Latest research on Automation Dependency - Regulator please note..
Join Date: Feb 2005
Location: UK
Age: 85
Posts: 697
Likes: 0
Received 0 Likes
on
0 Posts
We have reached the stage where currently the majority of airline pilots are those who have proceeded through the art of flying following a passion and perhaps starting with SE aircraft, in other words they are ‘pilots’ in the way we understand the word.
We might have to accept that the operation of commercial aircraft in the future will best be serviced by ‘aircraft operators’, men or women who are skilled in the operation of the systems that fly aircraft. There is little doubt that the systems can fly an aircraft safely and handle most situations that will occur in flight, in fact many military aircraft can only be flown by their computers under certain circumstances. It may well be that the people operating the aircraft will take control of the systems in the case of an unusual situation rather than take control of the aircraft - flying commercially by holding the joystick, as we know it – may be a thing of the past.
I can certainly see a scenario in the not too distant future where freight is carried by pilotless aircraft.
It’s a bit like the camera, sure very few people understand f numbers but the digital cameras do it all for them and, to be honest, can take pictures as well balanced as where speed and exposure had to be determined by the photographer. Indeed, how many people who use complex computer programs understand basic programming, no they are skilled at using the program and lack nothing by not being able to program. That doesn’t in any way prevent those who want to use film and manual cameras from doing so and enjoying the process. (I can still develop film and I can program so I know how frustrating it can be to find that these skills are no longer necessary to get good results)
So aircraft will crash under certain circumstances but who is to say that any different result would have been gained by the aircraft being operated by a ‘proper’ flier as opposed to an ‘operator’.
I am not advocating this and just don’t ask me to fly as a passenger on one, but I think it will happen - I am sure that it is only passenger attitude that is holding this back.
We might have to accept that the operation of commercial aircraft in the future will best be serviced by ‘aircraft operators’, men or women who are skilled in the operation of the systems that fly aircraft. There is little doubt that the systems can fly an aircraft safely and handle most situations that will occur in flight, in fact many military aircraft can only be flown by their computers under certain circumstances. It may well be that the people operating the aircraft will take control of the systems in the case of an unusual situation rather than take control of the aircraft - flying commercially by holding the joystick, as we know it – may be a thing of the past.
I can certainly see a scenario in the not too distant future where freight is carried by pilotless aircraft.
It’s a bit like the camera, sure very few people understand f numbers but the digital cameras do it all for them and, to be honest, can take pictures as well balanced as where speed and exposure had to be determined by the photographer. Indeed, how many people who use complex computer programs understand basic programming, no they are skilled at using the program and lack nothing by not being able to program. That doesn’t in any way prevent those who want to use film and manual cameras from doing so and enjoying the process. (I can still develop film and I can program so I know how frustrating it can be to find that these skills are no longer necessary to get good results)
So aircraft will crash under certain circumstances but who is to say that any different result would have been gained by the aircraft being operated by a ‘proper’ flier as opposed to an ‘operator’.
I am not advocating this and just don’t ask me to fly as a passenger on one, but I think it will happen - I am sure that it is only passenger attitude that is holding this back.
BOAC;
Well I think your point is very well taken, and I hesitated when I wrote that because I know that far from dumbing this generation down, I believe it has been shown to increase hand-eye work and certain cognitive skills in logical thinking, insight into software responses, etc. I have never tried/done MS Flight Sim but I understand it is a very good teacher for some things. You just don't get that adrenaline punch and the blood rush to the head after knowing you've just about killed yourself and perhaps a lot of others plus written off a beautiful airplane and put your company at risk in doing so...
I think what was on my mind was "the whole enchilada..." - those skills plus the ability to think and build and aggressively use the crap detector, and I know you see that.
The present privileging of certain skills, (those fostered by gaming, etc, - and note!...no judgement here in re gamers), over others, (traditional Wright-Bros etc), is not "progress" as we think of the phenomenon, it "is what it is" without any innate value attached or necessity to support/legitimate its existence or power over our thinking. Someone created it, (many here may remember "ET"), and some complained but we all got used to it - it could have been different.
It is a response to a particular and specific technological culture (out of many possible cultures we could have had) and not one in which what has unfolded is in any way a necessary evolution of design. In other words, nothing mandates how "automation" has been done; it is one way, and it could have been different.
We could have had the reverse kind of "automation, (and I have discussed this and it was considered as long ago as the early '80's according to a paper I have been sent written by an airline pilot so this thinking isn't new or novel), - iow, the other way around - pilots fly the airplane manually in short-term, to remain cognitively-engaged and through manipulations, physically situationally-aware, and for long-term flight a bread & butter autoflight/FMS system is employed while the "real automation" is latent, quietly monitoring "normal". "Normal" is the categorized gathering of all flight regimes (and perhaps even ground ops), which has been defined by the analysis of thousands upon thousands of flights, not necessarily on the same type of aircraft, under all circumstances out of and into all kinds of airports under all manner of approaches - we can get as sophisticated as we wish.
The analysis would define and "algorithm-ize, (apologies to software people), the data to create a detailed set of defined boundaries of flight and possibly ground ops, (again, not type specific - that would be a separate set of "interventions"), of "what human pilots do to maintain 'normal' " under all these circumstances.
When the boundaries of normal are approached, a response which mimics "heightened awareness" begins and at some point, "Abnormal" is defined from a set of disparate circumstances including all the obvious ones we are quite familiar with and know occur, (TAWS-EGPWS/TCAS/Stall/Ground Collision/Wrong Runway, etc) but could include first sensing of a non-stable approach as even some rudimentary FOQA software already do, or signs of loss of control, (Tripoli, Bahrain) and begin resolution not by taking over but by visual and audio warnings which themselves would be modelled after what humans would do/say.
