Turkish airliner crashes at Schiphol
Join Date: Dec 2008
Location: DXB
Posts: 513
Likes: 0
Received 0 Likes
on
0 Posts
Loss of confidence sets in and soon the pilot convinces himself that twiddling the automatics and pressing buttons is easier than hand flying.
In the Turkish accident the FO's reaction to the stick shaker was only to increase thrust, without simultaneously disconnecting the AP...
That's also a point I was trying to make in the BA038 thread where the PF didn't take manual control as the automatism was maintaining an unadapted path.
Preferring an unachievable path controlled by an automatism over adapted manual control of a visual, stable and powerless short final demonstrates the existence of such reluctance.
Using mixed manual and automatic inputs isn't a good idea. If you loose AP take manual thrust control and vice versa.
lederhosen, it is difficult to disagree with your observations of airmanship (# 2643). Mainly because you constrain yourself to a single view of an accident – the person model, which usually concludes with blame. Without elaboration there is little opportunity for debate; we have to fix airmanship – how?
First what do you mean by airmanship – it’s probably different to my understanding and use.
I like Tony Kern’s model – discipline, skill and proficiency, knowledge, situation awareness, and judgement. Thinking skills are both within and link these points; critical thinking, how we think, why we think, and awareness of bias.
Using such a framework it should be possible to debate what aspects were absent or which link failed, thus contributing to the accident.
Remedial activities can be identified and assessed for practicality and cost effectiveness.
In this instance, these almost certainly include organisational aspects of training and operations, and thence economics. This immediately opens a wider view, alternative models of accidents, contributions, and thus options for improving safety.
Simple statements of ‘poor airmanship’, ‘fly the aircraft’, ‘CRM failure’, etc, do little to progress the cause of safety. They often stifle debate or indicate a closed mind or withdrawal from the subject “I know what happened – it wont happen to me’.
This is not necessarily wrong, providing the individual understands why ‘it’ won’t happen to them, and what they are going to do about it, but if not, then the attitude could represent a lost opportunity to improve … their airmanship?
Accident Models.
Critical Thinking .
Perspectives on Human Error.
Airmanship training for modern aircrew.
Fostering successes rather than reducing failures.
Attention to safety and the psychology of surprise.
First what do you mean by airmanship – it’s probably different to my understanding and use.
I like Tony Kern’s model – discipline, skill and proficiency, knowledge, situation awareness, and judgement. Thinking skills are both within and link these points; critical thinking, how we think, why we think, and awareness of bias.
Using such a framework it should be possible to debate what aspects were absent or which link failed, thus contributing to the accident.
Remedial activities can be identified and assessed for practicality and cost effectiveness.
In this instance, these almost certainly include organisational aspects of training and operations, and thence economics. This immediately opens a wider view, alternative models of accidents, contributions, and thus options for improving safety.
Simple statements of ‘poor airmanship’, ‘fly the aircraft’, ‘CRM failure’, etc, do little to progress the cause of safety. They often stifle debate or indicate a closed mind or withdrawal from the subject “I know what happened – it wont happen to me’.
This is not necessarily wrong, providing the individual understands why ‘it’ won’t happen to them, and what they are going to do about it, but if not, then the attitude could represent a lost opportunity to improve … their airmanship?
Airmanship is a personal attitude to flying, why we do it, how we do it. Airmanship must grow with training, experience, and personal exposure. It is not just about staying alive or not bending the airplane or yourself, it is about walking off the airfield knowing that you have both performed and crafted an activity. You have been totally aware of what you have done and why you enjoyed it, and at that point you owe nothing to anyone. Tony Hayes, CFI Brisbane Valley Leisure Aviation Centre.
Critical Thinking .
Perspectives on Human Error.
Airmanship training for modern aircrew.
Fostering successes rather than reducing failures.
Attention to safety and the psychology of surprise.
Join Date: Jul 2002
Location: UK
Posts: 3,093
Likes: 0
Received 0 Likes
on
0 Posts
One is that the pilot is a necessary evil, a liability if you like, who invariably makes incorrect decisions & can get easily overloaded in high workload situations. In this scenario, technology is the final line of defense. The goal of this camp is to restrict the input of the fallible human to the minimum & to eventually get the pilot out of the cockpit (literally) as soon as the flying public will accept it.
