Indonesian Air Asia 8501 and the State of the Aviation Training Industry
I worry about our industry. I worry about it a lot. I worry about it because we focus on training to the almost complete exclusion of understanding. I worry because our industry tries to cure symptoms rather than causes. I worry that our industry trains because it can measure the outcome of training and ignores understanding because it can’t.Dutch folklore tells the story of a young boy who discovers a small hole in a sea dyke. Realising that this could quickly grow into a catastrophic failure, leading to country wide flooding, he plugs the hole with his finger. Despite the bitter cold he remains there all night until a search party finds him in the morning and makes the essential repairs. His quick thinking saves his country from catastrophe.
Our industry is starting to resemble finger plugging on a grand scale.
We have accidents in which crew fail to recover an aeroplane from an excessively steep angle of bank or pitch. Our response: plug the hole with ‘upset recovery training’.
We have accidents in which pilots fail to recover from a stall. Our response: plug the hole with additional stall prevention training.
We have accidents in our industry in which pilots crash on approach because they fail to monitor their air speed. Our solution: plug the hole with additional mode awareness training.
We even have accidents in which pilots seem to be incapable of flying the aircraft without the autopilot.
In every case the response to such accidents, all caused by fundamental deficiencies of pilot aptitude, seems to be additional supplementary ‘training’. We never seem to question the inadequacies of our pilots in terms of failure of basic pilot education.
We are completely reactive. We stand and look at the dyke, ready to plug the next hole as soon as it appears. But we never seem to understand that the solution is to build a stronger, better dyke.
Anticipated abnormal conditions
The role of the commercial pilot has changed dramatically over the last four decades. Initially pilots were the sole and thus indispensable means by which an aircraft could get airborne, navigate safely to its destination and land safely. Now, all these actions can, or could, be done with automation. So now the role of the pilot is to act as the long stop – to intervene when the automatics fail or can no longer cope.
But all the evidence suggests that pilots are increasingly incapable of doing so.
Just a few tens of seconds after the autopilot quit, the crew of Air France 447 had grossly mishandled the aircraft and allowed it to stall. The same inability to fly the aircraft after autopilot failure doomed Indonesia Air Asia flight 8501.
What will be the industry’s response? More ‘hands-on’ flying time in the simulator? We don’t seem to be questioning the aptitude of the pilots who progress as far as type rating. Or, if we do, then our examination of their basic skills falls short. The deficiencies are there and are plain to see if we dare to look hard enough.
The most fundamental attempts to address these built-in deficiencies are enshrined in the philosophies of the multi-pilot licence (MPL) and in Airbus’ underlying philosophy of automation.
The MPL emphasises training, training and more training on the aircraft type to be flown. It is a move away from generic flight experience on light aircraft in favour of a very high level of demonstrated competence on a particular aircraft in a particular role.
The Airbus philosophy runs even deeper. It has designed aeroplanes that essentially fly themselves unless the crew actively intervene. And even when they do, Airbus limits the range of interventions, makes them as simple as possible and guards them with underlying levels of even deeper automation. The implicit assumption being that at least some crews need all the help they can get. Or even that the aircraft needs to protect itself from misguided or incompetent crew interventions. As a result, almost anyone could be ‘trained’ to operate an Airbus aircraft.
A very high level of demonstrated competence backed up by a very high level of automation and automatic protections should serve to deal with the vast majority of potentially catastrophic situations
But training cannot deal with every eventuality. In particular it cannot deal with the totally unexpected and unanticipated because training is designed to teach specific skills which relate to expected and typical situations. Training, which is the primary (almost exclusive) focus of the pilot ‘training’ industry, is all about developing skills to deal with routine events and skills to deal with anticipated abnormal conditions.
Up to a point this approach is valid. A very high level of demonstrated competence backed up by a very high level of automation and automatic protections should serve to deal with the vast majority of potentially catastrophic situations. But both these approaches are fundamentally part of the hole-plugging approach. They don’t help to build a stronger dyke.
Unfortunately the water held back by this particular dyke is not a placid pond. It is a sea which contains wild, unexpected, random, multiple failures, exceptional conditions and bizarre pilot actions which lead to entirely unanticipated emergencies.
These are conditions for which the training and hole-plugging approach cannot work. A very highly trained pilot who demonstrates high levels of competence in trained skills is not necessarily properly equipped to cope with completely unexpected and unanticipated problems. The tools needed to cope with such dire circumstances are knowledge and understanding (in addition to other skills such as resilience, determination and grit).
In the words of one educational academic:
Knowledge and understanding allows us to explain observations and to make predictions. It helps us to prevent false hypotheses and to improvise and adapt in unexpected situations. Furthermore, knowledge and understanding are essential if we are to avoid unintentional mistakes caused by ignorance.
In the words of another:
Education and training are located at the two ends of a continuum of personnel development. Training is concerned with improving performance on the job for expected and typical situations. Education is concerned with increasing powers of observation, analysis, integration, understanding, decision making, and adjustment to new situations. (Flippo 1961).
In other words knowledge and understanding provides us with tools powerful enough to cope with the unknown.
