"To err is human": differing attitudes to mistakes in EK and Turkish accidents
Thread Starter
Join Date: Jan 2008
Location: Geneva
Posts: 188
Likes: 0
Received 0 Likes
on
0 Posts
"To err is human": differing attitudes to mistakes in EK and Turkish accidents
SLF here. As one who regularly puts his safety in your hands, I come humbly to your forum with observations and questions from the two current threads on the Turkish Airlines accident in Amsterdam, and the Emirates tailstrike and near-accident in Melbourne. (Moderators: I'm posting this in R&N as it relates directly to two threads here, and concerns what I think is an "item that may be of interest to professional pilots".)
Although the respective investigations continue, the preliminary reports and the lively armchair analysis on PPRuNe suggest that both the EK and TK cases involved serious errors by the flight crew. In the Turkish case, the crew failed to notice abnormally decreasing airspeed, resulting in a fatal crash. In the Emirates case the crew apparently used an incorrect aircraft weight in their take-off calculations, very nearly resulting in a fatal crash.
In both cases the mistake was serious enough, in itself, to cause a crash. And in both cases, there was no layer of redundancy to catch the mistake.
But the responses from pilots on the respective threads here have been curiously different. The emerging consensus on the TK accident thread is that the crew made an inexplicable and inexcusable error, all on their own. No systems/management failure, no fatigue issue, nothing. Some put it more bluntly than others, but the prevailing view is perhaps best summed up in this thoughtful comment from bjornhall:
In contrast, the response to the Emirates accident has been more along the lines of "there but for the grace of God go I", with lots of discussion of the possible role of fatigue, management problems, etc. One gets the impression that, in the view of pilots, this is a very different and less serious sort of "mistake" than in the TK case.
In general, as far as I can see, there are three basic types of "mistake":
1. Deliberately ignoring or contravening procedures, cutting corners, etc (e.g. as in the Garuda 737 accident, for which the captain was recently jailed).
2. Mistakes that "anyone" can make (where "anyone" presumably means "any properly qualified and reasonably competent and experienced airline pilot").
3. Mistakes that, although inadvertent, are so egregious or just plain stupid that no qualified pilot should make them ever, in an entire career.
Type 1 is not my concern here, although I know there is another whole debate over the application of criminal sanctions in safety-critical fields.
Type 2 mistakes are, or should be, covered by a system of checks and redundancies, both human and mechanical, so that when (not if) they happen, they are caught and rectified without major consequence.
Type 3 mistakes have no such coverage, since they are to be "trained out" of the human, and are presumably thought to be so unlikely as to require no further precaution or remedy.
It is fairly clear that the "PPRuNe consensus" (if such exists) views the Turkish accident as a type 3 mistake. The consensus is less clear on the EK accident, but many of the comments seem to view it as type 2, and there is a lot of sympathy for the crew.
Why the difference? Is there a difference? And if the EK case really is a type 2 mistake, shouldn't there be redundant systems to catch it?
Although the respective investigations continue, the preliminary reports and the lively armchair analysis on PPRuNe suggest that both the EK and TK cases involved serious errors by the flight crew. In the Turkish case, the crew failed to notice abnormally decreasing airspeed, resulting in a fatal crash. In the Emirates case the crew apparently used an incorrect aircraft weight in their take-off calculations, very nearly resulting in a fatal crash.
In both cases the mistake was serious enough, in itself, to cause a crash. And in both cases, there was no layer of redundancy to catch the mistake.
But the responses from pilots on the respective threads here have been curiously different. The emerging consensus on the TK accident thread is that the crew made an inexplicable and inexcusable error, all on their own. No systems/management failure, no fatigue issue, nothing. Some put it more bluntly than others, but the prevailing view is perhaps best summed up in this thoughtful comment from bjornhall:
Of course we all know that humans are susceptible to errors, the best pilots can make the worst mistakes, the system is only made safe by trapping and mitigating such errors, not by relying on them never occuring, etc etc etc. I doubt anyone contributing here lack that understanding.