Last edited by DozyWannabe; 3rd Feb 2010 at 14:56.
Join Date: Apr 2003
Location: Germany
Age: 76
Posts: 1,561
Likes: 0
Received 0 Likes
on
0 Posts
Not enough information...
I think we need to read the final transcript to see what the ATC clearance given to the accident crew was. Okay, the line-up was at 5.5 nautical miles, obviously well above the glideslope if they were still at 2 thousand feet (most probably with a full "fly down" GS indication, so not very much useful information there plus a full-scale deflection usually mandates a go-around according to SOPs) but why did they end up with that and then what did they do about it?
You know that it used to be "pilot error" most of the time there was a crash, unless perhaps a wing came off or the damned thing exploded in mid-air. Before we knew about wind-shear several accidents were down to "pilot error" when a highly experienced and qualified crew flew into the ground on short final for no obvious reason.
Now we have "poor airmanship" as the modern version of that, all too often. Just as accidents are usually the product of a chain of events, what we call "poor airmanship" can also have multiple causes, when I think we need to keep back a bit to take a broad view. Otherwise that all-too-human tendency to just think "Well that guy was a muppet but I am the ace of the base so that I should be okay, finish palaver!" takes over, when we learn nothing useful.
The Dornier 328 had a particular tone for autopilot disengagement called a "cavalry charge," a sort of weird little blurp-blurp that made me think the German cavalry must have been on a diet of beans there, and when things were not too busy I used to do a bit of hand flying. These automation-fetishists I flew with would groan and look at me, when I would just tell them that I was getting ready for my next sim session (about the only excuse that was plausible).
"But you just came back from one and the next one isn't until six months from now!" Yeah, but it's good to be able to hand fly, just in case you need to. Otherwise you would see someone trying to input stuff into the FMS for a last-minute change of plans, head down punching those little buttons when prudence suggested that heads-up looking out might be the way to go sometimes. But for that you do need to be comfortable hand-flying your aircraft, yes...
Generations of young students have been told that "the autopilot can fly the aircraft better than you can," when I never thought that the way to go.
You know that it used to be "pilot error" most of the time there was a crash, unless perhaps a wing came off or the damned thing exploded in mid-air. Before we knew about wind-shear several accidents were down to "pilot error" when a highly experienced and qualified crew flew into the ground on short final for no obvious reason.
Now we have "poor airmanship" as the modern version of that, all too often. Just as accidents are usually the product of a chain of events, what we call "poor airmanship" can also have multiple causes, when I think we need to keep back a bit to take a broad view. Otherwise that all-too-human tendency to just think "Well that guy was a muppet but I am the ace of the base so that I should be okay, finish palaver!" takes over, when we learn nothing useful.
The Dornier 328 had a particular tone for autopilot disengagement called a "cavalry charge," a sort of weird little blurp-blurp that made me think the German cavalry must have been on a diet of beans there, and when things were not too busy I used to do a bit of hand flying. These automation-fetishists I flew with would groan and look at me, when I would just tell them that I was getting ready for my next sim session (about the only excuse that was plausible).
"But you just came back from one and the next one isn't until six months from now!" Yeah, but it's good to be able to hand fly, just in case you need to. Otherwise you would see someone trying to input stuff into the FMS for a last-minute change of plans, head down punching those little buttons when prudence suggested that heads-up looking out might be the way to go sometimes. But for that you do need to be comfortable hand-flying your aircraft, yes...
Generations of young students have been told that "the autopilot can fly the aircraft better than you can," when I never thought that the way to go.
Join Date: Dec 2008
Location: DXB
Posts: 513
Likes: 0
Received 0 Likes
on
0 Posts
Generations of young students have been told that "the autopilot can fly the aircraft better than you can,"
PEI 3721 I am unsure why you would want to disagree with a significant level of blame being attributed to the captain of this flight. As has been done to death in numerous other posts I agree that there were other contributing factors. What has surprised me and others is that the initial investigation seems to place most of the blame elsewhere, on Boeing and ATC in particular.
