Automation Bogie raises it's head yet again

[Centaurus]

Europe Regulators To Curb Autopilots


By ANDY PASZTOR


(WSJ) European air-safety regulators, stepping up warnings about excessive dependence on cockpit automation, have told pilots of two widely used Airbus jetliner models to avoid routinely using their autopilots in certain emergencies.


The directive issued last week by the European Aviation Safety Agency, or EASA, specifically covers emergency procedures for more than 1,000 Airbus A330 and A340 jets in the event of major airspeed-sensor malfunctions. More than two-thirds of the two-engine, wide-body planes are flown by European or Asian airlines.


More broadly, the move reflects growing concerns by international safety experts about the hazards of undue reliance by pilots on on-board automated systems, following a spate of incidents pointing to pilot mistakes and confusion stemming from improper use of automated safety aids.


Safety officials believe many pilots flying a wide range of jets-across the U.S. as well as other regions-may need additional training to cope effectively with emergencies when autopilots or automated thrust-control systems are unavailable or can't be trusted.


"Given the increasing dependence on automation" in the latest generation of jetliners, according to Deborah Hersman, chairman of the U.S. National Transportation Safety Board, "you've got to have pilots who are prepared for all scenarios, especially those in which they have a lack of information" from typical flight-management sources.


If two of the three external-airspeed indicators on the Airbus models singled out by last week's directive provide unreliable readings, autopilots and automated engine settings are designed to instantly disconnect. But instead of routinely re-engaging both at the first sign of trouble to try to control the aircraft, EASA now requires pilots to wait at least 30 seconds to determine if reliable information is flowing into the automated systems.EASA's directive also follows the recent release of a long-awaited U.S. air-safety study emphasizing that commercial pilots tend to abdicate too much responsibility to on-board computers, partly because most current training programs emphasize that automated systems generally are more adept at handling in-flight emergencies.


Crews also must perform various other checks to ensure that the automated systems will issue appropriate orders once they are switched back on.


When speed sensors feed unreliable data to flight-management computers, EASA said the result can be autopilots issuing abrupt and "inappropriate" climb or descend commands, which may "constitute an unsafe conditions." Regulators patterned the directive after a separate safety warning issued by manufacturer Airbus, a unit of European Aeronautic Defence & Space Co.


The move comes after a series of dangerous airspeed-measurement failures and unpredictable or dangerous autopilot commands on Airbus A330s in recent years. The sequence of events that led to the June 2009 crash of an Air France Airbus A330 flying to Paris from Rio de Janeiro apparently started with unreliable airspeed indications and autopilot difficulties. No formal cause has been determined for the crash, which killed 228 people.

............................................................ ......................

Fine words telling us what we already know. Pity the regulators don't insist that operators under their control come up with a viable solution to automation dependancy

[John Farley]

If two of the three external-airspeed indicators on the Airbus models singled out by last week's directive provide unreliable readings, autopilots and automated engine settings are designed to instantly disconnect.

My understanding of what should then happen in this eventuality is that the pilots should manually maintain pitch and roll attitude and the normal power setting for the stage of flight while considering their options for continuing the flight with degraded airspeed information.

This is easier said than done for pilots taken by surprise - especially at night, turbulence or cloud, especially when they are at high altitude and heavy. In these conditions a limited safe speed range is available requiring pretty careful and accurate handling in circumstances where the pilots (necessarily) will have little recent currency and practice.

In my view following such airspeed anomalies it would be better for the 'system' not to disengage but to degrade to a basic attitude and thrust hold mode accompanied by a pretty serious warning about what was going on.

[InstinctsError]

..........Automation ........ the word fires up a battle among some others. but... again it should be only an assistance and aid measure to cope up with the whole scenario and not just the whole means of the unnecessary reliance on the circuit boards ........

Agreed .. skills dont equal a programme :}

[aterpster]

John Farley:

In my view following such airspeed anomalies it would be better for the 'system' not to disengage but to degrade to a basic attitude and thrust hold mode accompanied by a pretty serious warning about what was going on.

Sometimes, hand-flying using basic attitude instrument flying skills and a global view of all available information will usually best sort out which is invalid information. (well, perhaps not with the lack of basic skills that seem more prevalent these days.):eek:

The accident report of the December 1, 1974, crash of a NWA 727 in New York should be required reading for all of today's active crewmembers. They forgot to turn on the pitot heaters and when "the stuff hit the fan" they got fixated on the bad information to the exclusion of attitude instrument flying.

[Capn Bloggs]

The accident report of the December 1, 1974, crash of a NWA 727 in New York should be required reading for all of today's active crewmembers. They forgot to turn on the pitot heaters and when "the stuff hit the fan" they got fixated on the bad information to the exclusion of attitude instrument flying.
The two 757 accidents in South America ( Aeroperu 603 (http://en.wikipedia.org/wiki/Aeroper%C3%BA_Flight_603) and Birgenair 301 (http://en.wikipedia.org/wiki/Birgenair_Flight_301)) are closer to the modern magenta-line era and are sickening.

[HazelNuts39]

The EASA AD discussed in the newspaper article is available here (http://ad.easa.europa.eu/ad/2010-0271)

[Mac the Knife]

"In my view following such airspeed anomalies it would be better for the 'system' not to disengage but to degrade to a basic attitude and thrust hold mode accompanied by a pretty serious warning about what was going on."

Whether is is medical or aeronautical, automated systems (patient ventilators for example) should degrade gracefully, sustaining basic functions rather than just cutting out. The greater reliance on automation the more graceful the degradation needs to be rather than dumping the surprised operator directly into a manual/raw data situation.

[SNS3Guppy]

That's all good and well, if the system is able to accurately define what is causing the problem. If it's not an input error, not a malfunction of the autopilot computer itself, then one is left with other problems that can produce similar results, but which will continue to be a problem should the autopilot degrade to a lesser state of control.

Eliminating the autopilot completely takes all the problems out of the look, returns control to the pilot, and allows the matter to be resolved using a procedure.

Attempting to simply fall back to basic autopilot functions when the issue is a runaway stabilizer trim, for example, may only escalate the problem. Taking autopilot away and allowing the crew to manually fly eliminates the trend, allows the crew to restore equilibrium, and then address the problem.

I agree wholeheartedly that stepping "back into the loop" in turbulent night conditions in the top of a cloud layer or a mountain wave can be a challenge, but then isn't this the reason for having a crew on board in the first place? Fly and control the airplane. Make decisions that the automation cannot. Evaluate. Troubleshoot.

Raw data training these days is often seen as redundant, and in many cases, "raw data" still includes use of the flight director. Crews should be comfortable transitioning between manual flying, and automation, equally proficient in both, and this practice should be approved and encouraged by operators.

[John Farley]

Chaps

I well appreciate that by manually flying the aircraft you may be able to diagnose what info is right or wrong but I would prefer the crew to be able to gather their wits about them first, check what all the attitude indicators are saying and then take over and start investigating.

Is this not better than having the aircraft unexpectedly dumped on you with some turbulence induced roll and pitch away from cruise already happening?

OK some oldies were necessarily trained to cope with this but a lot of people today may have many other useful skills but they don't include tricky handling and before you know it a UA is off and running.

[jcjeant]

Hi,

• The July 17, 2007, an Airbus A320 TAM left the runway 35L at Congonhas Sao Paulo (Brazil), killing 199 people. In its final report on Brazilian CENIP says the concept of automation of the A 320 does not allow drivers to always have a good knowledge of the situation:



The Concept of the automation in the A-320 Does not Always Allow The Pilot to know EXACTLY how the Operation are Being Performed by The System.



On 27 November 2008, the crew of an A 320 lost control of his aircraft at low altitude over the sea near Perpignan. The accident victim was 7. During this flight, blocking two probes incident led to the miscalculation of the characteristic speeds (and Vαprot Vαmax), rejection of ADR 3, the direct passage into law and had rendered inoperative automatic compensation. This succession of automation in a very complex system has not been seen by the crew. The pilots did not understand what was happening. In its final report recommended that the BEA very diplomatic way to the European Agency for Aviation Safety Agency (EASA) to ensure that pilots are better informed of the situation at Airbus reconfiguration of systems:



BEA recommends that EFSA carries out a study to improve safety standards system certification warning crews during reconfiguration of flight control systems



• The June 28, 2010, EASA alerted operators of aircraft of type A 330 / A 340 that in some cases a failure of one engine on take-off could lead to a surge toward the ground and limit the potential for driver counter the phenomenon:



This condition leads to a movement of The Elevator To The zero position, Which has inducer pitch down movement INSTEAD of a pitch up movement Needed to lift off. In addition, it leads to a limitation of The Pilot control is pitch axis and limits capacity to The Pilot Count The pitch down movement Düring this flight phase, Which year deriving their unsafe condition.



• The August 24, 2010, during a flight between Khartoum and Beirut, an Airbus A320 suffered a power outage announcement of imprecisely to the pilots and caused the uncontrolled activation of the trim of the rudder. The AAIB recommended that British crews Airbus A 320 are informed:



Safety Recommendation 2010-092: It Is Recommended That all operators of Airbus alert A320-series aircraft of The Possibility That year Electrical Power Generation System Fault May Not Be On The ECAM annunciated clearly ", and May lead to uncommanded rudder trim operation.