But not monitoring airspeed on final is too basic to fall under the "to err is human" heading. The system shall be pilot proof, not idiot proof. Sweeping every display of pilot error under the same "human error" blanket risks making one blind to an industry wide training problem, if such a problem exists.
But not monitoring airspeed on final is too basic to fall under the "to err is human" heading. The system shall be pilot proof, not idiot proof. Sweeping every display of pilot error under the same "human error" blanket risks making one blind to an industry wide training problem, if such a problem exists.
In general, as far as I can see, there are three basic types of "mistake":
1. Deliberately ignoring or contravening procedures, cutting corners, etc (e.g. as in the Garuda 737 accident, for which the captain was recently jailed).
2. Mistakes that "anyone" can make (where "anyone" presumably means "any properly qualified and reasonably competent and experienced airline pilot").
3. Mistakes that, although inadvertent, are so egregious or just plain stupid that no qualified pilot should make them ever, in an entire career.
Type 1 is not my concern here, although I know there is another whole debate over the application of criminal sanctions in safety-critical fields.
Type 2 mistakes are, or should be, covered by a system of checks and redundancies, both human and mechanical, so that when (not if) they happen, they are caught and rectified without major consequence.
Type 3 mistakes have no such coverage, since they are to be "trained out" of the human, and are presumably thought to be so unlikely as to require no further precaution or remedy.
It is fairly clear that the "PPRuNe consensus" (if such exists) views the Turkish accident as a type 3 mistake. The consensus is less clear on the EK accident, but many of the comments seem to view it as type 2, and there is a lot of sympathy for the crew.
Why the difference? Is there a difference? And if the EK case really is a type 2 mistake, shouldn't there be redundant systems to catch it?
I Have Control
Join Date: May 2004
Location: North-West England
Posts: 49
Likes: 0
Received 0 Likes
on
0 Posts
No difference.
Both Type 3 mistakes. IMHO. TK have a very poor safety record, (in the public domain) so bias may well creep in. Mind you, this is not the first T/O screw-up on EK's record. JNB A340 incident springs to mind.
I think your thread is valid, and thought-provoking.
I think your thread is valid, and thought-provoking.
Join Date: Apr 2002
Location: FUBAR
Posts: 3,348
Likes: 0
Received 0 Likes
on
0 Posts
I think your summing up of the "consensus" is a fair representation of how most people see it. Personally, I can imagine a situation arising ,particularly in a training environment, that could lead to one "approaching" the Turkish scenario. What I find harder to justify is taking my eye off the ball long enough for it to develop as catastrophically as it did. Regretably, whichever way you dress it up, it really was one god- all -mighty loss of situational awareness. 
As I understand it, the EK incident was caused fairly simply by getting a number wrong.
I agree with you entirely that there should be a system of checks and balances in place to stop it happening, if there is, and it was circumvented/ignored then this becomes a Type 3 accident as the error becomes a bit less "forgiveable", but it seems the system allowed this error to slip through too easily by far.
Another accident that is difficult to categorise is the ditching off Sardinia of the TunisAir ATR42. In this case a simple system of crosschecks would have shown up the ridiculous discrepancy in the fuel uplift, but either the system didn't call for the check, or it was simply omitted/misunderstood. . . again teetering between Type2 & 3.
I hate to say it here also, for fear of being flamed, but had it been a Lufthansa/Air France/BA rather than Turkish aircraft involved, perhaps slightly more understanding might have been shown.
Unfortunately, whether we like it or even care to admit it, we are all guilty in life to a greater or lesser extent, of just a teensy-weensy bit of "casual racism" from time to time. Alas that is truly imprinted in the nature of human beings ,as any thorough research can demonstrate. It is a primeval instinct, which we we don't consciously seek, & can only at best partially ignore.
Interesting question you pose though
As I understand it, the EK incident was caused fairly simply by getting a number wrong.