As an experienced 737 captain I feel a number of serious errors were made by the crew. I do not believe that they were blameless in the outcome or displayed particularly good airmanship. Let me argue my point using your prefered model of airmanship:
Discipline: The crew carried out an unstabilised approach below a level Boeing recommends going around. They failed to complete required actions such as checklists in a timely fashion.
Skill and proficiency: The go-around / stall recovery was flown in such a manner that the aircraft crashed. This is a standard sim exercise and is the first item in the maneuvers section of the QRH. As has been pointed out they could have flown manually at an earlier point and the outcome would have been different.
Knowledge: They seem to have been taken by surprise by the 737's strong pitch up tendency when applying max power. They do not seem to have realised how long it would take the aircraft to decelerate when capturing from above.
Situational awareness: The crew accepted a tight turn onto the ILS and were too slow in recognising the problem. In the recovery they became overloaded and failed to recognise the problem with the throttles.
Judgement: The captain seems to have let the trainee get into a situation where recovery was a high risk situation and indeed proved to be beyond his own ability. He is reported to have been focused on the landing checklist, whilst no one was adequately monitoring the performance of the autoflight system.
Oakape has described many of the underlying issues in his excellent post. We live in an imperfect world. Our industry has become significantly safer due to automation. The flip side is that many pilots are now unable to deal with basic problems due to automation dependance.
These poor guys paid the ultimate price. I hope we can all learn something from what happened.
As an experienced 737 captain I feel a number of serious errors were made by the crew. I do not believe that they were blameless in the outcome or displayed particularly good airmanship. Let me argue my point using your prefered model of airmanship:
Discipline: The crew carried out an unstabilised approach below a level Boeing recommends going around. They failed to complete required actions such as checklists in a timely fashion.
Skill and proficiency: The go-around / stall recovery was flown in such a manner that the aircraft crashed. This is a standard sim exercise and is the first item in the maneuvers section of the QRH. As has been pointed out they could have flown manually at an earlier point and the outcome would have been different.
Knowledge: They seem to have been taken by surprise by the 737's strong pitch up tendency when applying max power. They do not seem to have realised how long it would take the aircraft to decelerate when capturing from above.
Situational awareness: The crew accepted a tight turn onto the ILS and were too slow in recognising the problem. In the recovery they became overloaded and failed to recognise the problem with the throttles.
Judgement: The captain seems to have let the trainee get into a situation where recovery was a high risk situation and indeed proved to be beyond his own ability. He is reported to have been focused on the landing checklist, whilst no one was adequately monitoring the performance of the autoflight system.
Oakape has described many of the underlying issues in his excellent post. We live in an imperfect world. Our industry has become significantly safer due to automation. The flip side is that many pilots are now unable to deal with basic problems due to automation dependance.
These poor guys paid the ultimate price. I hope we can all learn something from what happened.
Join Date: Jul 2002
Location: UK
Posts: 3,093
Likes: 0
Received 0 Likes
on
0 Posts
chuks:
Generations of young students have been told that "the autopilot can fly the aircraft better than you can,"
Generations of young students have been told that "the autopilot can fly the aircraft better than you can,"
Join Date: Jan 2008
Location: Kent
Age: 65
Posts: 216
Likes: 0
Received 0 Likes
on
0 Posts
Is Schiphol part of the problem?
As someone who's usually found in 14a, waiting for my bag of peanuts, I have nothing useful to contribute to a technical analysis of the accident.
However, having followed this affair at a respectful distance, I'm struck by something that never quite makes it to the level of proper discussion. Pilots consistently comment on the nature of Schiphol ATC as being crisp, hyper-efficient, and expecting equal efficiency on the part of the pilot. There's an implication that flight crew are routinely asked to accept instructions demanding a level of skill that perhaps pushes them out of their comfort zone.