I recall the first think someone did when trying to get the attention of the captain who was fixated on getting in and in a very unstable approach was to call the captain's name to break the fixation and then called for a go-around. Now we can imagine all kinds of humourous situations unfolding to audibly get someone's attention, but it worked. There are others. In fact, here's an idea... If unsuccessful and the boundaries of safe flight are nearby, the system could interevene as does the Boeing 777 and, though more aggressively, the AB series.
I am well aware of the large holes in all this and how far we have come in the present mode of privileging software over pilots while the human is relegated to the role of Monitor. A lot of this has to do with cost and the industry's desire to "get rid of expensive pilots and replace them with cheap software and black boxes", (that process actually began with the Boeing 767, which was built orignally for 3 pilots but came with seats for 2...I remember the shock very clearly when I first heard it). Because the profession of Pilot has been cheapened in so many ways, the social/psychological/philosophical "web" of all the connections which make the profession work well have been shaken and haven't stopped quivering yet. I don't like the present trend but here we are, and a number of "competency accidents" later.
These ideas aren't new, but when they were thought of in the early 80's they weren't doable. Today they are.
Sir George, "I believe one can demonstrate airmanship through the automatics not in spite of them." - Precisely.
Bear;
On autopilots vamoosing, yes, it is a surprise for sure and the present priorities dull, not sharpen response. But that is the very nature of flight...years of boredom, etc., and when it is truly serious, denial takes its place in the first few moments just as the higher cognitive skills may. Like hijackings and other such interventions, the industry has rightly chosen to resolve most of those problems "on the ground"...here, through design, redundancy, and so on.
Autopilots only vamoose only because they are engaged in the first place. They are a supplement to manual flight, not a substitute though both reliability and airline policy breed complacency.
Perhaps it is possible, perhaps not, but there is no autopilot system in existence which will handle serious jet upsets. At some point, the autoflight is going to disconnect and hand the airplane back to the pilot. And in fact all autopilots vamoose given the opportunity.
That said, the Air Canada A319 upset near Calgary is instructive. If I recall, while the autopilot disconnected, the report states that if the pilots had left the controls alone (use of rudder was heavy, requiring a change of the vertical stab if I recall), the airplane would have righted itself using the remaining flight-control laws - going from memory here and haven't looked it up.
Because most mechanical/technical/performance problems have been solved on the ground by the designers, when they do come along they are indeed a shock. I flew 35 years without a single engine failure, had one tire-rim failure on a DC9, one cabin compresser explode on a DC8, and only one serious hydraulic failure on an A330.
All those years and I can remember all the failures because they were so rare. The rest of the stuff that happened was just ordinary stuff that pilots handle all the time and no one ever knows except the incident reporting system and FOQA. I think that is most airline pilots' experience.
The "58 ECAMs" drill is "normal" for such a serious failure. There are a lot of systems to secure, landing data to re-calculate and communications to look after, (ATC, Flight Attendants, Company) and it all takes time. The Emergency Electrical Configuration or a 2-system hydraulic failure will take 45 minutes in the sim. There is no rush to get the airplane on the ground whether a 777 or 330, not unless there is uncontrollable smoke or known fire or known structural damage.
Given what the crew (and we, through the photos) could see, I wouldn't judge this failure as serious structural one in terms of a wing or engine breaking away mandating an immediate landing, but that's just me. I haven't been trained on the B777 so I don't know if the procedures are similar for such emergencies - perhaps someone here can talk about it.
funfly;
I don't think so.
The reason why I think this is not because it is not doable. It is doable right now. The Airbus is an entire digital platform and can be "remoted".
The problem is again, cognitive, not technical or social, (fear of getting on transportation with no operator). An airplane is not a streetcar or an elevator. The multitude of highly-complex decision-making which cannot yet be mimicked, is the hurdle. We may ask ourselves about the present A380 incident and try to come to terms with the design requirements which would have assessed the damage, created a course of action and carried it out, all without human intervention. It is always risky to say such stuff in one's own time, but I don't think pilotless flight is possible in the sense that it would be the routine, commercial way the world's airliners would be operated.
PJ2
Does MS FltSim and gaming actually provide skills with which to address "computerized flight"? Clearly I would suggest that such a notion is foolish. - I disagree - from what I have seen, in terms PURELY of system operation, they provide excellent skills. You did specify "computerized flight". So, back to the first question?
I think what was on my mind was "the whole enchilada..." - those skills plus the ability to think and build and aggressively use the crap detector, and I know you see that.
The present privileging of certain skills, (those fostered by gaming, etc, - and note!...no judgement here in re gamers), over others, (traditional Wright-Bros etc), is not "progress" as we think of the phenomenon, it "is what it is" without any innate value attached or necessity to support/legitimate its existence or power over our thinking. Someone created it, (many here may remember "ET"), and some complained but we all got used to it - it could have been different.
It is a response to a particular and specific technological culture (out of many possible cultures we could have had) and not one in which what has unfolded is in any way a necessary evolution of design. In other words, nothing mandates how "automation" has been done; it is one way, and it could have been different.
We could have had the reverse kind of "automation, (and I have discussed this and it was considered as long ago as the early '80's according to a paper I have been sent written by an airline pilot so this thinking isn't new or novel), - iow, the other way around - pilots fly the airplane manually in short-term, to remain cognitively-engaged and through manipulations, physically situationally-aware, and for long-term flight a bread & butter autoflight/FMS system is employed while the "real automation" is latent, quietly monitoring "normal". "Normal" is the categorized gathering of all flight regimes (and perhaps even ground ops), which has been defined by the analysis of thousands upon thousands of flights, not necessarily on the same type of aircraft, under all circumstances out of and into all kinds of airports under all manner of approaches - we can get as sophisticated as we wish.