I agree with you entirely that there should be a system of checks and balances in place to stop it happening, if there is, and it was circumvented/ignored then this becomes a Type 3 accident as the error becomes a bit less "forgiveable", but it seems the system allowed this error to slip through too easily by far.
Another accident that is difficult to categorise is the ditching off Sardinia of the TunisAir ATR42. In this case a simple system of crosschecks would have shown up the ridiculous discrepancy in the fuel uplift, but either the system didn't call for the check, or it was simply omitted/misunderstood. . . again teetering between Type2 & 3.
I hate to say it here also, for fear of being flamed, but had it been a Lufthansa/Air France/BA rather than Turkish aircraft involved, perhaps slightly more understanding might have been shown.
Unfortunately, whether we like it or even care to admit it, we are all guilty in life to a greater or lesser extent, of just a teensy-weensy bit of "casual racism" from time to time. Alas that is truly imprinted in the nature of human beings ,as any thorough research can demonstrate. It is a primeval instinct, which we we don't consciously seek, & can only at best partially ignore.
Interesting question you pose though
Join Date: Jan 2008
Location: southwest
Age: 78
Posts: 287
Likes: 0
Received 0 Likes
on
0 Posts
Schoolchildren
To anyone trying to be objective, it's saddening to hear Turkish (could be Egyptian, Brazilian.....) pilots unions, national politicians and local press instantly claiming "it wasn't our fault", like kids in the playground.
That attitude is so irritating that, rightly or wrongly, it surely helps turn opinion against the parties they are trying to defend.
That attitude is so irritating that, rightly or wrongly, it surely helps turn opinion against the parties they are trying to defend.
Join Date: Apr 2002
Location: FUBAR
Posts: 3,348
Likes: 0
Received 0 Likes
on
0 Posts
Indeed it does, and "national" characteristics (as that undoubtedly is what inability to accept loss of face may be characterised as is in many cases) is part of the reason that we DO differentiate between accidents depending on the nationality of the parties involved. To do so may be "politically incorrect", but to fail to do so is to be blind to the fact that we are indeed all "different" in ways that are both good and bad. Unfortunately, these days it is seen as a crime to say so, but we don't all react the same, and we shouldn't pretend otherwise.
Join Date: Aug 2006
Location: various hotels
Posts: 7
Likes: 0
Received 0 Likes
on
0 Posts
A 2 person crew aircraft has one pilot flying (PF) or monitoring and responsible for what the autopilot (to include autothrottle) is doing, and the other pilot monitoring (PM) the actions the PF and aircraft. An approach would typically have have the PM calling out airspeed (and sink rate) at set points based on a Vref, typically at 500' AGl and again at 100' AGL. It is, however, incumbent on the PM to call out any serious deviation from the planned or "targeted" approach speed or a higher or lower than normal sink rate at any time (at my company Vref + 5 is the minimum target airspeed allowed on the approach) along with deviations from glideslope(path) and localizer or course. Going below Vref with a decreasing trend would be very serious and would certainly generate a call. This is really very basic airmanship and speaks to a lack of basic situational awarness on the part of the crew.
The Takeoff Weight entry is more of a problem. Suffice it to say there are a number of cross checks made in an effort to trap a mistake, but if the payload weight is wrong everything else will be wrong.
The Takeoff Weight entry is more of a problem. Suffice it to say there are a number of cross checks made in an effort to trap a mistake, but if the payload weight is wrong everything else will be wrong.
Join Date: Aug 2006
Location: Sri Lanka
Posts: 41
Likes: 0
Received 0 Likes
on
0 Posts
According to your definition both are type 2 mistakes in my opinion, as both should have been caught by the system or the other crew-member. The EK crew were just lucky they got away with it, the TK guys were not. Having said this, we all make type 2 mistakes once in a while, there is nobody who has never made one. There are of course a lot of contributing factors leading to a type 2 mistake, but it should never result in a serious incident or accident. When this happens the system or the other crew-member has failed. Safe flying to all of you!