Is this the case? If so, does it put pressure on pilots to accept slightly scary, non-standard instructions, in order not to look like wimps? In a macho profession like flying (with apologies to female flight crew), are there likely to be a number of pilots who'll attempt to do things that are really taking them beyond their work-load capacity, for fear of losing face in front of all the other guys?
However, having followed this affair at a respectful distance, I'm struck by something that never quite makes it to the level of proper discussion. Pilots consistently comment on the nature of Schiphol ATC as being crisp, hyper-efficient, and expecting equal efficiency on the part of the pilot. There's an implication that flight crew are routinely asked to accept instructions demanding a level of skill that perhaps pushes them out of their comfort zone.
Is this the case? If so, does it put pressure on pilots to accept slightly scary, non-standard instructions, in order not to look like wimps? In a macho profession like flying (with apologies to female flight crew), are there likely to be a number of pilots who'll attempt to do things that are really taking them beyond their work-load capacity, for fear of losing face in front of all the other guys?
lederhosen;
Understand your POV and agree.
The "wandering into philosophy" is, however, necessary though less necessary (in my opinion) in the THY accident than others.
In other words, this is a recognition of broader factors in terms of antecedents and not a transfer of cause.
Put another way, if we remain stuck in the cockpit and don't examine antecedents (in terms of training, safety culture etc), then change is not possible because, unless there is a special case of capability (Colgan case) or "rogue pilot", the process of the identification of cause may stop at the "symptom" level.
I don't see PE_3721 disagreeing with your view, - a view with which I strongly agree. For reasons given, his post is, in this approach, well worth reading along with the links provided, as is Oakape's excellent and succinct contribution.
PJ2
Understand your POV and agree.
The "wandering into philosophy" is, however, necessary though less necessary (in my opinion) in the THY accident than others.
In other words, this is a recognition of broader factors in terms of antecedents and not a transfer of cause.
Put another way, if we remain stuck in the cockpit and don't examine antecedents (in terms of training, safety culture etc), then change is not possible because, unless there is a special case of capability (Colgan case) or "rogue pilot", the process of the identification of cause may stop at the "symptom" level.
I don't see PE_3721 disagreeing with your view, - a view with which I strongly agree. For reasons given, his post is, in this approach, well worth reading along with the links provided, as is Oakape's excellent and succinct contribution.
PJ2
Join Date: Apr 2003
Location: Germany
Age: 76
Posts: 1,561
Likes: 0
Received 0 Likes
on
0 Posts
I have had people who probably should know better telling me of the superiority of the autopilot for a long time now. When we moved up to an aircraft with an FMS then we were supposed to become a sort of "flying computer programmer," servant to the FMS, itself master of the autopilot. "Airplane not doing what you want it to?" Go head down into the keyboard to reprogram the FMS rather than just kick out the autopilot to hand-fly an ILS, for instance.
For a real-life example of what I am writing about, we would set up for an ILS, inbound to the FAF in NAV mode with APP armed. Well, this wonky African ILS had sidelobes the size of Pamela Anderson's bazongas and our trusty electronic servant would catch one, switch from NAV to APP, turn onto the final approach miles off course but then lose the sidelobe... when our computer programmers would lose the plot! "What is it doing now?" would be the plaint. Umm, well, if you just look out the window you can see the destination airport over there to our right at 2 o'clock, the FAF easily identifed by that big patch of green at 3 o'clock, the only one around, so that perhaps we should just revert to hand-flying for the PF while the PNF gets this stupid thing sorted out... what we used to call "airmanship"?
No, no, no! That's "flying like a bush pilot" (my approach to life, basically since I am one of those for the most part) and there must be an easy way to just type in the FAF ident in the FMS and then line-select DIRECT TO... No, that doesn't seem to be working since we are still on a heading to Antarctica, basically a long way from our point of intended landing now passing our 2:30. Hmm... What mode are we in here? Oh, yeah... with the over-riding logic being that the autopilot could fly better than a human, when I really, really never thought so. Not least, show me an FMS computer that is as powerful as the human brain. Okay, some of our human brains are a bit addled but still...