We acquire understanding from a combination of training and education. But, how many times does one hear the term ‘pilot education’ or hear of the ‘pilot education industry’? Never.
Education is the necessary condition for understanding many things by participating with them or by experiencing them.
It doesn’t have to be a purely intellectual undertaking.
Education is the necessary condition for understanding many things by participating with them or by experiencing them. It doesn’t have to be a purely intellectual undertaking. For example solo flying during initial pilot ‘training’ is not, by definition, training. The trainee is learning, by themselves, often by exploration: ‘what happens if I do this….oops”
Quite often the solo pilot is learning what NOT to do as much as practising what she/he should do. And by learning these things the solo pilot is also acquiring understanding.
Many of the deepest ingrained lessons we pilots have learned about flying came from things we did when flying solo. Silly mistakes, small but frightening episodes where we inadvertently became lost, or lost control, or became disoriented. From these learning experiences came deep understanding of how to avoid doing the same thing again and how to suppress panic and extract ourselves methodically and safely from dangerous situations.
This is one form of education. It is immensely powerful, utterly essential, and virtually unmeasurable.
There are other circumstances where education does have to be academic rather than practical. For example the swept wing stall.
In many cases deliberately stalling a swept wing aircraft is simply too dangerous to contemplate. The problem is compounded because the swept wing stall tends not to be accurately modelled in simulators. The result is that we don’t practice it. In fact we don’t even get to see it.
This is when academic education must step in. In this example, we need to educate young pilots to understand the swept wing stall. Most particularly they have got to understand that it is a very different beast from a straight wing stall.
The final line of defence
The point about education is that the emphasis is on understanding rather than performance. A pilot who is ‘trained’ will competently recognise the stall warning and recover the aircraft before it stalls. (This is a nice measurable performance standard which our industry will instinctively seize upon). But the pilot who is educated properly about swept wing stall and understands it in depth will be able to imagine what it might be like to be in one and thus, hopefully, recognise it – even though he has never experienced it.
Through no fault of Air France or the crew themselves, the pilots in charge of AF447 had done no practical training on recognising and recovering from a fully developed swept wing stall. In fact their only practical training had been in light aircraft in which (like all of us) they learned precisely the wrong lessons about stalling for someone destined to operate swept wing aircraft. In their light aircraft they learned that a stall is easily recognisable by the sharp onset of heavy buffet and nose drop. Neither of which occurs at the start of a swept wing stall.
So really their only chance of defending themselves against this threat (the failure to recognise that they were in a swept wing stall) was to have absorbed and understood the theory of swept wing stall in class. If, like me, they received only cursory instruction on swept wing stall then some of the conditions for that accident had already been laid down.
Indeed I can speak of this from my own experience. As a young F4 Phantom Pilot I twice unknowingly entered a swept wing stall. Only recently after thinking deeply about AF447 did I realise that I had done so. On both occasions I inadvertently pitched the aircraft way above its maximum permitted angle of attack. I knew I had done so because I could see the AoA needle pinned against its stop. On both occasions I instinctively pitched nose down to regain a normal angle of attack. What I couldn’t understand at the time was why, given such high alpha, I ‘hadn’t stalled’ the aircraft. With my imperfect knowledge and understanding I believed that a stall would always be preceded by heavy buffet and pronounced nose drop – neither of which were present.
Only recently have I come to understand the truth and maybe through that understanding I can glimpse a small part of what happened in the cockpit of AF447 on that fateful night.
Like the crew of AF447 I was a very ‘skilled’ pilot by virtue of the extensive training I had received and the many flying assessments in which my ‘competence’ had been measured and quantified.
But skill, although necessary, is not of itself a sufficient condition for complete pilot competence.
Everywhere I look in our industry I see less emphasis on education and understanding in favour of pure training. And in many cases the training itself is becoming more specific and more focused on narrowly defined competencies.
As a publisher of aviation ‘theory’ books I am now more convinced than ever that a pilot’s final line of defence is knowledge and understanding. If we are to embark on the path of less experience of basic flying in favour of highly structured training for defined competencies then we absolutely must devote more time to education and understanding.
We at Padpilot place much more emphasis on deep understanding rather than mere acquisition of facts.
We treat the EASA pilot exams’ learning objectives as useful guidance but we don’t allow ourselves to be confined by them.
As a result we at Padpilot place much more emphasis on deep understanding rather than mere acquisition of facts. We treat the EASA pilot exams’ learning objectives as useful guidance but we don’t allow ourselves to be confined by them. In that respect we are deeply conscious of the cautionary words of one educationalist:
Sometimes training within education subverts the enterprise, such as training to pass an exam by mastering a set of tricks. This training is aimed at mastering the skill of passing an exam.
So now, for example, when our books describe swept wing stall we include video of a real life fatal example and we wreath it about with warnings and cautions in red type. For us it’s the start of a long road to a better system for pilot education.
Nevertheless I worry about our industry. I worry about it a lot. I worry about it because I don’t think we understand how important it is to understand. I worry because it’s only a matter of time before we run out of fingers with which to plug the dyke.