Join Date: Mar 2000
Location: Arizona USA
Posts: 8,571
Likes: 0
Received 0 Likes
on
0 Posts
According to your definition both are type 2 mistakes in my opinion, as both should have been caught by the system or the other crew-member. The EK crew were just lucky they got away with it, the TK guys were not. Having said this, we all make type 2 mistakes once in a while, there is nobody who has never made one. There are of course a lot of contributing factors leading to a type 2 mistake, but it should never result in a serious incident or accident. When this happens the system or the other crew-member has failed.
There is normally quite a lot of finger pointing going on at PPRuNe...mostly toward non-European operators which adds little to the specific safety discussion.
Ain't likely to change, either.
Join Date: Dec 2008
Location: PDX
Posts: 72
Likes: 0
Received 0 Likes
on
0 Posts
(Self-loathing freight here) TK's airspeed is something that pilots have always monitored, so a lapse there would be Type 3 while perhaps EK's data entry would at one time have been performed by an engineer in a three-crew cockpit (so, Type 2?).
I do not know if a navigation error should be considered Type 2 because that is something that the navigator used to do in a four-crew cockpit. . . .
Edit: transposed 3 and 2--now fixed. I guess that's why my chair doesn't have rudder pedals.
I do not know if a navigation error should be considered Type 2 because that is something that the navigator used to do in a four-crew cockpit. . . .
Edit: transposed 3 and 2--now fixed. I guess that's why my chair doesn't have rudder pedals.
Last edited by fotoguzzi; 4th May 2009 at 17:50.
Join Date: May 2005
Location: dubai
Posts: 375
Likes: 0
Received 0 Likes
on
0 Posts
''Type 3 mistakes have no such coverage, since they are to be "trained out" of the human, and are presumably thought to be so unlikely as to require no further precaution or remedy.''
Can't speak for TK but the EK accident was definitely Type 3
1- This is the third instance of such an input error in the past 7 years at EK alone (there have probably been many more input errors that were never captured because they did not cause a problem).
2- Input errors into a computer are common, although less so for takeoff data.
3- There may have been contributing factors, such as cockpit interruptions, fatigue or MFF flying, which counter the ''trained out'' possibility. 'Cockpit interruptions' has already been pointed out by the LOSA evaluations, yet no further precautions or remedies were instituted and even now, very little has been done to address this issue. Fatigue is an ongoing problem at EK, one that is hard to quantify especially if there is no intention on the part of management to even acknowledge the possibility. MFF is debatable for sure but it is also a possibility. Add the three together, and who knows the result!
Can't speak for TK but the EK accident was definitely Type 3
1- This is the third instance of such an input error in the past 7 years at EK alone (there have probably been many more input errors that were never captured because they did not cause a problem).
2- Input errors into a computer are common, although less so for takeoff data.
3- There may have been contributing factors, such as cockpit interruptions, fatigue or MFF flying, which counter the ''trained out'' possibility. 'Cockpit interruptions' has already been pointed out by the LOSA evaluations, yet no further precautions or remedies were instituted and even now, very little has been done to address this issue. Fatigue is an ongoing problem at EK, one that is hard to quantify especially if there is no intention on the part of management to even acknowledge the possibility. MFF is debatable for sure but it is also a possibility. Add the three together, and who knows the result!
Join Date: Dec 2007
Location: france
Age: 75
Posts: 74
Likes: 0
Received 0 Likes
on
0 Posts
In the TK accident, the pilots made mistakes, that's right. But we cannot ignore that the aircraft was not really fit to fly :
"The data show instances of left radio altimeter malfunctions on some of the nine previous flights. In the recorded cases, the autothrottle also entered the retard mode above the intended flare altitude, and the thrust levers moved to idle, because of a malfunction of the left radio altimeter on two of the nine flights. The data of these flights are being investigated."