***



Because of the complexity of hyper Airbus, it may happen that the pilots did not understand what the automation and worse, they do not know what to do to get out of a situation that they no longer control !

Source:
L?hyper complexité des systèmes a des limites qu?Airbus ignore. : Les dossiers noirs du transport aérien (http://henrimarnetcornus.20minutes-blogs.fr/archive/2010/12/26/l-hyper-complexite-des-systemes-a-des-limites-qu-airbus-igno.html)

[SNS3Guppy]

OK some oldies were necessarily trained to cope with this but a lot of people today may have many other useful skills but they don't include tricky handling and before you know it a UA is off and running.

I've never seen a training program that didn't include unusual attitude recoveries and or jet upsets. I finished recurrent training at the end of last month, and it certainly included upset recoveries, and automated, as well as manually flown procedures.

During an approach into Kabul this month, a generator failed, and tripped off, taking that side of the split bus with it, as well as the essential bus. As it was going, the autopilot began react erratically, the control column began to vibrate and shake, and then the yoke began to rock back and forth with a high frequency vibration somewhat like an aileron buzz.

The quick solution was to disconnect the autopilot and handfly, while troubleshooting the problem using checklists and procedures.

The autopilot in the airplane I fly will drop from "command" to "manual," providing basic functions, during certain types of malfunctions. However, when in doubt, get rid of it and fall back to basic flying. It is, after all, something one should have mastered at the student pilot level.

[parabellum]

get rid of it and fall back to basic flying.


Agreed. Also pray that your FO is more than a 250 hour lad/lass who is straight out of flying training.

The message that comes out loud and clear from the original post is that some of todays pilots, whilst relying heavily on automation, are not doing enough to fully learn and understand how it does, or doesn't work.

How many times have you been in the SIM and seen someone, when given a simple problem, get themselves into an awful mess through a lack of system knowledge?

Maybe it is not so much an individuals fault but more symptomatic of todays flying training?

[Dengue_Dude]

As an ex flight engineer (now), I've taught glass cockpit for quite a few years.

Observations like 'you are now inundated with information' have now run full circle. For many years it was regarded as a good thing.

I suspect now it's not viewed as being the bonus it was designed to be. Even in my last years of flying, I was conscious that the levels of spacial awareness seemed to be reducing as crews 'followed the coloured lines'.

I think glass cockpits are brilliant, but I hope that we still will keep hold of those fundamental skills, the ability to use mental arithmetic and the motivation to keep those skills honed.

Sadly, I don't think that's happening. I think we're allowing a culture of complacency to develop despite all the courses preaching the contrary.

We'll see. I suspect I'll be the subject of a wave of rhetoric by outraged crew, but I flew for over 35 years and that's what I saw.

[BOAC]

Is it time to wake it up again?

http://www.pprune.org/safety-crm-qa-emergency-response-planning/379780-computers-cockpit-safety-aviation.html

(Best ignore all the Islamabad stuff at the end.)

[Gretchenfrage]

Sir, my need is sore.
Spirits that I've cited
My commands ignore.

(J.W.v. Goethe's Sorcerer's Apprentice)

[protectthehornet]

all that STUFF is built by engineers for engineers

BUT what a pilot NEEDS is:

an easy way to make sure the plane is flying properly

that it is on the propert course

that it is performing properly in terms of mechanical health

that threats to the plane can be dealt with (IE weather radar)

I imagine that much of today's cockpit information could be dealt with in other ways, for example, on takeoff...a computer voice says POWER set, engines nominal

freeing the pilots to look out the window

but, whatever method, pilots, airlines, governmental agencies, must allocate resources to make sure pilots can handle problems, including problems that are handled in an impromptu fashion...making things up as you go if there is no time to do a checklist

[SNS3Guppy]

How many times have you been in the SIM and seen someone, when given a simple problem, get themselves into an awful mess through a lack of system knowledge?

Generally if people get into trouble for systems knowledge or lack thereof, it's from failure to follow procedure. When people begin trying to out-think the airplane rather than pick up a checklist and work through a problem, they get in trouble.

I do see people get into trouble by trying to follow through with the automation when they really ought to get rid of it. I had a sim partner once that could fly the airplane very well, but who insisted on doing everything on the autopilot. I understood that he wanted to learn that panel and work it well, and I understood that he'd been counseled to let the autopilot work for him as much as possible in the upcoming checkride. The problem was that whenever he started to get behind things, trying to fix it through the autopilot only made it worse. It might be that the descent wheel didn't start him down as fast as he'd have liked on the approach, and ended up high. Trying to catch up by increasing the rate would see him missing his MDA. Perhaps it was being a little fast or a shirt turn onto the localizer, and then blowing through the approach. In each case, he saw the problem coming, and disconnecting the autopilot would have fixed the problem for him by allowing him to handfly and correct the problem. Sometimes trying to work through something with the autopilot is a little like trying to run a race on stilts. It's easier sometimes, and sometimes more effective, to remove the stilts, shut off that autopilot, and fly.

[Dani]

While I fully appreciate the discussion whether automatics are good or bad or both, and I can understand some people's concern about pilots loosing pilot's skills, I must say that this AD has nothing to do with it!

If you read the AD carefully (and if we are talking about the same one, namely the one HazelNuts39 presented us), it's not about automation but about correct failure finding in case of one or several Air Data failures. This can happen in any aircraft, even in steam-driven cockpits. Needless to say, that in most multiple AD failures, the autopilot disconnects by itself. The problem pilot's routinely have with such type of failures in the cockpit is not the lack of manual skills, but the wrong way of identifying the correct AD source or the insufficient use of pitch and power.

Modern cockpit have even more advantages than older ones, i.e. they offer more independent sources to troubleshoot (e.g. GS indication from FMS).

[Capn Bloggs]

Maybe it is not so much an individuals fault but more symptomatic of todays flying training?
Absolutely agree. It is the fault of the regulators and company Check and Training that is the root cause. A pilot cannot and should not be expected to work out what he needs to know/what techniques/skills he should have. He didn't design the aeroplane and he didn't dream up the concept of flight. Our industry and society in general gets better because the lessons learned by earlier generations are used by current day society, not by pilots having to individually "gain" all that experience, first-hand, over and over again.

From where I sit at least, there has been a fundamental failure of the hierarchy to ensure that pilots are equipped with the skills and knowledge they need to operate in the magenta-line world. It's not as though this problem has just popped up and people have been caught out. The writing has been on the wall for years but has been ignored.

I must say that this AD has nothing to do with it!

I disagree. This is all about crews being unfamiliar enough with handflying that they get themselves into bother when they have to so they leave hte AP in for too long. A crew who routinely hand flies in a less-automated aircraft will cope much better with a loss of airspeed because they are used to it (and probably know the powers and attitudes like the back of their hand); they are much more likely to get rid of the AP earlier.

[Metro man]

Pull out the QRH on the A320 and you will find a table of power settings and attitudes for unreliable airspeed.

On my previous aircraft I had a good idea of these as I had to apply them myself every time I flew, now I need to refresh my memory as I just don't use them so often.

An advanced autopilot is a major bonus during an emergency as it reduces the workload of simply keeping the aircraft flying safely, enabling more attention to be devoted to dealing with the problem. These days airspace is more complicated and congested. Tolerences are more critical and setting up an approach much more involved.

In the 1960s the controller could probably find you a vacant block of airspace to play around in while you sorted out what was going wrong with your B707, followed by a radar vector straight on to the ILS.

These days you will need to maintain tolerances in a hold and sequence into a STAR, following speed and altitude constraints. Don't forget you still need to run the emergency checklists and procrdures as well.

I certainly wouldn't want to be operating early generation jets in modern airspace.

[Centaurus]

An advanced autopilot is a major bonus during an emergency as it reduces the workload of simply keeping the aircraft flying safely, enabling more attention to be devoted to dealing with the problem. These days airspace is more complicated and congested. Tolerences are more critical and setting up an approach much more involved.

Depends entirely what sort of emergency or just merely the garden variety "Non-Normal" like a generator, pack, uncommanded pressurisation and so on.

The discussion is more about an undue attraction to automatics like the example I observed when a crew took up an instrument approach holding pattern using the magenta line and heading bug to trace their flight path over the magenta line holding pattern.

Pity that a minor programming error placed the holding fix which was a VOR, some 15 miles from the actual VOR and that over the fix the VOR/DME showed 15 DME which you would have thought would have picked up by the crew as somewhat unusual. At the same time the VOR/RMI needle was pointing to where the actual VOR was 15 miles away.

Fortunately it was in the simulator since the real aircraft would have eventually flown into a nearby mountain caused by the crew blindly following a false position magenta line. The reluctance of some crews to cross check frequently with basic navaids during even a normal flight is symptomatic of the inherent dangers of automation complacency. It is plain laziness.