The analysis would define and "algorithm-ize, (apologies to software people), the data to create a detailed set of defined boundaries of flight and possibly ground ops, (again, not type specific - that would be a separate set of "interventions"), of "what human pilots do to maintain 'normal' " under all these circumstances.
When the boundaries of normal are approached, a response which mimics "heightened awareness" begins and at some point, "Abnormal" is defined from a set of disparate circumstances including all the obvious ones we are quite familiar with and know occur, (TAWS-EGPWS/TCAS/Stall/Ground Collision/Wrong Runway, etc) but could include first sensing of a non-stable approach as even some rudimentary FOQA software already do, or signs of loss of control, (Tripoli, Bahrain) and begin resolution not by taking over but by visual and audio warnings which themselves would be modelled after what humans would do/say.
I recall the first think someone did when trying to get the attention of the captain who was fixated on getting in and in a very unstable approach was to call the captain's name to break the fixation and then called for a go-around. Now we can imagine all kinds of humourous situations unfolding to audibly get someone's attention, but it worked. There are others. In fact, here's an idea... If unsuccessful and the boundaries of safe flight are nearby, the system could interevene as does the Boeing 777 and, though more aggressively, the AB series.
I am well aware of the large holes in all this and how far we have come in the present mode of privileging software over pilots while the human is relegated to the role of Monitor. A lot of this has to do with cost and the industry's desire to "get rid of expensive pilots and replace them with cheap software and black boxes", (that process actually began with the Boeing 767, which was built orignally for 3 pilots but came with seats for 2...I remember the shock very clearly when I first heard it). Because the profession of Pilot has been cheapened in so many ways, the social/psychological/philosophical "web" of all the connections which make the profession work well have been shaken and haven't stopped quivering yet. I don't like the present trend but here we are, and a number of "competency accidents" later.
These ideas aren't new, but when they were thought of in the early 80's they weren't doable. Today they are.
Sir George, "I believe one can demonstrate airmanship through the automatics not in spite of them." - Precisely.
Bear;
On autopilots vamoosing, yes, it is a surprise for sure and the present priorities dull, not sharpen response. But that is the very nature of flight...years of boredom, etc., and when it is truly serious, denial takes its place in the first few moments just as the higher cognitive skills may. Like hijackings and other such interventions, the industry has rightly chosen to resolve most of those problems "on the ground"...here, through design, redundancy, and so on.
Autopilots only vamoose only because they are engaged in the first place. They are a supplement to manual flight, not a substitute though both reliability and airline policy breed complacency.
Perhaps it is possible, perhaps not, but there is no autopilot system in existence which will handle serious jet upsets. At some point, the autoflight is going to disconnect and hand the airplane back to the pilot. And in fact all autopilots vamoose given the opportunity.
That said, the Air Canada A319 upset near Calgary is instructive. If I recall, while the autopilot disconnected, the report states that if the pilots had left the controls alone (use of rudder was heavy, requiring a change of the vertical stab if I recall), the airplane would have righted itself using the remaining flight-control laws - going from memory here and haven't looked it up.
Because most mechanical/technical/performance problems have been solved on the ground by the designers, when they do come along they are indeed a shock. I flew 35 years without a single engine failure, had one tire-rim failure on a DC9, one cabin compresser explode on a DC8, and only one serious hydraulic failure on an A330.
All those years and I can remember all the failures because they were so rare. The rest of the stuff that happened was just ordinary stuff that pilots handle all the time and no one ever knows except the incident reporting system and FOQA. I think that is most airline pilots' experience.
The "58 ECAMs" drill is "normal" for such a serious failure. There are a lot of systems to secure, landing data to re-calculate and communications to look after, (ATC, Flight Attendants, Company) and it all takes time. The Emergency Electrical Configuration or a 2-system hydraulic failure will take 45 minutes in the sim. There is no rush to get the airplane on the ground whether a 777 or 330, not unless there is uncontrollable smoke or known fire or known structural damage.
Given what the crew (and we, through the photos) could see, I wouldn't judge this failure as serious structural one in terms of a wing or engine breaking away mandating an immediate landing, but that's just me. I haven't been trained on the B777 so I don't know if the procedures are similar for such emergencies - perhaps someone here can talk about it.
funfly;
I can certainly see a scenario in the not too distant future where freight is carried by pilotless aircraft.
The reason why I think this is not because it is not doable. It is doable right now. The Airbus is an entire digital platform and can be "remoted".
The problem is again, cognitive, not technical or social, (fear of getting on transportation with no operator). An airplane is not a streetcar or an elevator. The multitude of highly-complex decision-making which cannot yet be mimicked, is the hurdle. We may ask ourselves about the present A380 incident and try to come to terms with the design requirements which would have assessed the damage, created a course of action and carried it out, all without human intervention. It is always risky to say such stuff in one's own time, but I don't think pilotless flight is possible in the sense that it would be the routine, commercial way the world's airliners would be operated.
PJ2
Paxing All Over The World
Just a regular Pax
Bergerie1
This mirrors what I have seen in general commerce over the past 20 years. What I saw in Telecoms/IT was the move away from planned maintenance and prevention to fire fighting when it went wrong. Of course, general IT has the benefit that when it stops going round - it is already sitting on the ground. 
BUT it can bring a company to a juddering halt and I have seen it often times.
Then I saw the manager apply lots of money to fix the problem. He/She did not get criticised (by the mgmt) for not having spent the money on better maintenance, better quality kit, better staff to prevent the problem in the first place - just got praise for fixing it! The staff, meanwhile, saw them spend at least as much money as we had asked for before and been refused - and them watch them get praised! So, staff get disaffected, unappreciated and drift away. Others have to stay for the sake of their income and pension.
