How can we stop it from happening again to the best of our ability?
Posted in another forum in response to the question:- “What might an Operations or Training manager learn from the AF 447 accident?”
General points:
1. Remind ourselves that it is not possible to understand, and thus predict, how humans will react in all situations, and that …
2. Humans cannot understand all of the interactive aspects of new technology in their operating environment (man, machine, and organisation); either from an operators view or that of a manager / regulator.
3. The operating environment has many threats which are believed to be contained, but can still pose serious problems due to changes in the aircraft, or the operation, or human behaviour, e.g. ice crystals, new aircraft type, operating closer to Cbs. The industry at large and individually we must be aware of the hazards due to change, assumption, and complacency.
Specific points:
A. Add knowledge of ice crystals to the Cb threat, reinforce the need to avoid Cbs by a large margin – more than any ‘legal’ minimum distance.
B. Re-evaluate situation assessment training, and surprise / stress management behaviours.
C. Re-evaluate SOPs for flight instrument failures and flight with unreliable airspeed (UAS). What do crews need to know to determine the difference?
• Determine in what circumstances each SOP might be used, state the assumptions made, and need to consider alternative actions.
• Reduce the complexity of drills – is there a need for a table of attitude vs wt vs thrust vs altitude for UAS. Confirm what is important and why – what are the assumptions.
D. Asses the crew’s dependency on automation; does this affect the currency of hand flying skills. Question if the use of the Fight Director detracts from basic instrument flying skills, or knowledge / rules of thumb for aircraft attitude / power setting for various stages of flight.
E. Consider what drills / crew action might be required in critical situations if either the crew misidentifies the situation, or with good awareness, acts incorrectly: ‘what if’; is there an 'undo' option'.
F. Report and share all incident / event data; follow up all technical investigations with a HF view. Share data with other operators, and seek data from other operators and consider applicability in your operation.
G. Require that aircraft type training identifies the significant differences between FBW and conventional aircraft control, and that the crew are trained in these features, e.g trim followup (autotrim) – when to check / use manual trim, lack of / differnt control force feedback or change of force with speed – what alternative crosschecks could be used.
Key items:
Safety management, safety culture, a learning culture, professionalism, reduce complexity.
Remember that certification requirements only provide a minimum standard.
The company Philosophy and Policy should reflect the need for a safety margin in all aspects of operations, publish this in Procedures (and training material), and Practice this both in training and operations – then check. PPPP
Epilogue:
Occasionally the industry encounters situations beyond the limits of certification, these are ‘black-swans’ where the industry depends, either consciously or not, on the human rescuing the situation. We celebrate many notable successes. Unfortunately we have to suffer failures because the situation is beyond human capability; this hurts our pride, beliefs, and our professional standards, yet correctly we search for a solution, we have to keep on doing our best.
In such circumstances the human is still best placed to evaluate and judge the situation; but the human might benefit from some generic human training to improve awareness, managing surprise, and recall knowledge; aspects of higher professional standards. (Excerpt from an earlier post to this forum).
And remember … (taken from a related blog) “… the software quit before the human.”