When in the simulator, you see an experienced pilot furiously programming a visual circuit into the FMC, when in the first place all he was asked for, was a take off climb to 1500 ft, turn downwind at 30 degrees angle of bank, maintaining 2 mile spacing from the runway clearly visible from his seat using raw data and manual flying with no autothrottle, turn base and land - all in CAVOK - then you have now seen the archetypal automatics dependant pilot.

[haughtney1]

Just out of interest...

Is the birdy (FPV we call it on the 777) purely derived through the IRS's? or is there some ADC input?
My company has done the delightful thing of removing the performance inflight section of the ORH on the 777.

[Jabiman]

Air Crash Observer: Pilots out of Control on Autopilot (http://aircrashobserver.com/e107/news.php?extend.503.3)
In March 2010 India's aviation regulator said that most pilots in the country are not adequately trained to fly an aircraft in the automated mode, which is used on half of all domestic flights. "From now on they should fly their aircraft only in the manual mode until they train their pilots to operate the automated technology," said the Directorate General of Civil Aviation.
I wonder how many other nations this applies to also and if the bean counters have their way then it could happen elsewhere.

[N1EPR]

All pilots should have the ability to take over from an autopilot and be able to control the aircraft AND THINK at the same time.

Manual flying skills can only be acquired and kept by practice. I first discovered this as we transitioned to glass cockpits. After a few months of mostly autopilot use I found my manual flying skills were not up to my standards. I then started doing more manual flying and less autopilot use which corrected the problem. Since manual skills may be needed at altitude as well as other times your practice should include cruise too. By manual flying I mean no autopilot or autothrottles.

It must be realized that if your manual skills are rusty it will take more mental capability to fly the aircraft and leave less mental prowess to think about your next course of action or how to correct whatever problem you are dealing with. Your manual skills should be almost second nature. Do what ever it takes to accomplish that and when the day comes and its your turn to do your best to save a ship you will be ready.

[SNS3Guppy]

I wonder how many other nations this applies to also and if the bean counters have their way then it could happen elsewhere.

Indian aviation is a constant source of entertaining issues, such as widespread logbook fraud.

It's not a "beancounter" issue. No airline attempts to save money by failing to train it's crews on aircraft systems (including automation). It doesn't cost any more; there's nothing to be saved by not emphasizing full useage. Indian aviation has never been the gold standard for training, competence, veracity, honesty, or transparency. In a country where everybody in the business likes to call themselves "commander this," or "colonel that," "wing commander," "air commander, or some other ridiculous and pretentious title, you have a cultural issue, not an accounting one.

The thread is about over-reliance on automation, not about lack of training in it. The problem, if indeed such is the case, isn't lack of training in automation, but if anything not enough emphasis on basic raw flying skills. This may be owing in some cases to operators that insist on the maximum usage of automation to provide the smoothest flight for passengers, to some pilots who simply use it more out of preference or habit.

A change that I noted during my last recurrent session was that we've eliminated the requirement to conduct a raw data approach during the checkride. This wasn't an economic issue, but a change incorporated by the FAA to reflect real-world conditions; new procedures were added with changes in additional authorizations. No training costs were saved, but changes were incorporated. Personally, I'd prefer to see us retain the raw data procedures, but it's not my call. It's not the "beancounter's" call, either.

"Beancounters" don't run training departments.

[Loose rivets]

At the end of the day, those pilots that have spend years hand flying, will not have a mental demand-overload if they suddenly have to grab the controls.

The thousands of hours I spent on Viscounts, Heralds and F27s, left me with an autopilot somewhere in my brain. After 5 years of not flying, I suddenly found myself height-holding to the thickness of the needle, without really thinking about it. I was typical of every single pilot I flew with in those days.

Summer nights on Viscounts without radar were something of a baptism of fire for young pilots. The guy in the left was just glad he didn't have to avoid flack, so when in the middle of a CB, the horizon bar became lost behind the edges, he would think little of it. It was also a good education in containing one's fear.

I would hate those days to return, but what has taken its place is some kind of hybrid operation - neither skilled master of the machine, or possessor of a computer that is capable of calling on a wide enough spectrum of learned data. It is not a desirable compromise.

[Dani]

Capn Bloggs:
I disagree. This is all about crews being unfamiliar enough with handflying that they get themselves into bother when they have to so they leave hte AP in for too long.

Then I'm afraid you haven't read the concerned AD. It says:

AD No.: 2010-0271:
When there are significant differences between all airspeed sources, the
flight controls of an Airbus A330 or A340 aeroplane will revert to alternate
law, the autopilot (AP) and the auto-thrust (A/THR) automatically disconnect,
and the Flight Directors (FD) bars are automatically removed.

It is - in this case - not about leaving the AP in for too long, but for re-engaging it again before proper trouble shooting has been done. This is a clear issue of proper failure handling and good CRM and not about over-reliance on automatics.

Dani

[FlightDetent]

While I fully appreciate the discussion whether automatics are good or bad or both, and I can understand some people's concern about pilots loosing pilot's skills, I must say that this AD has nothing to do with it!

Thank you Dani, voice of reason at last. Loose Rivets, you have one too sir.

Loss or lack of manual skills is an issue, but Colgan will not be prevented by mandating 1500 hrs. Why not 1550? Recognizing approach to stall and proper stall recovery is mandatory task pre-solo, at 10 hrs total.

FTOs are historically called schools but they are quite far from it, normal businesses run for profit. Why fight the syndromes and not cure the diesease? Do teach pilots in traning and screen them properly? LH does it right, tight selection, stringent pass rules and 250 FOs are ready for the job in a novice position. Any airline could do the same, should they decide to do so.

Low hour RYR FO in Ciampino landed 737 uneventfully single-pilot, I applaud him/her. The pilot was trained for such task, and performed well. There's nothing heroic about it yet it speaks clearly about his training.

LH again, cross-wired sidestick, FO saves the day by using a learned technique.

Kos hard landing clearly uncovered discontinuity in follow-up on training progress. No amount hand flown of hours will substitute that. Yes, the trainee may be slow to acquire skill and need more hours to reach the target, but ultimately it is the pass-grade process which failed.

Prescribing 500 simulator landings will not cure that, faster learners will be unnecessarily tied to the box at immense cost and pilot who needs 600 of them will still be held short.

My 2c
FD (the un-real)

PS: As noted in the beginning loss of manual flying is an issue. I believe it is not the issue which bends metal. Let's read again on A340 in Toronto, AA in Jamaica, Garuda in Jogjakarta and Southwest in Midway. Or Little Rock, mother of all that (still) goes wrong. Drilling holes in the sky, even manual raw data NPAs in limiting conditions will not improve that. Sure it builds confidence but overconfidence is what kills people. Pilot that can professionally perform V1 cut and OEI landing as well as handle manual reversion is ready to fly the line, ready to learn the line. Provided that his previous training did not leave terra incognita elsewhere in the required knowledge.

[BOAC]

It is - in this case - not about leaving the AP in for too long, but for re-engaging it again before proper trouble shooting has been done. This is a clear issue of proper failure handling and good CRM and not about over-reliance on automatics. - part I I agree with, but the last phrase (my bold) not. We need to identify (and gently coax away) the source of the mindset that needs to re-engage the A/P with 'unseemly haste'.

[FlightDetent]

We need to identify (and gently coax away) the source of the mindset that needs to re-engage the A/P with 'unseemly haste'.
Exactly as you speak, BOAC, all agreed. The automation is not the problem, it is a just a simple tool. When NDBs came, all learned that the tool has limitations, i.e. the night shift (my english vocabulary reached limits here). Same for ILS false lobes. Training against autopilot misbehaviour is only half the story. Back to basics, it is a triangle: man-task-machine. Once this is explained, cockpit automation becomes just another tool just like a power steering on a car. Those not trained for AP use may face dire consequences one day. Not the consequences of AP, but consequences of their inproper understanding.

FD (the un-real)

[Metro man]

In March 2010 India's aviation regulator said that most pilots in the country are not adequately trained to fly an aircraft in the automated mode, which is used on half of all domestic flights. "From now on they should fly their aircraft only in the manual mode until they train their pilots to operate the automated technology," said the Directorate General of Civil Aviation.

Now that will be interesting, manually flying new generation aircraft, designed to be operated with the automatics.

Congested airspace, overworked controllers, poor infrastructure in most of the country and the regulator wants to remove the automation and have everyone flying by hand.

Get rid of all the expat pilots and rapidly promote the youngsters with little experience in either manual or automated environments.

Now where did I put my wizz wheel ?;)

[henra]

John Farley
In my view following such airspeed anomalies it would be better for the 'system' not to disengage but to degrade to a basic attitude and thrust hold mode accompanied by a pretty serious warning about what was going on.

I fully agree !!!
I'm glad to see this proposed by someone who knows what he is talking about and who himself surely wouldn't need this help but appreciates that it would improve overall safety.
It would buy the surprised crew precious time to sort things out.

IMHO it is not really helpful in today's airliners that once the regular automation gives up, a possibly mis- trimmed aircraft in dangerous (and potentially unclear) attitude and maybe config is handed back to completely puzzled pilots.