I'll bet that there are folks in every single carrier in the world who recognise that pattern. My reason for mentioning it is to say that the airlines are just going down the same slippery path as the rest of commerce.
PJ2 #34. Your Old/New photography simile is VERY good and so easily grasped. My nephew was fixated by aircraft from the outset and went up for the first time when he was 13. He's now 34 and RHS of a 738 in the Southern Hemisphere and he likes the 738 for it's blend of the old and new.
Bergerie1
We have never honoured sufficiently all those people who kept doing it right and were never noticed because through good planning, attention to detail, constant practice, and good airmanship they prevented potentially hazardous situations from becoming a catastrophe.
BUT it can bring a company to a juddering halt and I have seen it often times.Then I saw the manager apply lots of money to fix the problem. He/She did not get criticised (by the mgmt) for not having spent the money on better maintenance, better quality kit, better staff to prevent the problem in the first place - just got praise for fixing it! The staff, meanwhile, saw them spend at least as much money as we had asked for before and been refused - and them watch them get praised! So, staff get disaffected, unappreciated and drift away. Others have to stay for the sake of their income and pension.
I'll bet that there are folks in every single carrier in the world who recognise that pattern. My reason for mentioning it is to say that the airlines are just going down the same slippery path as the rest of commerce.
PJ2 #34. Your Old/New photography simile is VERY good and so easily grasped. My nephew was fixated by aircraft from the outset and went up for the first time when he was 13. He's now 34 and RHS of a 738 in the Southern Hemisphere and he likes the 738 for it's blend of the old and new.
Join Date: Jun 2009
Location: US
Posts: 497
Likes: 0
Received 0 Likes
on
0 Posts
Automation is fine if it is doing what you want it to do. Why let it run away if it isn't? Just disconnect it. Pushing buttons when things go amuck might solve the problem but disconnecting and hand flying will fix it every time if you know how. If you don't know how, get another job. The automation is there to help you, not take care of incompetent pilots.
Join Date: Jun 2009
Location: US
Posts: 497
Likes: 0
Received 0 Likes
on
0 Posts
I think that is the problem. If they don't hit the right buttons in time they are screwed. That is why it is nice to have a qualified crew instead of these students. A qualified crew could disconnect and make an easy approach, the students would just mash buttons and see what happens. We have a lot of qualified pilots out there so hiring students to fly aircraft they are not ready for is really stupid. Yes they are cheap, but they aren't qualified.
Person Of Interest
Join Date: Jan 2000
Location: Keystone Heights, Florida
Age: 68
Posts: 842
Likes: 0
Received 0 Likes
on
0 Posts
Trying to ease my mind
PTH...Off topic, but have you ever lived in Annapolis, Md., owned a sailboat and one day take two tired crewmembers to your home one night on our day off...if so, please PM me...
If not...I agree 100% with you...A DC-9 or 727 in good hands hold no torch to this new computer stuff...
If you're not the pilot I'm thinking of, sorry, but I learned my (and your) trade because of mentors like you...BTW, what were you doing in 1968 after getting out of the Navy???...If I'm on target...
If not...I agree 100% with you...A DC-9 or 727 in good hands hold no torch to this new computer stuff...
If you're not the pilot I'm thinking of, sorry, but I learned my (and your) trade because of mentors like you...BTW, what were you doing in 1968 after getting out of the Navy???...If I'm on target...
Jabiman;
It leaves them in the same place one is in when, after spending thousands of dollars and hundreds of hours hard work on learning a foreign language and having your very first encounter with reality in that country and not just ordering tea.... perhaps a complex, emotional situation that is rapidly unfolding with the person in front of you, (buying bananas and arguing the price, dealing with a car accident, dealing with a robbery or a heart attack, etc etc), and you can't understand what the (angry, sick?) person is saying to you and you need "the words" to explain what you're trying to accomplish but can't find them because of all the commotion, excitement, distractions, idioms and wild gestures that they didn't (and couldn't) teach you in language school. That's where it leaves you.
And for me, that's where the MCPL leaves a 250hr pilot in an airliner in which the captain is relying upon you when reality strikes.
PJ2.
So where does that leave those pilots who go straight to TR and then into a jet and whose employer encourages them to rely on automation as much as possible?
And for me, that's where the MCPL leaves a 250hr pilot in an airliner in which the captain is relying upon you when reality strikes.
PJ2.
Last edited by PJ2; 8th Nov 2010 at 01:05.
Join Date: Sep 2004
Location: Vancouver, B.C.
Posts: 10
Likes: 0
Received 0 Likes
on
0 Posts
We can postulate over the potential negative consequences of Automation dependency and the 'dumbing' down of basic piloting skills.
We can concern ourselves with the MPL's abbreviated curriculum and the 250 hr 'virgins' ability to avail him/herself in a situation but we certainly cannot dispute the irrevocable fact that accidents rates during the first decade of the 21st century have remained constant , more than we would like but nevertheless ,not on the increase.
Let's move forward , learning , embracing and adapting.
But most importantly - not regressing
Mister Approach
We can concern ourselves with the MPL's abbreviated curriculum and the 250 hr 'virgins' ability to avail him/herself in a situation but we certainly cannot dispute the irrevocable fact that accidents rates during the first decade of the 21st century have remained constant , more than we would like but nevertheless ,not on the increase.
Let's move forward , learning , embracing and adapting.
But most importantly - not regressing
Mister Approach
Mister Approach;
These are noble sentiments. You will not find anyone here disagreeing with your wishes.
