. The report identifies blatant errors and places them in context of an operational accident; unfortunately this was not an operational accident and thus the investigation has chosen an inappropriate primary cause. There is also a short fall in the human factors investigation by failing to give an understanding, or at least a theory, as to why the crew shut down both engines.
The report gives the primary cause of the accident thus: (my italics, assuming ‘initial’ means primary from the translation)
The initial (primary) cause of the accident was the acceptance by the crew of the approach clearance although they were not prepared to it, namely the absence of preparation of a go-around. It led the crew to perform a series of improvised actions that ended in the prohibited override of the primary stop on the power levers.
Even with the care shown in this report why do investigators in general fall back on the operating crew for being the primary cause of an accident?
This crew was the unfortunate (possibly deficient), last link in the error chain. This accident was waiting to happen; any one of a number of previous crews who, if they too had violated the flight idle restriction (which may have been common practice) during the critical 16 secs during gear selection at low altitude may have lost control. It would at best be speculative to suggest that any other crew would have handled the unique flight characteristics resulting from full in-flight reverse and the associated confusing engine displays in any better way than did the accident crew.
Thus the primary cause from an engineering perspective could have been stated as the loss of control due to both engines giving reverse thrust or a combination of reverse and engine shutdown.
Yet in reality the primary cause was the complete failure of the safety management system, from the authority at one extreme to individual crew at the other. The safety system was riddled with assumptions, allowed violating behavior by both individuals and organisation, and lacked the use of error checking or mitigating techniques.
The aircraft crashed because the right engine remained in reverse, the left engine was shut down, and then the right engine was shut down. The failure to explain or present a theory as to why the crew shut down both engines deprives the industry of further understanding human error. (I accept the difficulties due to lack of recorded data).
As the result of assumption, error, personal failing, and many other reported circumstances, both engines were in reverse in the air. The crew sensed that something was abnormal (seat of the pants), deceleration, pitch down. They certainly heard abnormalities in prop speed (probably the dominant cue) and had confusing engine indications: very high prop rpm, increasing engine speed with flight idle selected. The crew apparently detected the error, the recovery action was unsuccessful (right engine remained in reverse).
Immediately prior to engine shutdown the crew would have had a gross thrust abnormality, the right engine remained in reverse, but the left was recovering to give positive thrust, thus the crew might have been expected to detect a right hand engine failure (dead leg dead engine from rudder input), but why was the left engine was shut down first, and then later the right?
The engine indications were split, the left engine had lower values than the right; there was no evidence that the crew detected an engine failure or called for shut down drills. Thus was the left engine shutdown by mistake?
The problem was compounded by the short time period; was the right engine shut down because of prop over speed - the failure to recover from beta range. Then why shut down both engines, was each engine was shut down by different crew members? It is the details of these actions that are the key factors that led to the crash.
If both propellers had been recovered to normal flight range then the aircraft may well have continued in a flyable condition i.e. the accident would not have happened; yet this issue is not included in the discussion, or the reason for the right propeller failing to recover from beta range (except in appendix 22). A recommendation should be at least to brief crews that having entered beta range if they then required full power (baulked landing) the power levers should only be moved forward carefully avoiding slamming to the forward stop.
The answers to the human factor issues could reside in the training programs, in previous / basic training, or other experiences including
firewalling power levers or rushing engine shutdown, but these aspects were not investigated in depth. For generic information see the
the full PSM+ICR report ;human factors issues at 4.2.12 and section 8; ops issues at 4.2.11; training at 7.2 and 7.2.4.
This accident is a lesson for all who think that by passing one defense (flight idle stop) with reliance on another, that they will not be exposed to risk. Whatever the aircraft or engine type, always stay within the limitation of the AFM.
This was a tragic accident that matched most if not all issues that were identified by the PSM+ICR project report. Also the background and the circumstances have been described previously by J Reason in “Human Error” and “Managing the Risks of Organizational Accidents”; unworthy epitaphs for those who lost their lives.