Admittedly it requires reliable attitude information but if that is lost also manual flying won't save the day....

[cessnapete]

This trend has been a long time coming.
I was admonished in BA by the Route Check Capt 10 years ago for hand flying to 10000ft in severe CAVOK departing from a very sleepy Carib airport.
He said on the B744, we should always use maximum automatics to avoid overloading the NHP!.
I believe that on the BA B777 even when manual flying, use of the A/T is mandatory, so you can't win.

[Centaurus]

Friend of mine undertook his first simulator assessment session (A320)after arrival with an Indonesian low cost carrier. He had 16,000 hours mainly 757/767 and a type rating A320 simulator only. Cleared for take off with left turn (simulator) under visual day conditions. Just before bending the aircraft around to the designated heading at 2000 ft, he glanced over his shoulder in a reflex action to check all clear for traffic.

Immediately the Indonesian check captain stopped the simulator and demanded an explanation of why the pilot checked all clear before commencing the turn. The astonished pilot said it was good airmanship to have a look if visual conditions and that he had done this for years.
The check captain said "You should never look outside for traffic- you should always rely on TCAS to warn you of traffic". Another case of automation gone mad. And they walk among us in Indonesia....

[PA-28-180]

" he glanced over his shoulder in a reflex action to check all clear for traffic. "

Don't know about the rest of this crazy world, but this is EXACTLY how I was trained during my initial PPL flight training!
ALWAYS look BEFORE turning/climbing/descending (as much as possible) - radar/TCAS is great....but NOT fool proof! :ugh:

[infrequentflyer789]

The BA B777 even when manual flying, use of the A/T is mandatory, so you can't win.

How the heck does that work in the event of unreliable airspeed ? A/T carries on setting thrust based on known bad air data ?!

Or is it that sometimes A/T will drop out, but you can't force it to drop out ? I thought one of the key issues of the great A v B FBW debate was that B allowed you to override it ?

[FLCH]

The check captain said "You should never look outside for traffic- you should always rely on TCAS to warn you of traffic". Another case of automation gone mad. And they walk among us in Indonesia....

I guess the buzzards and other large birds of prey have transponders strapped to them !

If that's the attitude, then it's just a matter of time before there's a mid air collision.

[protectthehornet]

just how quickly have we forgotten the Brazil tragedy, in which TCAS didn't work because of lack of transponder signal from one plane.

just be a good pilot...look around...oh and for birds...if you bank sharply, you may take the engines out of the plane of the bird formation.

think about it.

[wiggy]

The BA B777 even when manual flying, use of the A/T is mandatory, so you can't win.

How the heck does that work in the event of unreliable airspeed ? A/T carries on setting thrust based on known bad air data ?!



No need to get worked up about A vs. B or not being able to disconnect the Autothrottle. For normal ops the BA 777 SOP is indeed to use the Autothrottle at all times when it's available and functioning correctly, however in the event of unreliable airspeed indications one of the several initial actions is to disengage the Autothrottle and use "manual" power.

[Caygill]

It must be realized that if your manual skills are rusty it will take more mental capability to fly the aircraft and leave less mental prowess to think about your next course of action or how to correct whatever problem you are dealing with. Your manual skills should be almost second nature. Do what ever it takes to accomplish that and when the day comes and its your turn to do your best to save a ship you will be ready.Very true, such basic skills are and need to be "fully automated" tasks. No pilot (or whatever operator of almost any complex set of tasks) would be able to fly efficiently without having these skills as totally action models. The psychological load would simply be to great if every single action would be a considered response to a megaload of sensory inputs.

Having lost your level of "human automation" will not only make you slow to react, it also will make you bad to handle any additional stress factors.

[Piltdown Man]

I'm going to make myself a target now...

I don't think this is an "over reliance on automation" issue. It is a "What do you do if your air data system supplies some or totally corrupt data?" problem. No matter what aircraft you fly - glass, steam or otherwise, you have to have something to hang your hat on. A modern aircraft is capable of generating flight path data independent of the air data system and maybe we should start to use this information to get back to "Straight and Level." From a stable platform which is under control (and probably being manually flown) it should be possible to start to work out what has failed and take appropriate action.

PM

[Safety Concerns]

piltdown man, at last someone with the vision to see beyond the usual nonsense about automation.

man and machine is the safest way forward not man or machine.

[Loose rivets]

There was a guy on the NOVA?? or some such program, who showed us what he would have liked the 447 PF to be doing. He suggested 5 degrees nose up, and a specific power setting. Clearly, this is only the start of containing the situation.

The two and four engined turboprops I flew were all much of a muchness in terms of basic control, but when on a slippery jet, it was quite a different story. One thing on the BAC 1-11, was the unpleasant resonant frequency of the wings. They really would twang sometimes, and the FD 108 would become a vertical bar of colours for several seconds. This made accurate interpretation of the pitch very difficult if not impossible in the short term. Needless to say, this could be repeated over and over for several minutes and there were times it seemed automation might be able to do a better job. However, there was one thing the sensors didn't take into account, and that was vertical acceleration.

Firstly, I have to say I don't know if modern kit inputs vertical acceleration data.

I know we have it drummed into us to forget seat-of-the-pants flying, and believe the instruments, but in our scenario, we are already down to basics. Suggesting a set pitch and power leaves me a little worried. I found anticipating the power needed, by the feel of the vertical acceleration was something that put me ahead of the game. There is no way in hell you can look to EPRs, HP rpms or N1s for such human input, it has to be initiated very quickly and with only moderate accuracy.

I hasten to add, this movement had to be contained within a small range.

Throttle position was available and meaningful to us on those simple machines, and when I read of throttles that don't move, and of insistence on A-T, I just feel glad I was flying in a different era. And what is also obvious, it needs the kind of practice a box fixed to the planet can never provide. Pilots must be allowed to fly their aircraft on a regular basis - especially in rough air.

[fokker1000]

Sadly, FBW newbies haven't had the good luck to fly aircraft that require basic power, attitude equals performance.
If it's all gone wrong this will still keep you out of trouble, or reduce your risk of getting into more of it IMHO.
Auto thrust with non moving levers is not the best idea. I've flown AC with both and dislike them [non movers], especially on rough old night into a nastey island with a big 'hill' next to the runwayand your speed trend arrow doing the hokey kokey!!!!!:eek:

[breguet]

I have flown Boeing and Airbus and, I do not care if the Thrust levers move or not as long as I see the N1s and EPRs in my normal scan.

Flying a FBW aircraft, is exactly the same as a conventional aircraft. And nothing prevent the pilot to disconnect the AP ( and the Autothrust) and fly a manual approach with the FD Off, unless the Cie decides that AP, FD and A/T shall be On all the time which is not a good idea because it prevents pilots to maintain their manual skills...

When the approach is "rough" due to high winds and turbulence, I disconnect A/T and AP and fly it maually it is a lot easier, in my view, because you have a better control on speed.

[parabellum]

My first jet was the BAC1-11 and following the advice of an old but not bold pilot I always had a copy of that page of the ops manual that showed the table for flight with unreliable airspeed on my clipboard. Carried it through until my final flight on the -400 and because I looked at it fairly often, it being constantly 'in view', I subconsciously committed just a few of the speed, body angle and power settings to memory, it can't hurt.

I still maintain that a large part of the current problem with automation is that it is not taught and practiced sufficiently to the point that pilots are competent and confident to utilise the automation throughout its envelope; that is a management and training issue and in almost every case will come back to a budget issue.

[bubbers44]

When automation works it is fantastic. When it quits working turn it off because it can hurt you. Manual flying skills should still be required for all airline pilots. Airlines that rely on automation for 250 hr pilots to fly passengers is crazy. Simple loss of airspeed procedures for all airliners gave a chart for weight, altitude, attitude and power to survive. Air France put their flight into the Atlantic with their Airbus recently. Did they have that training? Why do we have to dumb down our procedures to hire unqualified junior pilots. We have plenty of qualified pilots in the US to fill these positions but for a reasonable salary. We are not short of pilots, just short of airlines that will pay them what they should. I only know about the US so hope European pilots are equally qualified.

[bubbers44]

I'm talking about the new hire FO's. Not captains.

[bugg smasher]

through until my final flight on the -400


Staring that right in the face now P, any words of wisdom as I take my final stroll, shortly, for the last time, off the mega-ship flight deck that has been my home, my light, and my reason for living, forever and always?

This is not good, nobody ever told me it would end, I am devastated...

[protectthehornet]

bubbers...everyone has more time to type because they are so autopilot dependent.

I know you know what flying is all about...I mean it.

BUT I really miss the pride one took in their hand flying and while hand flying...they managed to navigate and even eat a ham sandwich.

sheesh...

[parabellum]

Try and enjoy your final trip as much as possible, I picked a four sector SIN-AMS-JFK-AMS-SIN for my final run. Chatted up the hotels in advance and managed an even more superior room. Unbeknown to me my wife arranged a good photographer with airside clearence to do the whole caboodle in AMS, still photography. Check-in, briefing, the final walk down the jetway, external, flight deck, a/c departing so I have a permanent record. Once retired stay involved if you can, airshows etc. and other volunteer work, stay busy.