The reality however, is quite different.
The commercial aircraft accident statistics do not agree with your "irrevocable fact" statements regarding accident rates remaining constant, and "not on the increase." PTH voices it well - "Zero" is what the rate should be and if not, it is where this industry should be aiming.
The rise shown in the Boeing documents doesn't make a trend yet as it only began climbing in 2006. But it is neither flat at zero, nor is it decreasing. There are clear reasons for this and this and other threads here are telling the industry why.
Below are three graphs from the larger Boeing "Statistical Summary of Commercial Jet Airplane Accidents 1959 - 2009" pdf. The first two graphs show the worldwide commercial jet fleet fatal accident rate and the second shows the same for only the US and Canada. For those who wish to compare which airplanes are involved, the third graph shows the Worldwide Commercial Jet Fleet Hull Loss and Hull Loss with Fatalities rates. The curves are slightly flatter for the worldwide fatal accident rate but sharply higher in the US and Canada.
PJ2
(better image quality can be had in the original report)
Accident Rates and Onboard Fatalities by Year
US and Canadian Operators Accident Rates by Year
Accident Rates by Airplane Type
We can concern ourselves with the MPL's abbreviated curriculum and the 250 hr 'virgins' ability to avail him/herself in a situation but we certainly cannot dispute the irrevocable fact that accidents rates during the first decade of the 21st century have remained constant , more than we would like but nevertheless ,not on the increase.
Let's move forward , learning , embracing and adapting.
But most importantly - not regressing
Let's move forward , learning , embracing and adapting.
But most importantly - not regressing
The reality however, is quite different.
The commercial aircraft accident statistics do not agree with your "irrevocable fact" statements regarding accident rates remaining constant, and "not on the increase." PTH voices it well - "Zero" is what the rate should be and if not, it is where this industry should be aiming.
The rise shown in the Boeing documents doesn't make a trend yet as it only began climbing in 2006. But it is neither flat at zero, nor is it decreasing. There are clear reasons for this and this and other threads here are telling the industry why.
Below are three graphs from the larger Boeing "Statistical Summary of Commercial Jet Airplane Accidents 1959 - 2009" pdf. The first two graphs show the worldwide commercial jet fleet fatal accident rate and the second shows the same for only the US and Canada. For those who wish to compare which airplanes are involved, the third graph shows the Worldwide Commercial Jet Fleet Hull Loss and Hull Loss with Fatalities rates. The curves are slightly flatter for the worldwide fatal accident rate but sharply higher in the US and Canada.
PJ2
(better image quality can be had in the original report)
Accident Rates and Onboard Fatalities by Year
US and Canadian Operators Accident Rates by Year
Accident Rates by Airplane Type
Thread Starter
Join Date: Apr 2005
Location: Australia
Posts: 1,414
Likes: 0
Received 0 Likes
on
0 Posts
A few years ago I was tasked to give dual instruction to a student pilot on a Cessna 152. He already had ten hours of dual but had not yet gone solo. His regular instructor was on leave hence I got called in.
We strapped in and I waited for him to begin his pre-start cockpit checks. In an apologetic voice he said he had forgotten to bring along his checklist. I said no worries, just go ahead and start the engine. After all he had flown several trips in the Cessna.
He then said he did not know how to start the engine without a checklist to guide him. He was so embarrassed and I felt sorry for him while mentally gritting my teeth that his instructor had taught him to rely solely on a checklist as a Good Thing. In the event we finally got airborne. After the session was finished and after the student had taxiied in I suggested he should park the brakes and shut down the engine. Embarrassed silence after which he said he was sorry but he did not know how to shut down the engine in the Cessna 152 without reference to checklist guidance.
At a different level, this is what is happening in some parts of the airline industry where new (and not so new) pilots have lost the confidence that would enable them to fly safely without the crutch of the autopilot and it's associated goodies
We strapped in and I waited for him to begin his pre-start cockpit checks. In an apologetic voice he said he had forgotten to bring along his checklist. I said no worries, just go ahead and start the engine. After all he had flown several trips in the Cessna.
He then said he did not know how to start the engine without a checklist to guide him. He was so embarrassed and I felt sorry for him while mentally gritting my teeth that his instructor had taught him to rely solely on a checklist as a Good Thing. In the event we finally got airborne. After the session was finished and after the student had taxiied in I suggested he should park the brakes and shut down the engine. Embarrassed silence after which he said he was sorry but he did not know how to shut down the engine in the Cessna 152 without reference to checklist guidance.
At a different level, this is what is happening in some parts of the airline industry where new (and not so new) pilots have lost the confidence that would enable them to fly safely without the crutch of the autopilot and it's associated goodies
Join Date: Sep 2004
Location: Vancouver, B.C.
Posts: 10
Likes: 0
Received 0 Likes
on
0 Posts
'we've been flying now for 107 years...as of dec 17 this year...and the accident rates should be zero...not steady.'
We've been practicing medicine for close to 5000 years yet 100's of thousands of people perish every year as a result of avoidable medical error , I can guarantee that most Physicians and primary healthcare providers would suggest that the figure should be zero as well and so do I.
Aviation has been remarkably safe and it continues to be so, it is the ideology behind the 'zero' accident rate that propagates this notion of infallibility .
I wish for a zero accident rate , I hope that the research into the accidents attributable to the over dependency in automation will serve to educate future generations of so called 'playstation' pilots but will we really ever have an infallible system capable of producing zero accidents ? I hope so , if any industry can do it , aviation can, but my reasoned , rational side suggests that we will never have perfection. If automation is the reason for our imperfections as pilots , then should we abandon it? Go back to the basics?