"The data show instances of left radio altimeter malfunctions on some of the nine previous flights. In the recorded cases, the autothrottle also entered the retard mode above the intended flare altitude, and the thrust levers moved to idle, because of a malfunction of the left radio altimeter on two of the nine flights. The data of these flights are being investigated."
Join Date: Apr 2002
Location: FUBAR
Posts: 3,348
Likes: 0
Received 0 Likes
on
0 Posts
Don't really think the A/T selecting idle thrust when you don't particularly want it means "the aircraft was not really fit to fly"
A bit of a pain in the ass perhaps, but finally, the A/P & A/T are merely there to do your job for you. If they don't do it as you wish (& it is incumbent upon you to check they do) you switch them off & DIY.
Really truly lets drop this cr@p.
Chances are the reason previous crews didn't write up the radalt prob in the tech-log wasn't to avoid making work for the ginger-beers, but more likely because they simply didn't notice.
On a Cat1 approach to a manual landing it is a total non event, so let's quit trying to blame the A/T or RADAlt for the fact the crew (for whatever reason still to be established ) were not flying, nor monitoring the performance of the A/P, A/T and more pertinently , the A/C.
END OF STORY.
A bit of a pain in the ass perhaps, but finally, the A/P & A/T are merely there to do your job for you. If they don't do it as you wish (& it is incumbent upon you to check they do) you switch them off & DIY.
Really truly lets drop this cr@p.
Chances are the reason previous crews didn't write up the radalt prob in the tech-log wasn't to avoid making work for the ginger-beers, but more likely because they simply didn't notice.
On a Cat1 approach to a manual landing it is a total non event, so let's quit trying to blame the A/T or RADAlt for the fact the crew (for whatever reason still to be established ) were not flying, nor monitoring the performance of the A/P, A/T and more pertinently , the A/C.
END OF STORY.
Join Date: Dec 2000
Location: Somerset
Posts: 78
Likes: 0
Received 0 Likes
on
0 Posts
Some of the non-pilots who have been reading about this accident seem to be having difficulty in working out just why the pilots amongst us are so horrified by what appears to have happened.
It is the very basic need to "watch the shop", for at least one of the pilots to be keeping an eye on the instrument panel, that very quickly becomes second nature.
I can only describe what went on, in non-aviation terms, as having increased ones speed when driving to say 70 mph - and then closing ones eyes! And waiting for the bang.
If someone had told me that an accident like this would one day take place, well, I just wouldn't have believed them!
It is the very basic need to "watch the shop", for at least one of the pilots to be keeping an eye on the instrument panel, that very quickly becomes second nature.
I can only describe what went on, in non-aviation terms, as having increased ones speed when driving to say 70 mph - and then closing ones eyes! And waiting for the bang.
If someone had told me that an accident like this would one day take place, well, I just wouldn't have believed them!
Join Date: Nov 2000
Location: Dre's mum's house
Posts: 1,432
Likes: 0
Received 0 Likes
on
0 Posts
Automation, in modern glass cockpit aircraft, is dutifully dumb: tell the airplane to fly to Point X with a mountain between the airplane and Point X and it will, very accurately, impact with said mountain (Cali).
Sit back and watch the thrust levers come back to idle thrust and, with the autopilot engaged ( and no Alpha Floor protection), it will fly a textbook full stall manoeuvre.
We are in danger of losing sight of the big picture: the "Playstation" generation are now sitting in the right seat, and the left seat, of modern commercial aircraft: "what's it doing now" is becoming more common.
Aviate, navigate, communicate...stick to the basics.
A 737 / A320 is a glorified C150 and you can always go back to basics.
Sit back and watch the thrust levers come back to idle thrust and, with the autopilot engaged ( and no Alpha Floor protection), it will fly a textbook full stall manoeuvre.
We are in danger of losing sight of the big picture: the "Playstation" generation are now sitting in the right seat, and the left seat, of modern commercial aircraft: "what's it doing now" is becoming more common.