Sorry, considerable thread drift!:O

[Centaurus]

I don't think this is an "over reliance on automation" issue.
Maybe, maybe not. But when a first officer on a sophisticated jet transport has never done a profile descent from top of descent using DME versus height, then that is because of over-reliance on automation.

[Piltdown Man]

No, that's called poor training and standards.

PM

[BOAC]

No, that's called poor training and standards. - isn't that what this is all about? 'Over-reliance on automation' - just WHERE do you think that comes from then?

[stepwilk]

"Once retired stay involved if you can, airshows etc. and other volunteer work, stay busy."

Unashamed plug, but I stay involved by volunteering as a youth-mentor member of the Tuskegee Airmen Inc.'s Red Tail aviation program. Look into it. And yes, I'm Caucasian.

[ExSp33db1rd]

I don't disagree with any of the above - even the Last Trip advice !! - but let's not forget that every generation has it's own problems to address, in ways we 'oldies' can only imagine now.

I was once told that I would never make a navigator until I'd been over Berlin with the shells coming through the cockpit as I tried to get 3 different drift readings from the drift sight as the pilot tried to keep the aircraft steady on heading, to assess the wind for the flight home.

I never had to do that.

Later I despaired of my own nav. students wrestling with the sextant, and Grid navigation, over the Pole.

They never had to do that.

Along came INS, and then FMS and now GPS, they don't even have to fight HF radio - do they ? ( Delhi, Delhi, this is Bombay - Shut Up Delhi ! )

I blame Bill Gates, life was easier before computers - even in the shops !!( at least people could do mental arithmetic )

[Metro man]

I bet most of you lot never had to hand swing a propellor either. ;)

[protectthehornet]

hey metroman...do they call you, "Lefty"? (just kidding).

everyone should be able to fall back to a previous method or technology...TCAS out? look out the damn window.

even lindbergh lost his compass system at one point...so he put the north star above his left shoulder and still found Dingle Bay ok!

[pool]

Interesting thread, interesting arguments.

I go with what I have experienced the last few years. Some examples:

- In our big company there is not a weekly incident report without unstabilized approaches. The overwhelming reason given is "short cuts and short vectors". Now either there is a lack of capability to calculate distance vs. altitude or a lack of capability to fly the aircraft down slightly higher angles. Mostly both.
- When a AT goes u/s, the approaches often go to at least "uncomfortable". A frenzied scan between the speed on the PFD and the N1/EPR on the engine display leads to pitches between 0°and 8°, bank of more than 10° l/r on the final ILS stage.
- Visual approaches and circlings are treated as demanding procedures (when I started aviation they were the more simple approaches than CatII ...). I observe an abundance of special techniques with the basically obsolete FMC, like drawing 3nm circles, abeam points, altitude reminders etc. Additionally there is a teaching to use/program the FD throughout the whole procedure. Although it might help a bit, people often try to fly it more than looking outside. All this ends in an over reliance on such reminders and on the FD and consequentially a loss of general overview of the simple approach happens.
- The AP is mostly switched off at ~500ft. Initially the FD is religiously followed, but soon it is slightly lost, ~200ft, because the lookout only starts there. The aiming point gets lost down the runway, with very little corrective action, as the "feeling" for the aircraft could never be established in such a short and already fully configured time. The landings are often smooth, but seem somewhat uncontrolled and mainly long.

I do not want to venture into discussing why, just a personal observation to sustain the dire need for attention to the issue of the thread, the issue of automation correlated to the loss of skill.

[FlightDetent]

- isn't that what this is all about? 'Over-reliance on automation' - just WHERE do you think that comes from then?

Ok, I'll bite the bait, with tongue in cheek and a twink in the eye.

Are you implying, that the "magenta FMS whizz no-pants-in-seat" set is actually not a legion of underachieveing brats but rather a product of a generation of greyhair examiners and course developers who (under pressure to reach the required output levels and cost competetivness) took the easy way out. Using automation without need to explain what it does and what never was intended to do? :)

Yours,
FD (the un-real)

[A37575]

In Australia, a regional operator flying Saab 340 "strongly recommends" (my italics) the autopilot be engaged at 200 feet (min engagement height) after take off and disengaged at 50 feet on landing. This is taking use of automation to a ridiculous degree and does nothing for flight safety. On the contrary. As this airline specialise in training its own cadets from zero hours to right seat of the Saab, it is living proof that manual flying is seen by this operator as an unnecessary impost. Despite numerous warnings from international flight safety authorities that automation dependancy is an increasing concern, clearly some operators don't see it that way.

The local regulator for this region is not interested and in fact has indicated it favours more use and understanding of automation. Nothing at all from these hallowed halls about the need for manual flight competency :mad:

[RVF750]

This thread can go on forever....

The FMS magenta track line makes npa's utterly piss-easy to fly manually, it really riles me that anyone can cock these up with a damn autopilot engaged too.

As mentioned above, it is horses for courses, we fly automated airl;iners nowadays and they are designed to fly through the automatics. flight Managers look at FDA stats and draw obvious conclusions- more "events" occur when hand flying or with automation turned off. They have KPIs to lower "events" and logic tells them to stop people hand flying to lower the events.

Counter logic say to make folk hand fly more- but would lawyers agree? This is the quandry the Managers face. Don't forget they might only fly one day a month themselves, with F/Os terrrified to be stuck in a flightdeck with them and perrforming accordingly.....

Hmm.

[BOAC]

Ok, I'll bite the bait, with tongue in cheek and a twink in the eye. - the short answer to your post (TIC) is 'Yes'.

In my book, the "generation of greyhair examiners and course developers" should take on the task (- and I'm sure the good ones do) of ensuring that the "legion of underachieveing brats" is made to understand basic flying, including how to plan and fly a descent without a computer.

[JamesT73J]

I suppose it's a case of 80/20, and this pervades a huge amount of modern education. Background information is pared-down to a minimum, and if the end goal - line first officers - is all that matters, then hard faith in the automatics makes absolute sense, if only from a management point of view. Of course, the random nature of the real world periodically cares little for stats, modelling, or theory, and this is where man still has a big role to play when something out of the ordinary happens.

I think automation is amazing, but I've never flown anything larger than a C150. I still marvel at the precision and grace of modern airliners. Even as passengers, we have become disconnected from the experience of flight; a recent trip on a 777 was more like a magic carpet than an aircraft. A quick look around the cabin during takeoff will show that most people won't even flinch at being recipients of over 100 years of knowledge and skill that can propel a 350 tonne object into the air, and 10000 miles away in a few hours, and repeat the trick endlessly. Familiarity is the root of indifference.

And yet...

The laws of physics are unchanged, and indifferent. The same information is being presented in the same form (more or less) to the pilot, and Airbus say their products fly just the same as anything else. Is there anything really new here?

[RAT 5]

In my book, the "generation of greyhair examiners and course developers" should take on the task (- and I'm sure the good ones do) of ensuring that the "legion of underachieveing brats" is made to understand basic flying, including how to plan and fly a descent without a computer.

It would be a great pleasure to do so,.....but...SOP's preclude it, or management attitudes do. Further, if you tried, on the line, to instill some semblance of what is being discussed, it would take a very long time as the foundations have not been laid in the TQ training. That has been minimised and diluted as a cost saving measure. Students paying for themselves can not afford the 3 or 4 extra sim sessions which should be there for basic handling skills. Companies paying for the training don't want to do more than required. They seem content to have robotic trained monkeys up front. Pilots who can not be trusted to fly visual approaches without significant LNAV VNAV backup. I always thought the most useful tools in a visual approach were Mk.1 eyeball and a consumate understanding of pitch & power. Someone said the physics of a jet and a Cessna are similar. Correct. Give the beast the correct amount of energy and it will do what you want; give it more or less and it will do something surprising. That means you're not in control, which being the PF is rather disappointing. George can do a wonderful job; it can also bite your rear-end. So to follow the suggestion of teaching raw data descents to a descending low drag visual circuit as the norm, which it was in the clock & dials day, would be a real thrill and frowned upon very heavily. It'd take a full summer season. I hear from FDM experts that in CAVOK most G/A's follow screwed up visuals. Amazing. Solution? discourage the troops from doing them as a time/fuel saving method because if they mess it up any saving has been reversed. Encourage robtic techniques. Trouble then is when command upgrade time comes, the TRE expects to see some initiative, strong situational awareness, ability to multi-task in non-normal scenarios. If the basic foundations are not there to let you relax and manage the a/c too much energy and capacity is used during the tasks and it all breaks down.
I fly with F/O's who at 26 have 3000hrs and are expecting a command shot soon. How do % fail rates compare between companies? Long & short-haul. But also compare this to the era of 7-10 years in a company, 5000hours and the likely hood to have seen and learnt from some grey haired old farts during the various interesting scenarios that inevitably threw themselves in your face during your apprenticeship. Now, there seems an assumption that 4 years, 3000hrs, a couple of OK prof checks and it's time for 4 stripes. I find that some SFO's coming up for consideration, are still reluctant to make the first call when PF. I always throw the ball into their court, but they still ask what "I'd like to do." Your choice, mate; I'll tell you if I disagree. That attitude is not an option on a command course.
I did hear 1 chief TC say he would like to teach better handling techniques, but there is no time. With companies having bases all over the world it is very difficult to keep tabs on standards. So, dum it down to what you know works; use AFDS as amuch as possible; write a detailed SOP book and hope everyone plays the game. I can understand that solution; it just very soul destroying. I learnt BOAC's quoted techniques on the line. It was the way everyone flew, and most were good at. They'd all come through he same mill. Shame times have changed.
It will be interesting, but perhaps not possible, to hear what TC's in 10 years are saying. They will have come from this generation of teachings and be passing it on. I wonder what they will find on the command courses in the future. I wonder if the checking syllabus will have changed. In 15 years the CAA's will be staffed by people from todays teaching regieme. Will they change anything?