If we can logically conclude otherwise then it would be logical to embrace our new reality , understand our limitations and in turn attempt to mitigate potential traps.
Mister Approach
We've been practicing medicine for close to 5000 years yet 100's of thousands of people perish every year as a result of avoidable medical error , I can guarantee that most Physicians and primary healthcare providers would suggest that the figure should be zero as well and so do I.
Aviation has been remarkably safe and it continues to be so, it is the ideology behind the 'zero' accident rate that propagates this notion of infallibility .
I wish for a zero accident rate , I hope that the research into the accidents attributable to the over dependency in automation will serve to educate future generations of so called 'playstation' pilots but will we really ever have an infallible system capable of producing zero accidents ? I hope so , if any industry can do it , aviation can, but my reasoned , rational side suggests that we will never have perfection. If automation is the reason for our imperfections as pilots , then should we abandon it? Go back to the basics?
If we can logically conclude otherwise then it would be logical to embrace our new reality , understand our limitations and in turn attempt to mitigate potential traps.
Mister Approach
Join Date: May 2007
Location: Western Pacific
Posts: 721
Likes: 0
Received 0 Likes
on
0 Posts
Automation is not the problem. The miss-use of automation & the lack of automation monitoring are the problems. Along with over zealous cost cutting in training, which leads to over dependance on automation as a patch to try and cover up the shortcomings in the pilots these training systems produce. It's not usually the individual pilot's fault, the fault generally lies within the system.
When you combine this over dependance on automation with a lack of adequate knowledge of the automation systems, again caused by shortfalls in training, you start to see acidents like some of the recent ones that have been discussed on the various PPRuNe forums of late.
I recently had a TRE stand in front of a recurrent ground school I was in & state that there are at least three pilots on the flight deck of a B777. The Captain, the First Officer and the Autopilot. This kind of thinking will invariably lead to disaster sooner or later. Autopilots are not pilots, nor were they designed to be.
When you combine this over dependance on automation with a lack of adequate knowledge of the automation systems, again caused by shortfalls in training, you start to see acidents like some of the recent ones that have been discussed on the various PPRuNe forums of late.
I recently had a TRE stand in front of a recurrent ground school I was in & state that there are at least three pilots on the flight deck of a B777. The Captain, the First Officer and the Autopilot. This kind of thinking will invariably lead to disaster sooner or later. Autopilots are not pilots, nor were they designed to be.
Join Date: May 2007
Location: Western Pacific
Posts: 721
Likes: 0
Received 0 Likes
on
0 Posts
The autopilot is not a pilot. It is an aircraft 'manipulator' that only manipulates roll, pitch & thrust (I know - autoland. Lets not get too complicated!). And it only manipulates under the guidance of the pilot (read MCP selection).
To call it a pilot can, and sometimes does, lead an inexperienced, incompetent or complacent pilot into believing that he/she can 'hand over' to it & drop out of the loop of 'piloting' the aircraft for a period of time. You can't. And if you do, the stage is set for some unexpected event to place the aircraft in a situation that may not be recoverable, given the circumstances & crew ability and/or experience on the day.
It does not think, reason, draw on experience, make judgement calls, discuss, use CRM, learn, remember, feel or the myriad of other things we do as human beings. And, unless you remain completely in the loop, it does not provide feedback.
It is simply a tool used to assist the pilot as he/she 'pilots' the aircraft. Like an FMC, or a radio, or EICAS.
To call it a pilot can, and sometimes does, lead an inexperienced, incompetent or complacent pilot into believing that he/she can 'hand over' to it & drop out of the loop of 'piloting' the aircraft for a period of time. You can't. And if you do, the stage is set for some unexpected event to place the aircraft in a situation that may not be recoverable, given the circumstances & crew ability and/or experience on the day.
It does not think, reason, draw on experience, make judgement calls, discuss, use CRM, learn, remember, feel or the myriad of other things we do as human beings. And, unless you remain completely in the loop, it does not provide feedback.
It is simply a tool used to assist the pilot as he/she 'pilots' the aircraft. Like an FMC, or a radio, or EICAS.
Join Date: Jul 2000
Location: London
Posts: 1,256
Likes: 0
Received 0 Likes
on
0 Posts
Think more along the lines of mode confusion and better training to prevent it.
Also the accident rate is very low and has been for a number of years. It tends to be forgotten that a steady rate, with many more aircraft flying, equals more crashes and hull losses.
Also the accident rate is very low and has been for a number of years. It tends to be forgotten that a steady rate, with many more aircraft flying, equals more crashes and hull losses.
Mister Approach;
Over the past fourty years, (essentially the beginning of the '70's), the move towards privatization and deregulation of the economy, particularly in the United States, the need to strike a balance between "safety" and "profitability" has been extremely difficult to come to terms with. A perfectly safe airline doesn't fly at all, and an airline that is responding to the pressures for short-term profit by cutting costs to the bone without due regard for the effects of such cuts isn't going to be in business for long and likely wouldn't be a safe carrier to travel on.
That is exactly what the industry has been doing over the last fourty years to get the accident rate down as low as it is already.
The approach comes in many forms..."Error-trapping procedures", "Crew Resource Management", "Standard Operating Procedures", "Training Standards" and the industry has also responded with brilliant developments such as TAWS-EGPWS, (Don Bateman, Honeywell - google him), TCAS, Runway Incursion and Ground Position and yes, even high-level autoflight/autothrottle systems.