I am happy to wait for the final report on this one but I could easily imagine a situation where some low-time, inexperienced guy is faced with a fairly complicated aircraft and a rushed approach into a very, very busy airport, when he reverts to what he has been told and just lets the autopilot (and the autothrottle) get on with the job of flying the aircraft.
Not strictly part of the professional's world (thank God) but true enough is that I have encountered quite a few wealthy American owners of complex aircraft who cheerfully admit to being pretty much incapable of hand-flying on instruments so that they count on turning on that magic box of tricks, their autopilot, in IMC and letting it fly the aircraft.
No professional should ever go that far but that is the direction of the trend vector, I think. Less-capable because lower- and lower-time with less and less practical experience young pilots (look at the background of the Colgan crew to see what I mean) are flying more and more capable aircraft under more and more pressure. The systems and the SOPs are taken to be enough to ensure safety when it's pretty clear it should take a bit more than that except that no one wants to pay what that should cost. This is like enjoying motoring in a Ford Pinto, cheap and fun until the crunch comes yet with the level of loss still acceptable.
I have had all-inclusive holiday flights where I literally had never heard of the operator of the Airbus or Boeing I and my family were on, regulated by Turkey or Cyprus or God knows where but, hey, it was cheap!
For a real-life example of what I am writing about, we would set up for an ILS, inbound to the FAF in NAV mode with APP armed. Well, this wonky African ILS had sidelobes the size of Pamela Anderson's bazongas and our trusty electronic servant would catch one, switch from NAV to APP, turn onto the final approach miles off course but then lose the sidelobe... when our computer programmers would lose the plot! "What is it doing now?" would be the plaint. Umm, well, if you just look out the window you can see the destination airport over there to our right at 2 o'clock, the FAF easily identifed by that big patch of green at 3 o'clock, the only one around, so that perhaps we should just revert to hand-flying for the PF while the PNF gets this stupid thing sorted out... what we used to call "airmanship"?
No, no, no! That's "flying like a bush pilot" (my approach to life, basically since I am one of those for the most part) and there must be an easy way to just type in the FAF ident in the FMS and then line-select DIRECT TO... No, that doesn't seem to be working since we are still on a heading to Antarctica, basically a long way from our point of intended landing now passing our 2:30. Hmm... What mode are we in here? Oh, yeah... with the over-riding logic being that the autopilot could fly better than a human, when I really, really never thought so. Not least, show me an FMS computer that is as powerful as the human brain. Okay, some of our human brains are a bit addled but still...
I am happy to wait for the final report on this one but I could easily imagine a situation where some low-time, inexperienced guy is faced with a fairly complicated aircraft and a rushed approach into a very, very busy airport, when he reverts to what he has been told and just lets the autopilot (and the autothrottle) get on with the job of flying the aircraft.
Not strictly part of the professional's world (thank God) but true enough is that I have encountered quite a few wealthy American owners of complex aircraft who cheerfully admit to being pretty much incapable of hand-flying on instruments so that they count on turning on that magic box of tricks, their autopilot, in IMC and letting it fly the aircraft.
No professional should ever go that far but that is the direction of the trend vector, I think. Less-capable because lower- and lower-time with less and less practical experience young pilots (look at the background of the Colgan crew to see what I mean) are flying more and more capable aircraft under more and more pressure. The systems and the SOPs are taken to be enough to ensure safety when it's pretty clear it should take a bit more than that except that no one wants to pay what that should cost. This is like enjoying motoring in a Ford Pinto, cheap and fun until the crunch comes yet with the level of loss still acceptable.
I have had all-inclusive holiday flights where I literally had never heard of the operator of the Airbus or Boeing I and my family were on, regulated by Turkey or Cyprus or God knows where but, hey, it was cheap!
Last edited by chuks; 3rd Feb 2010 at 18:09.
Join Date: Oct 2009
Location: The Isle Dordt
Posts: 290
Likes: 0
Received 0 Likes
on
0 Posts
Pilots consistently comment on the nature of Schiphol ATC as being crisp, hyper-efficient, and expecting equal efficiency on the part of the pilot. There's an implication that flight crew are routinely asked to accept instructions demanding a level of skill that perhaps pushes them out of their comfort zone.