[BOAC]

Rat - a fine summary of where we are and where we are going.

[fantom]

We are already there and it is far too late to turn the boat around. We have seen it coming for some time.

Those who decry automation need an answer and there will be none forthcoming.

BOAC and I (and others) suffered the trials of selection; now, a credit card will do...

[BOAC]

the trials of selection; now, a credit card will do... - at least the day the c/card buys someone into the RAF/FAA/AAC is some way off.

[stepwilk]

Not long ago, on this forum, I took polite exception to an apparently professional pilot's joking remark that to go up, you pull the stick back and to go down you pushed it forward. What's so hard about that?

I don't think he had any idea what I was talking about when I suggested that the throttle/power lever was actually the up/down control. Several others jumped in to complain that I was talking nonsense. "Try telling that to somebody taking off," one snarked.

[aterpster]

stepwilk:

I don't think he had any idea what I was talking about when I suggested that the throttle/power lever was actually the up/down control. Several others jumped in to complain that I was talking nonsense. "Try telling that to somebody taking off," one snarked.

I don't either. The arguments are endless.

[stepwilk]

Try staying on a glide slope with the yoke alone. That's an argument I'll be delighted to watch.

Certainly there was no argument in my day, and in the 1960s and '70s, that's how we were taught. Have the physics changed?

[737Jock]

Stepwilk in an airbus it is exactly that: to go up you pull the stick to go down you push, the thrust levers don't move and stay in the climb detent.
Flying for dummies.

Offcourse in theory you are absolutely correct, it's excess thrust that enables an aircraft to climb or lack of thrust that makes it go down.

Another weird fact about airbus, even with manual thrust you still need to pull the stick to climb and push to descent. Pitch/power coupling is "prevented" by autotrim. Leaving the stick in neutral but playing with thrust levers would just result in speed changes.
This is why I find the airbus is a bitch to fly in turbulent gusty weather.

The airbus is simply a point the nose at where you want to go aircraft.

[PBL]

Here we go again! Stepwilk and Terpster, hang on to your hats!

It's not particularly physics, it's primarily a culture thing.

John Farley said it 6 months ago in http://www.pprune.org/tech-log/419924-aoa-instrumentation-2.html#post5791913.

The discussion goes on from there. Hurt's Aerodynamics for Naval Aviators is quite clear on the theme: Angle of attack is the primary control of airspeed in steady flight but his examples are unconvincing.

The discussion was not long, and was inconclusive, because to resolve it seems to require numbers which nobody was providing. But the issues are by no means simple.

PBL

[aterpster]

stepwilk:

Try staying on a glide slope with the yoke alone. That's an argument I'll be delighted to watch.

How about auto-throttles engaged but auto-flight not?

[G-SPOTs Lost]

So are you lot saying that an apprenticeship is now whats required to enhance safety?

Problem being airlines wont hire from airtaxi / corporate / jet charter as it doesn't fit their business model, they need to start doing to release that position for somebody else to train up.

You'll be a way up the corporate type rating ladder before you get to see an aircraft with VNAV and autothrottle - taken me 10 years! It sort of forces the issue regarding lack of skill as there is nothing to rely on.

I've always said that airline SOP's are not there just for safety's sake but also to account for the lowest common denominator in terms of ability - it appears they have stealthily changed emphasis from the former to the latter.

The great shame is that it will never change going forward.... you grandfathers of the green are wasting your breath, Airlines simply need more children of the magenta lest somebody doesn't get to AGP for a tenner.

What is interesting is that all the loco models are based on the Southwest model who don't entertain cadets or low time captains - we've nobody to blame but ourselves.

Get CTC to approach some Chieftain and Kingair operators - would the newbies cope....I'm not sure they would, they need the rigid SOP's to make it to the line.

[BOAC]

A simple plea! This thread is about 'Automation' - not the age old stick/throttle discussion.

Before we are all dragged screaming off into pedants' corner, and personal insults start hurtling around yet again, can those arguing about this please **** off somewhere else? :mad:

How about 'Private flying', or 'Flying Instructors' or 'Professional Pilot Training'? Plenty of choice.

[RAT 5]

BOAC: back on track indeed. The basic tests, LST & LPC, are ancient in todays world. We are allowed, encouraged, to use automatics for an NPA. Not to have 1 available would be a multi-failure on a modern jet. However, the LPC test manoeuvres are similar for many types. Do all smaller pax a/c have multi autopilots? 1 broken = manual approaches.
The only manual manoeuvre is the SE work. That worries me. It is negative training. In a multi-autopilot a/c, with an engine failure, why not fly the approach with the A/P. Managing the automatics in the non-normal scenario is what the real world needs. Are we trained for it, or could mis-managing the auto's be the final piece of cheese that lines up the holes? If we decide that is the way to go it has to come from the top = CAA's. The other solution, if things stay the same, is for companies to give more training/recurrency time in use of automatics in some very strange circumstances. Real training to a proficiency, not a fast pass at 1 secenario quickly followed by another and another, then a tick in the box that recurrency has been satisfied. The airline is responsible for ensuring the OVERALL proficiency of its PILOTS.
If the way forward in testing is to use automatics the way we should in the real world during NNC's, then the opposite should be true and companies would give extra time to develop manual flying skills. Same reasoning as above. In both cases it will cost the airlines time/money so it won't happen.
What about an LPC using automatics and an OPC using manual skills? I still believe the pax expect us to be a safe insurance and get them home safely when the sparks fly; or don't.
Further, given that we seem to agree the future is towards having system operators and less pilots, I'd like to see a complete re-think of airline's personal charateristics for student pilots. They ask for people who will be very over-educated for the tasks involved and probably of different motivation for the future life styles that they will inherit. Causes the wingeing syndrome we are so familiar with. Wrong type of people in wrong places. During an interview for Air Heathrow I was told they did not recruit F/O's, but captains, and not even captains but future managers. Pilot was not in the vocabulary. OK a A380/B747 captain might be more of a cruise liner captain personna than a LoCo B738 captain. Indeed it does require a different personna. Do airlines recruit for that? Not in my opinion. Another major carrier told me they did not require more than 1 experienced pilot on board and that captains flew the a/c and station managers made all other decisions. With ACARS & SAT phones they could always ask for help. Again, a very different enviroment than the F50/EMB135 6 sector a day pilot. Different horses for courses and different training worlds. Some very automated, others verfy basic. The Highlands & Islands type flying needs a different animal to the A380 pilot.
Is there any motivation to change the aviation world, anywhere? I find it sad when I read an airline does not recruit experienced F/O's or non-type rated DEC's. Could this the start of a trend towards cloning their own monkeys? Easier and cheaper, or is it in fact the modern version of all the major carriers (EU) who used to have their own training schools for raw recruits. get 'em young and mold into the way of our church. No bad habits. Brain washed in our mehtods. I still meet ex-major pilots who've spent their whole lives in one cosy nest and believe there is only one way to to do it, and their's is the best.
How long before ATC start flying the a/c and we do really become system monitors? They like to do it now by command. It's worse than being married. Do this, do that and no back chat. Trouble is that sometimes they ask for something which is not appropriate, but I still see Topgun over there get sucked into it and start to wonder if it'll work out OK. There is a hesitation to say, " I, the responsible commander, would prefer to do something else, as from where I sit this is not going to work out comfortably." That's a whole new thread and I'll leave that for another time.

[A37575]

In a multi-autopilot a/c, with an engine failure, why not fly the approach with the A/P.

But in the 737 Classic, the autopilot-flown single engine instrument approach, is neither fish nor fowl. The remaining throttle is operated manually and so are the rudders; while the ailerons and elevators and stab trim are on autopilot. In the simulator for example, we see the situation where the pilot changes the thrust manually to correct a speed trend. In turn, he then has to adjust the rudder position to correct for yaw.

Unless the correct amount of corrective rudder is fed in to counter yaw, the control wheel then turns automatically to hold the ILS localiser. This causes inevitable flight spoiler operation and thus unwanted drag. The whole event can then turn into what is beautifully described as a horses's arse. And that is because of crossed controls.