All these are on the technical side; the Human Factors side examines how people think, communicate, behave and how they respond to abnormal/emergency circumstances - it involves ergonomics, information display, checklist design, and a thousand other factors. This approach extends examination of organizational behaviours in root pathways to accidents. Google the pioneers of this work such as Charles Perrow, James Reason, Robert Helmreich and more recently, Sidney Decker, for examples of the hundreds of specialists doing this work. All these are currently existing mitigation responses in an attempt to lower the accident rate and they have worked - never perfectly, but the rate is lower now than it was in 1950 - dramatically so. But it is, as shown, now rising - the accident rate is turning around but it is too early to say whether it is a trend or a blip as the previous years' show.
The healthcare/medical profession is mentioned and has been discussed widely here at various times and places. Helmreich is doing work as others are in this area. See, "Systems Approaches to Surgical Quality and Safety: Human Factors Approaches in Medicine" for just one story on this.
While medicine is learning from aviation how to lower risk, reduce fatality rates, employ some of the mitigation approaches aviation has developed and so on, the healthcare industry has to develop their own methods for reducing the spectacularly-high fatality rate in medicine, (it is similar to, though higher, the automobile industry in the United States - about one fully-loaded B747 fatal accident every three days), and while that is a statement of fact, that statement, I know, is also an unfair comparison because the two endeavours are entirely different just in terms of problem diagnoses and the fact that medicine deals with patients and we deal with comparatively mechanically-uncomplicated machines and pretty straightforward operational circumstances. That's why medicine must develop its own mitigation strategies.
PJ2
Over the past fourty years, (essentially the beginning of the '70's), the move towards privatization and deregulation of the economy, particularly in the United States, the need to strike a balance between "safety" and "profitability" has been extremely difficult to come to terms with. A perfectly safe airline doesn't fly at all, and an airline that is responding to the pressures for short-term profit by cutting costs to the bone without due regard for the effects of such cuts isn't going to be in business for long and likely wouldn't be a safe carrier to travel on.
If we can logically conclude otherwise then it would be logical to embrace our new reality , understand our limitations and in turn attempt to mitigate potential traps.
The approach comes in many forms..."Error-trapping procedures", "Crew Resource Management", "Standard Operating Procedures", "Training Standards" and the industry has also responded with brilliant developments such as TAWS-EGPWS, (Don Bateman, Honeywell - google him), TCAS, Runway Incursion and Ground Position and yes, even high-level autoflight/autothrottle systems.
All these are on the technical side; the Human Factors side examines how people think, communicate, behave and how they respond to abnormal/emergency circumstances - it involves ergonomics, information display, checklist design, and a thousand other factors. This approach extends examination of organizational behaviours in root pathways to accidents. Google the pioneers of this work such as Charles Perrow, James Reason, Robert Helmreich and more recently, Sidney Decker, for examples of the hundreds of specialists doing this work. All these are currently existing mitigation responses in an attempt to lower the accident rate and they have worked - never perfectly, but the rate is lower now than it was in 1950 - dramatically so. But it is, as shown, now rising - the accident rate is turning around but it is too early to say whether it is a trend or a blip as the previous years' show.
The healthcare/medical profession is mentioned and has been discussed widely here at various times and places. Helmreich is doing work as others are in this area. See, "Systems Approaches to Surgical Quality and Safety: Human Factors Approaches in Medicine" for just one story on this.
While medicine is learning from aviation how to lower risk, reduce fatality rates, employ some of the mitigation approaches aviation has developed and so on, the healthcare industry has to develop their own methods for reducing the spectacularly-high fatality rate in medicine, (it is similar to, though higher, the automobile industry in the United States - about one fully-loaded B747 fatal accident every three days), and while that is a statement of fact, that statement, I know, is also an unfair comparison because the two endeavours are entirely different just in terms of problem diagnoses and the fact that medicine deals with patients and we deal with comparatively mechanically-uncomplicated machines and pretty straightforward operational circumstances. That's why medicine must develop its own mitigation strategies.
PJ2
Airmanship – filling the gaps in assumptions.
The reluctance to cite airmanship may be due to the difficulties in defining it or explaining how it might be taught (gained), when it fails, or is not used.
At best, airmanship is depicted with a model, a concept. The model shows what airmanship entails, but little of how it can be acquired, and less still of how its absence actually effects operations.
IMHO this is because the majority of the constituents are within the mind. How do pilots effect self discipline, what are the mental aspects of skill, how is proficiency (competency) judged, what entails requisite knowledge, a decision, judgment, etc.
Most airmanship attributes are judged by result, success / failure, but the process remains hidden.
We lack details of the data used in the report, but whatever assessments have been made they might only be interpretations of the crew’s behavior (their airmanship). The thoughts remain unknown – what was the pilot’s interpretation of the situation, on what basis was the course of action taken what was known / not known. Furthermore, there is the risk of hindsight bias.
Indications of problems with ‘airmanship’ might be found in broad categories of recent accidents, e.g. -
Speed related, particularly slow speed.
Disorientation – attitude instrument flying.
System mode of operation; human-technology interaction.
They all relate to displays, and thus aspects of awareness; note the link with Orasanu’s work #33.
As an alternative view of the problem, consider the technological changes in recent years.
Airspeed displays - dial vs tape (which way up); attitude formats – a simple display vs full screen with surrounding clutter (including mode annunciation).
We have lost the individualism of separate instruments, the new integrated displays might require greater interpretation.
The majority of the changes are evolutions, where each step has been accepted, sometimes judged superior at other times only acceptable in conjunction with other enhancements. All of these changes have been based on assumptions, the majority of which are unwritten, but each fraction of the industry ‘understands’ them – regulators, trainers, individuals – except like airmanship everyone has their version of the assumption and how the human fits in with it.
We have assumed that these changes will ‘improve safety’ whereas in reality (today’s problems) they may have hidden or moved the old errors – same low accident rate. Errors now appear in another form or elsewhere in the operation.