Aviate, navigate, communicate...stick to the basics.
A 737 / A320 is a glorified C150 and you can always go back to basics.
Gibon2;
For an SLF contribution to this discussion, you get an 'A'. Nice work.
Typing is useful to begin with, but very quickly the process becomes a human one and is therefore subject to enormous subtleties which defy categorzation. But I suspect you already know and appreciate that.
4Greens;
Along with Jim's book there are others equally if not more important in terms of human error, which I highly recommend anyone wishing to comprehend why this accident is so horrifying to professional aircrews and so misunderstood by non-pilots including engineers. Hopefully this thread will run it's course soon.
Nothing excuses what this crew did not do - fly the airplane. All the rest are unimportant details raised by those who won't/can't admit this fundamental fact.
Further, there is no understanding in this accident that is useful in the addressing of organizational, systemic or human factors accidents.
For an understanding of those kinds of accidents, (which aren't really accidents at all but happenings-with-precursors), the following references will be of interest:
Some titles are:
Normal Accidents: Living with High Risk Technologies,
Charles Perrow
The Field Guide to Understanding Human Factors
and
Just Culture,
- both by Sidney Dekker
Why-Because Analysis, (web-based)
Ladkin
Accident Precursor Analysis and Management: Reducing Technological Risk Through Diligence,
National Academy of Engineering
The Challenger Launch Decision:Risky Technology, Culture and Deviance at NASA,
Diane Vaughan
Organization at the Limit: Lessons From the Columbia Disaster
Starbuck & Farjoun
Beyond Aviation Human Factors,
James Reason
The Real Slim Shady;
And, perhaps in increasing frequency given the accidents we have seen thus far this year, "I think I'll just let it do what it wants...".
The training issues alone are substantial but checking issues as well as flight data programs are crucial. Without those tools, an airline is flying blind and every incident or accident will be a shock and a surprise.
For an SLF contribution to this discussion, you get an 'A'. Nice work.
Typing is useful to begin with, but very quickly the process becomes a human one and is therefore subject to enormous subtleties which defy categorzation. But I suspect you already know and appreciate that.
4Greens;
Along with Jim's book there are others equally if not more important in terms of human error, which I highly recommend anyone wishing to comprehend why this accident is so horrifying to professional aircrews and so misunderstood by non-pilots including engineers. Hopefully this thread will run it's course soon.
Nothing excuses what this crew did not do - fly the airplane. All the rest are unimportant details raised by those who won't/can't admit this fundamental fact.
Further, there is no understanding in this accident that is useful in the addressing of organizational, systemic or human factors accidents.
For an understanding of those kinds of accidents, (which aren't really accidents at all but happenings-with-precursors), the following references will be of interest:
Some titles are:
Normal Accidents: Living with High Risk Technologies,
Charles Perrow
The Field Guide to Understanding Human Factors
and
Just Culture,
- both by Sidney Dekker
Why-Because Analysis, (web-based)
Ladkin
Accident Precursor Analysis and Management: Reducing Technological Risk Through Diligence,
National Academy of Engineering
The Challenger Launch Decision:Risky Technology, Culture and Deviance at NASA,
Diane Vaughan
Organization at the Limit: Lessons From the Columbia Disaster
Starbuck & Farjoun
Beyond Aviation Human Factors,
James Reason
The Real Slim Shady;
"what's it doing now" is becoming more common.
The training issues alone are substantial but checking issues as well as flight data programs are crucial. Without those tools, an airline is flying blind and every incident or accident will be a shock and a surprise.
Last edited by PJ2; 5th May 2009 at 01:05.
Join Date: Jan 2008
Location: uk
Posts: 857
Likes: 0
Received 0 Likes
on
0 Posts
More concerning to me is that:
- this aircraft is built with nice big real moving controls to make it (allegedly) a lot more obvious to the pilot what the automatics are doing, compared to a certain other type of aircraft
- and (as you noted) the aircraft warned its handlers, using its big moving controls, of this suicidal tendency tendency on multiple previous flights
- and...