Even on a single engine autopilot flown circuit, it is not uncommon to see the aircraft turning (say) on to base leg with the PF using the heading select function of the autopilot. Meaning the control wheel turns but if yaw is allowed to occur due to incorrect rudder input it becomes quite interesting to see the control wheel turning in an opposing direction. Aircraft turning left under yaw and wheel turning right trying to correct the yaw due incorrect rudder pressure.

On the other hand a single engine circuit or single engine ILS flown manually throughout, presents no problem to a current and competent pilot because there is no mixing of manual input and autopilot input.

Personal opinion only, of course

[d105]

With regards to the "stay on glideslope without throttles" remark.

We are talking about the last 500 feet I presume? At my base the aim is full idle approaches until approaching the landing gate.

[Melax]

The accident report of the December 1, 1974, crash of a NWA 727 in New York should be required reading for all of today's active crewmembers. They forgot to turn on the pitot heaters and when "the stuff hit the fan" they got fixated on the bad information to the exclusion of attitude instrument flying. From ATERPSTER (Post # 4)
The accident report can be downloaded here: http://www.fss.aero/accident-reports/dvdfiles/US/1974-12-01-US.pdf

I had an eerie feeling when I looked up the crash site on GOOGLE Earth, there is still a large clear area...
http://img842.imageshack.us/img842/1559/nwa6231crashsite.jpg


Makes you more humble, there is absolutely no substitute for the basics.

[JamesT73J]

Makes you more humble, there is absolutely no substitute for the basics

Chilling reading how the picture can change. The comment that the elevator feel units were spoofed by the incorrect ASI readout and allowed +5G acceleration to be achieved with only a modest pull on the control column.

[RAT 5]

A37575: My point exactly. "Unless the correct amount of corrective rudder is fed in to counter yaw, the control wheel then turns automatically to hold the ILS localiser. This causes inevitable flight spoiler operation and thus unwanted drag. The whole event can then turn into what is beautifully described as a horses's arse."

And that is because of crossed controls.Teach the drivers to do it properly. Allowing cross controls during autopilt useage is bad piloting; it is not a reason not to use all aids at your disposal. The use of automatics is to reduce the workload and improve oversight of the operation. This is true in normal & non-normal ops. If the automatics are doing a bad job it does not mean it is the fault of the automatics.
For a SE ILS, if the rudder is trimmed correctly it will be the same for all configurations except when reducing thrust to slow down for more flaps. A little opposite rudder for a few seconds should not be beyond the wit of most. Thrust control is no more difficult than a manual approach. Mis applied rudder with autopilot on is obvious by looking at the aileron defelction.

Your comment about turning base I see all the time, but that is because people extend flaps in turns. Reduce thrust with no rudder input. If you increase thrust 3% then turn there is no need for rudder input and everything is smooth. SE turns in constant config is highly recommended and SOooo much easier.

It reminds of 1 airline I flew for on the classic years ago. In London TMA there were many step climbs of 1-2000'. On B767, years before, they had introduced a PIP software which, in LVL CHG gave a gentle climb for such small adjustments. The classic just went full power and screamed up. Those of us 'in the know' used V/S 1000fpm and taught it. The HOT did not like this as many pilots did not know the gotchas of V/S. He banned it and insisted on KISS, the use of uncomfortable LVL CHG. Guess what; with TCAS, and RVSM, it is now Eurocontrol SOP. So rather than train guys how to do it properly it was banned. Is that the correct attitude? Know what does what and know how to use it, then choose the best. Ignorance will bite you hard. I still see guys playing the piano on the AFDS trying to find the correct button to make happen what they want. AND they still look at the button and NOT the FMA. Detailed good complete training will always go a long way. It's the diluted version which will cause problems later.

[blimey]

Airbus manual flying in normal law relies heavily on automatics, and the switches also known as the thrust levers give no positional feedback when used in auto. When things go awry, therefore, it’s quite a big jump in workload to fly the aircraft in alternate or direct having only practiced what is essentially controlled stick steering for normal manual flying; and putting the control levers back to the approximate position they were in prior to disengagement is a non starter. It must have all looked so splendid and modern on the designers' desk. Hey ho.

Just back to edit a typo, and for d105 below, not sure an airbus would suit you!

[d105]

Out of interest. Which 'gotcha's' could occur using V/S in a 1000 feet step climb?

From your post I conclude both A/T and A/P were being used as you state LVL CHG would have the aircraft push it to full thrust.

So with both those engaged I fail to see what could happen in a 1000 fpm step climb. Unless you somehow start to retract your flaps in that climb without accelerating to up speed first.

Unrelated question for the Airbus drivers. When you fly your aircraft manually is there a difference in responsiveness in relation to speed?
On the 737 when you descent at 330 KIAS while flying manual the slightest movement on the yoke has the aircraft reacting to it immediately. As you slow down to UP speed the reaction of the aircraft to your inputs becomes more sluggish.

Is this simulated in the Airbus joystick? Or does the joystick feel the same at VMO and VCLEAN ?

Control feel is personally my first indicator something is wrong. If I would feel the aircraft becoming heavy in the nose but my trim is near 5° and my ASI indicates 320KIAS, then I know something is up.

[alf5071h]

The issue appears to be automation, but more so, advanced technology - automatics only provide muscle for some applications.
A view from previous discussions was that we have to live with ‘it’ (I stress ‘it’ as automation/technology); we have to get used to it, it’s a fact of modern operations. Some people are managing this, perhaps some not, and for the majority there is an uncomfortable in-between state with many pitfalls, often creating a feeling of ‘not being in control’.

Many safety problems involve ‘technological malfunction’; this is rarely due to a system fault, but more likely aspects of human interaction – not understanding the situation (including the technology) or not choosing an appropriate course of action.
As much as we think that we understand technology, most initial training only covers the basics of normal operations without serious ‘malfunction’, and don’t forget a ‘self’ malfunction.
Current automation adds complexity which furthers the need for evaluation, understanding, and choice; this requires greater operating knowledge which may only be acquired ‘on the job’.

Are we – the industry and individually, learning ‘on the job’? Is there encouragement (there’s no requirement), opportunity / time, personal willingness, or guidance / instruction available for this?
IMHO we are failing to learn about operating automation. Where we do learn, and a lot is accomplished in operation, it may not be directly relevant to the situations we face, or we are not learning in-depth to satisfy the needs of safety.

With modern automation/technology we appear to require more time to acquire knowledge in order to operate in the same manner as previously.
Thus calls for more training; but alternatives could be to change the operational environment, or the nature of the operation. The most likely solution is a combination of both, involving new ways at looking at training, operations, and particularly ourselves.

Many of the answers are within the ‘bleats’ about auto/manual flight. Stop and think about what ‘you’ can do about ‘it’, think about operating without technology, the basics of assessing the situation, understanding and planning without an ND / FMS. Prepare yourself for the unexpected - how you would manage the aircraft in a LHR hold without FMS / autos; it’s not just stick and throttle flying. Then with these skills automation might appear more logical.

“Give a starving man food today and he will be hungry tomorrow, teach a man to farm and he will feed himself” - - - Teach a pilot to operate / manage (think) as well as handling skills, then you provide a basis for future safety, new captains, and instructors.
Yes – there’s a gap between effecting any change and the reality of now – we have to ‘think’ about that ourselves – how do we operate and what do we know – or don’t know (what and how), and what are we going to do about that – professional self improvement. The industry won’t do that for you, these qualities cannot be regulated or inspected; they have to come from within.

[Diversification]

Over a year ago Airbus offered the airlines to buy an extension to their software that would help the pilots in situations where airspeed were unreliable. The software add-on was called BUSS. BackuUP Speed Scale, that replaces the pitch and thrust table. BUSS is said to be optional on A320, A330 and A340 but included in the basic software for A380 as part of the ADR monitoring functions.

[Sub Orbital]

D105
If you don't know the "gotcha's" with regards to using V/S in a step climb, then I never, ever want to be on your aircraft.

[Oakape]

But in the 737 Classic, the autopilot-flown single engine instrument approach, is neither fish nor fowl. The remaining throttle is operated manually and so are the rudders; while the ailerons and elevators and stab trim are on autopilot. In the simulator for example, we see the situation where the pilot changes the thrust manually to correct a speed trend. In turn, he then has to adjust the rudder position to correct for yaw.

Unless the correct amount of corrective rudder is fed in to counter yaw, the control wheel then turns automatically to hold the ILS localiser. This causes inevitable flight spoiler operation and thus unwanted drag. The whole event can then turn into what is beautifully described as a horses's arse. And that is because of crossed controls.




The 737 is perfect for picking up deficiencies in knowledge/training/skill due to this.

Most of the problems I have observed in the sim seemed to be due to pilots treating the autopilot as a 'pilot', rather than a tool to assist. This comes about for a number of reasons including laziness, lack of knowledge of A/P limitations, being overloaded, lack of adequate training or simply, incompetence - they need the A/P to do it for them because they can't do it themselves.