This is not a simple aspect of man–machine interface involving human factors. It is the understanding of human behavior with the machine, the gap between man and machine. A similar gap between aircraft certification and operational regulations, the gap between what management/training expects from procedures and what actually happens in normal operations, between safety and profitability, what we expect from humans and what the human actually accomplishes, and all these depend on context, which also is often an assumption.
There may not be a specific solution, but with a very wide view, the industry might benefit from considering some of the ‘hidden’ safety gaps in their assumptions.
Is airmanship expected to fill the gaps in these assumptions?
“Knowledge and error flow from the same mental sources, only success can tell one from the other.” (Ernst Mach, 1838-1916; “Knowledge and error”, 1905)
At best, airmanship is depicted with a model, a concept. The model shows what airmanship entails, but little of how it can be acquired, and less still of how its absence actually effects operations.
IMHO this is because the majority of the constituents are within the mind. How do pilots effect self discipline, what are the mental aspects of skill, how is proficiency (competency) judged, what entails requisite knowledge, a decision, judgment, etc.
Most airmanship attributes are judged by result, success / failure, but the process remains hidden.
We lack details of the data used in the report, but whatever assessments have been made they might only be interpretations of the crew’s behavior (their airmanship). The thoughts remain unknown – what was the pilot’s interpretation of the situation, on what basis was the course of action taken what was known / not known. Furthermore, there is the risk of hindsight bias.
Indications of problems with ‘airmanship’ might be found in broad categories of recent accidents, e.g. -
Speed related, particularly slow speed.
Disorientation – attitude instrument flying.
System mode of operation; human-technology interaction.
They all relate to displays, and thus aspects of awareness; note the link with Orasanu’s work #33.
As an alternative view of the problem, consider the technological changes in recent years.
Airspeed displays - dial vs tape (which way up); attitude formats – a simple display vs full screen with surrounding clutter (including mode annunciation).
We have lost the individualism of separate instruments, the new integrated displays might require greater interpretation.
The majority of the changes are evolutions, where each step has been accepted, sometimes judged superior at other times only acceptable in conjunction with other enhancements. All of these changes have been based on assumptions, the majority of which are unwritten, but each fraction of the industry ‘understands’ them – regulators, trainers, individuals – except like airmanship everyone has their version of the assumption and how the human fits in with it.
We have assumed that these changes will ‘improve safety’ whereas in reality (today’s problems) they may have hidden or moved the old errors – same low accident rate. Errors now appear in another form or elsewhere in the operation.
This is not a simple aspect of man–machine interface involving human factors. It is the understanding of human behavior with the machine, the gap between man and machine. A similar gap between aircraft certification and operational regulations, the gap between what management/training expects from procedures and what actually happens in normal operations, between safety and profitability, what we expect from humans and what the human actually accomplishes, and all these depend on context, which also is often an assumption.
There may not be a specific solution, but with a very wide view, the industry might benefit from considering some of the ‘hidden’ safety gaps in their assumptions.
Is airmanship expected to fill the gaps in these assumptions?
“Knowledge and error flow from the same mental sources, only success can tell one from the other.” (Ernst Mach, 1838-1916; “Knowledge and error”, 1905)
Join Date: Nov 2004
Location: Gangster Paradise, RSA
Posts: 230
Likes: 0
Received 0 Likes
on
0 Posts
Airmanship bla bla bla... Define it.... Most manufacturers will state not to troubleshoot. Taking the assumption you know exactly what the automation is supposed to do and it doesn't, disconnect fly manual. That is if you still know how.....Lot's of fancy talk around this subject, but it sounds like beating around the bush.
As for the wonderful accident rate compared over the years, compare traffic from 1950 and actual now....
As for the wonderful accident rate compared over the years, compare traffic from 1950 and actual now....
Maurice –
(1) Airmanship? Precisely my point ?
(2) The assumption is probably incorrect – in many instances crew’s do not know when to disconnect (change the level of automation) – either not appreciating the situation, or with understanding failing to act – full circle to Orasanu !
(3) Perhaps the problem is that we don’t know (understand) if we are beating around a bush or if it is any other object.
I hope that on reflection that you will realise that ‘rate’ considers the different traffic volumes.
A related problem for the industry is that with a constant low accident rate, yet increasing volume, the number of accidents will increase. Therefore in order to satisfy the ‘public perception’ of safety, the rate has to improve just to maintain the same number of accidents per year.
There are many problems in this area of safety management; see The paradoxes of almost totally safe transportation systems.
From this, another view of the ‘problem’ could be that it is a manifestation of applying conventional (error reducing) safety strategies to a system (via training) which is now in equilibrium, and thus requires error containment strategies. We need some errors in order to learn.
A failure or misapplication of CRM/TEM ?
(1) Airmanship? Precisely my point ?
(2) The assumption is probably incorrect – in many instances crew’s do not know when to disconnect (change the level of automation) – either not appreciating the situation, or with understanding failing to act – full circle to Orasanu !
(3) Perhaps the problem is that we don’t know (understand) if we are beating around a bush or if it is any other object.
I hope that on reflection that you will realise that ‘rate’ considers the different traffic volumes.
A related problem for the industry is that with a constant low accident rate, yet increasing volume, the number of accidents will increase. Therefore in order to satisfy the ‘public perception’ of safety, the rate has to improve just to maintain the same number of accidents per year.
There are many problems in this area of safety management; see The paradoxes of almost totally safe transportation systems.
From this, another view of the ‘problem’ could be that it is a manifestation of applying conventional (error reducing) safety strategies to a system (via training) which is now in equilibrium, and thus requires error containment strategies. We need some errors in order to learn.
A failure or misapplication of CRM/TEM ?