And yes, Turkish was asked whether it would accept an instruction that would require some skill to get to a correct landing. The captain was free to reject it, he could also be vectored behind BarbiesBoyfriend for landing, adding 10 minutes to the duration of the flight. These are the decisions that a captain has to make.
In this case the Swiss cheese conspired against the crew, carefully lining up all holes, throwing an automation failure their way... I am eager to see the final report, to enumerate the issues not mentioned in this thread
Join Date: Jan 2010
Location: London
Posts: 14
Likes: 0
Received 0 Likes
on
0 Posts
Forgive my simplistic approach (no pun intended).
The aircraft crashed because control was lost while carrying out a go-around (at very low speed).
The go-around was made in response to stickshaker activation.
The stickshaker activated as speed approached Vs.
The speed decayed (well) below Vref as the A/T was responding to left radalt (erroneous) data.
Would someone please advise why three pilots failed to monitor what was going on.
G15s
The aircraft crashed because control was lost while carrying out a go-around (at very low speed).
The go-around was made in response to stickshaker activation.
The stickshaker activated as speed approached Vs.
The speed decayed (well) below Vref as the A/T was responding to left radalt (erroneous) data.
Would someone please advise why three pilots failed to monitor what was going on.
G15s
Join Date: Dec 2008
Location: DXB
Posts: 513
Likes: 0
Received 0 Likes
on
0 Posts
They were too busy rushing checklists 
My guess is that neither the ATC nor the crew could guess if the aircraft would intercept loc & glide before or after 6 NM until it actually intercept the loc. This is certainly not the kind of vectoring that contribute to traffic fluidity and flight safety.
Was this flight the only aircraft to accept such tight vectoring? Was it the only one to intercept after the published FAP? What percentage of crew would have performed a go around at 5.5 NM?
Join Date: Mar 2006
Location: Choroni, sometimes
Posts: 1,974
Likes: 0
Received 0 Likes
on
0 Posts
Was this flight the only aircraft to accept such tight vectoring? Was it the only one to intercept after the published FAP? What percentage of crew would have performed a go around at 5.5 NM?
Last edited by hetfield; 3rd Feb 2010 at 20:09.
Join Date: Oct 2009
Location: The Isle Dordt
Posts: 290
Likes: 0
Received 0 Likes
on
0 Posts
My knowledge is mainly from listening to ATC communications, but it seems that many pilots accept the challenge of a "short approach" and make a nice landing. My estimate is that ATC offers several (two dozen?) a day and Turkish was the first one to go seriously wrong. (If the short approach does not stabilize quickly enough, go around...) I expect that the DSB will spend a few paragraphs on the practice in its report.
SFLY, I am ambiguous on the ATC role, because the acceptance of the short approach is a decision of the captain and I know that Amsterdam Approach is not pushing it. OTOH, ATC offers an approach that requires more skill to execute correctly than the standard instrument approach.
To all that missed the <irony> tags around my checklist remark: We'll know more when the final report is published, with a CVR transcript. Until then, we can only guess. I feel that some "rushing" played a role in the accident, but that is just speculation.
SFLY, I am ambiguous on the ATC role, because the acceptance of the short approach is a decision of the captain and I know that Amsterdam Approach is not pushing it. OTOH, ATC offers an approach that requires more skill to execute correctly than the standard instrument approach.
To all that missed the <irony> tags around my checklist remark: We'll know more when the final report is published, with a CVR transcript. Until then, we can only guess. I feel that some "rushing" played a role in the accident, but that is just speculation.
Guest
Posts: n/a
Relative to the "Philosophy", "Economics", and "Pilot Error" addressed by PJ2, I get the distinct feeling that his post may have gone unappreciated.
ATC may be the Ringmaster, but it is not inappropriate for a Lion to roar and say "Negative, Unable". Couldn't hurt in the long run, either, Lord knows a little respect for who does the flying would be nice.
bear
ATC may be the Ringmaster, but it is not inappropriate for a Lion to roar and say "Negative, Unable". Couldn't hurt in the long run, either, Lord knows a little respect for who does the flying would be nice.
bear