If this aircraft actually was, "not really fit to fly", then why did nine previous crews let it go back up ? Whichever way you look at it it looks like a huge chunk of human error turning what should have been a minor technical problem into a fatal crash.
Per Ardua ad Astraeus
Join Date: Mar 2000
Location: UK
Posts: 18,579
Likes: 0
Received 0 Likes
on
0 Posts
If there any mods around, can I make a plea that we do not immerse ourselves yet again in a relentless repetition of '"it was u/s", "why was it not snagged?" (soon we'll be asked about autolands here). PLEASE can you transfer any such posts to the already 'expired' Schipol crash thread?
PJ as usual says it for me -
"And, perhaps in increasing frequency given the accidents we have seen thus far this year, "I think I'll just let it do what it wants...".
The training issues alone are substantial but checking issues as well as flight data programs are crucial. Without those tools, an airline is flying blind and every incident or accident will be a shock and a surprise."
There are substantial lessons we all need to learn from recent accidents where 'automatics' have played a deadly part. Those who dismiss the Schipol crash as 'irrelevant' preach a highly dangerous, blaze and over-confident doctrine which serves no-one's interests but their own. In all these accidents, supposedly competent and experienced crews have allowed the unthinkable to happen. These events appear to be becoming less isolated than we had assumed they were. This is the major issue we need to address without delay.
Whether we have a type 1, 2 or 3 really is only of passing interest. This is 'Human Factors' in the raw.
). PLEASE can you transfer any such posts to the already 'expired' Schipol crash thread? PJ as usual says it for me -
"And, perhaps in increasing frequency given the accidents we have seen thus far this year, "I think I'll just let it do what it wants...".
The training issues alone are substantial but checking issues as well as flight data programs are crucial. Without those tools, an airline is flying blind and every incident or accident will be a shock and a surprise."
There are substantial lessons we all need to learn from recent accidents where 'automatics' have played a deadly part. Those who dismiss the Schipol crash as 'irrelevant' preach a highly dangerous, blaze and over-confident doctrine which serves no-one's interests but their own. In all these accidents, supposedly competent and experienced crews have allowed the unthinkable to happen. These events appear to be becoming less isolated than we had assumed they were. This is the major issue we need to address without delay.
Whether we have a type 1, 2 or 3 really is only of passing interest. This is 'Human Factors' in the raw.
Join Date: Mar 2007
Location: south east UK
Posts: 375
Likes: 0
Received 0 Likes
on
0 Posts
good thread. I agree with BOAC - lets not pollute this interesting discussion with pointlss comments about Rad alts and autolands.
well said - however thats why i think that the type 1, 2,3 issue on the opening post IS relevent. I know the EK error got through the system but in general, as pointed out, there are checks and balances to trap the type 1 and 2 errors. As demonstrate by the horror at the stunning level of 'incompetance' that cause the turkish crash, the type 3 error has no error traps on an operational basis. It is assumed that these errors are prevented by the training and checking regime. Quite why that appears to be no longer the case I don't know. Although to be fair, and risk a politically incorrect flaming, it does still appear that decent operators from enlightened societies are still preventing and trapping these sort of errors. And its important that we understand why, so that we don't go down the same road.
The reason why we tend to point fingers more at places like Turkey (and as pointed out - Egyption, Brazilian are another couple that spring immediately to mind) is not because we are all superior western sky gods, it is because a society that instantly mobilises and has government and press output such as we've seen from turkey, almost by definition cannot be 'safe' from a human factor point of view.
After all how can Turkish possibly learn, or train out whatever behaviour caused this accident if they refuse to acknowledge that it exists.
Couple this attitude with the authoritarian / subservient issue in many cultures whereby an FO dare not question the mighty captain then you have a real safety issue.
Why does this appear to happening more and more? who knows, perhaps its that the recent aviation expansion has thrust some operators and regulatory regimes into top divisions - where they don't really deserve to belong. it doesn't matter how new the aircraft are, how shiney the paint and what club (star alliance in this case) you belong to. If your regulator, and accident investigator thinks that its job is about face saving and shifting blame then you will never, ever get the human domain correct to trap those type 3 errors.
Originally Posted by BOAC
There are substantial lessons we all need to learn from recent accidents where 'automatics' have played a deadly part. Those who dismiss the Schipol crash as 'irrelevant' preach a highly dangerous, blaze and over-confident doctrine which serves no-one's interests but their own. In all these accidents, supposedly competent and experienced crews have allowed the unthinkable to happen. These events appear to be becoming less isolated than we had assumed they were. This is the major issue we need to address without delay.
The reason why we tend to point fingers more at places like Turkey (and as pointed out - Egyption, Brazilian are another couple that spring immediately to mind) is not because we are all superior western sky gods, it is because a society that instantly mobilises and has government and press output such as we've seen from turkey, almost by definition cannot be 'safe' from a human factor point of view.
After all how can Turkish possibly learn, or train out whatever behaviour caused this accident if they refuse to acknowledge that it exists.
Couple this attitude with the authoritarian / subservient issue in many cultures whereby an FO dare not question the mighty captain then you have a real safety issue.
Why does this appear to happening more and more? who knows, perhaps its that the recent aviation expansion has thrust some operators and regulatory regimes into top divisions - where they don't really deserve to belong. it doesn't matter how new the aircraft are, how shiney the paint and what club (star alliance in this case) you belong to. If your regulator, and accident investigator thinks that its job is about face saving and shifting blame then you will never, ever get the human domain correct to trap those type 3 errors.
Thread Starter
Join Date: Jan 2008
Location: Geneva
Posts: 188
Likes: 0
Received 0 Likes
on
0 Posts
Thanks all for the interesting comments so far. If I could just second BOAC's request to avoid rehashing here arguments that have been thoroughly played out in the respective accident threads. My purpose in starting this thread was to look at the different attitudes to the actions of the respective crews in the two accidents, and to explore what this might show about underlying beliefs and assumptions about human factors and safety.
Anyway, so far we have:
So much for the PPRuNe consensus! But still, it seems that many view the two errors as being of the same type. As PJ2 says, my types are at best approximations and at worst completely arbitrary. But this kind of rough categorisation can perhaps shine some light on the difficult cases at the margins - which are typically the ones that result in accidents. As captplaystation puts it:
For example, type 2 errors that are insufficiently covered by checks and balances, or type 3 errors that are easier to make and more common than is generally recognised.
As SLF, I am happy to accept the tiny risks of catastrophic mechanical or structural failure, extraordinarily unlucky combinations of multiple factors (e.g. crew forgets flaps AND flap warning fails), freak weather, terrorist attacks, midair collisions, etc. What worries me more is that I may be a typo away from a fiery death, or that the crew may just "forget" to fly the plane.
Worse than this, if I have understood correctly, is that there's no agreement on what to do about the problem. PJ2 says of the TK crash:
So what to do? Are you saying these kinds of risks cannot be reduced?
Anyway, so far we have:
Both Type 3 mistakes. IMHO.
According to your definition both are type 2 mistakes in my opinion
teetering between Type 2 & 3
As SLF, I am happy to accept the tiny risks of catastrophic mechanical or structural failure, extraordinarily unlucky combinations of multiple factors (e.g. crew forgets flaps AND flap warning fails), freak weather, terrorist attacks, midair collisions, etc. What worries me more is that I may be a typo away from a fiery death, or that the crew may just "forget" to fly the plane.
Worse than this, if I have understood correctly, is that there's no agreement on what to do about the problem. PJ2 says of the TK crash:
Further, there is no understanding in this accident that is useful in the addressing of organizational, systemic or human factors accidents.