Once they have engaged the autopilot, some pilots seem to be under the delusion that they don't have to fly the aircraft anymore because they have handed control over to another 'pilot'. They then spend their time focusing on checklists & trying to make decisions, while the ailerons are getting further & further towards the stops &/or the speed is rapidly decreasing.

What you usually end up with is the aircraft turning in the oposite direction to what is commanded by the heading bug & the pilot then saying "what's it doing?" & "the autopilot isn't working properly". He/she then disconnects it, which usually leads to immediate flight path control problems, particularly if the speed has become dangerously low.


Know what does what and know how to use it, then choose the best. Ignorance will bite you hard. I still see guys playing the piano on the AFDS trying to find the correct button to make happen what they want. AND they still look at the button and NOT the FMA.


Absolutely correct RAT 5! Most pilots don't seem to realise that pushing the button on the MCP is the third step in the process, not the first. You have to first decide what you want or need to achieve, then work out how you are going to use the various A/P modes to acomplish that, & only then, do you move to the MCP to engage the required mode.

And I agree, you must always, always, always check the FMA. Pushing the button is what you want; what is on the FMA is what you have got.

Again it's lack of knowledge of the system &/or lack of training in how to use it properly.

As for V/S, a lot of pilots seem to be either unaware of the appropriate time to use it, or scared of using it at all. Also, Flight Level Change seems to be the only mode some feel comfortable using during descent, even to the extent of using it instead of VNAV Speed when something has changed the descent profile. They don't seem to grasp that, put simply, VNAV Speed is Flight Level Change with some of the protections of VNAV path.

[rahul_mica]

CMD , VOR/LOC, 110.30 will in all probability take you down the centerline of ONE particular runway and preferably the one the ATC wants you to be on.

AP off, AT off, FD off :: you MIGHT bring in both the parallel runways, both the parallel taxiways the next door Airforce base and a local highway into the equation.
Now dont tell me that hasnt happened before! :\

[A37575]

AP off, AT off, FD off :: you MIGHT bring in both the parallel runways, both the parallel taxiways the next door Airforce base and a local highway into the equation.
Now dont tell me that hasnt happened before!

Only if you lack situational awareness. Only you will know that - not the auto pilot....

[sevenstrokeroll]

A3757 is quite right. Situational awareness is key. I wonder how many ''automated'' pilots bother to identify the localizer frequency. Garbage in Garbage out.

Oh, and that NW Airbus that overflew Minneapolis...it didn't take care of itself.

I do think one problem with autopilot use, especially in emergency ops in simulator training is this. .Union contractual language doesn't allow for compounding emergencies during check rides. Lose an engine and they can't fail the autopilot on you...in the sim. But in real life, well that's different.

[d105]

Well Sun Orbital, how about you enlighten me than?

Perhaps I'm not as schooled in use of V/S for a 1000ft step climb as you. Maybe that's due to the fact that, when I am in command, the aircraft will actually be under manual control ;)

A37: If you need to Autopilot to perform a basic flight manoeuvre you should be stripped of your license.

[sevenstrokeroll]

I can only imagine that the GOTCHAS in using vertical speed in a climb is that if you select a vertical speed beyond the energy capabilities/ceiling capabilities/ that you might stall.

I seem to recall a Mexicana DC10 that stalled due to inattention in this mode...is that what you meant?

of course selecting mach hold/IAS hold can result in a climb that is protected against a stall, but the nose can ''hunt'' if there are airspeed vagueries.

[aterpster]

sevenstrokeroll:

I wonder how many ''automated'' pilots bother to identify the localizer frequency. Garbage in Garbage out.

I haven't flown the state-of-the-art avionics. I know they display the identifier of the selected LOC or VOR. If that is displayed only after the avionics have "listened" to the identifier, then wouldn't that make it unnecessary for the "automated pilot" to take the time (and attention diversion) to concentrate on the Morse Code audio?

[misd-agin]

757/767 do not automatically display left and right ILS frequency on the 'map' mode. You set the center ILS and it tunes the left and right ILS. It's up to each pilot to then switch their nav display to the ILS mode to verify the correct frequency for their respective ILS.

Randomly the center ILS will not correct tune the left or right ILS. It's a problem with what the center ILS is displaying for a frequency vs. what is actually being tuned. You then have to adjust the center ILS frequency selector until the left and right ILS's have the correct frequency.

SOP's cover this - tune center ILS, verify on respective nav display by selecting ILS mode, verify via morse code.

[Capn Bloggs]

AND they still look at the button and NOT the FMA.
Reminds me of a couple of classic phrases from my instructor on my first electric jet:

"The MCP is the Rumour Panel. The FMA is the Truth panel."

and, regarding Magenta/Managed vs White/Selected:

"White Rules"

(with an extra bit of irony: he was South African.)

[d105]

I can only imagine that the GOTCHAS in using vertical speed in a climb is that if you select a vertical speed beyond the energy capabilities/ceiling capabilities/ that you might stall.

I seem to recall a Mexicana DC10 that stalled due to inattention in this mode...is that what you meant?

of course selecting mach hold/IAS hold can result in a climb that is protected against a stall, but the nose can ''hunt'' if there are airspeed vagueries.

This makes absolutely no sense in the case we are discussing. We were talking about "dangers" involved in utilising V/S mode to attain a 1000ft. step-climb in busy TMA's or during lower level climb-out.

I can only speak for the 737, but I fail to see how you would be able to put the aircraft into approach to stall while performing a 1000FPM step-climb with all automatics engaged without any malicious intend in the situations mentioned above.

[sevenstrokeroll]

d105



you fail to see...how about if an engine flamed out? or if the autothrottles were not working properly...

do you remember a 747 in altitude hold that lost one engine...slowed down, stalled, piece broke off, a safe landing at KSFO was accomplished...possibly went supersonic, possible spin

hey man, stuff happens

[bubbers44]

I saw that China Airlines 747 sitting at SFO for some time and wondered how the pilots could watch it happen without intervening. They must have been aware they lost an engine and needed a lower altitude.

[d105]

Sevenstroke,

You are joking? You are seriously suggesting that after engaging the aircraft into a 1000fpm step-climb both pilots would not even notice an engine flaming out or the A/T not moving with reducing speed?

Seriously? Because in those cases it wouldn't even matter if it were the A/P or the actual pilots flying that particular aircraft. It was doomed from the start.

My question still stands thus.

[sevenstrokeroll]

bubbers 44...I saw it at SFO too. WOW. how did the tail stay on??????

D105...I wish I were joking...but it really happened to this 747 crew. I'm sure you have heard of murphy's law...if it CAN happen, it WILL happen.

[DC10 Fever]

One poster has already mentioned Aeroperu 603

Aeroperú Flight 603 - Wikipedia, the free encyclopedia (http://en.wikipedia.org/wiki/Aeroper%C3%BA_Flight_603)

Although incredibly unfortunate, there is a transcript of the CVR where the Captain is clearly struggling, with the only sense coming from the FO.

The Captain's persistent intention of getting the AP back on is frightening when you're only getting erroneous readings back.

While near-invaluable nowadays, aircraft 'intelligence' is never going to be 100% accurate... ever! That's what the boys up front are there for aren't they ;)

As mentioned earlier there's a reliance on AP/AT/FMC among newer boys/girls, not necessarily with blame attributable to them, but basic understanding of concept seems to be overlooked.

[bubbers44]

As I recall they were level in altitude hold and an engine flamed out. A descent, clearance or not, is the only way to maintain control at that altitude. Aviate, navigate and communicate is what I was always taught. They missed the aviate part big time. I agree, it is amazing the tail stayed on after what they did to that 747. How do airline pilots do that? At least the old pilots.

[misd-agin]

So if they were climbing in VNAV PTH it would protect(assist) them from an engine failure ??? := :ugh:

[PJ2]

d105;

Read it, and weep for professionalism: China Airlines B747 SP Accident Report (http://www.rvs.uni-bielefeld.de/publications/Incidents/DOCS/ComAndRep/ChinaAir/AAR8603.html)

By your response I suspect you have tons of experience and a lot of wisdom and run a good operation. You seem however, again at least by your response, somewhat incredulous that these things can and do occur in professional commercial operations. If I may respectfully offer the suggestion that regularly reading accident reports brings the experience of others who were there, right to your front door and perhaps into mind when the next unexpected event occurs in your own operation. Just a thought.

bubbers44;
How do airline pilots do that?
Exactly.

How do experienced veteran crews, with a third crew member riding specifically there in the cockpit to keep an eye on the guages while the captain is training the other guy in the right seat, fail to maintain airspeed letting it bleed off 40kts over 100 seconds, and stall their aircraft at 400', killing them and six others?

PJ2

[d105]

I'm by far an experienced captain PJ2. But perhaps it is because I haven't just yet come to the point where I feel the left chair is more comfortable than my own bed that I find missing such things to be unbelievable.

I do admit not reading nearly enough accident reports and have bookmarked your link for the near future :)

@misd-again: Not sure if your reply was pointed at me. But to be clear, we're not discussing VNAV.

[bubbers44]

PJ2, I don't know but if the right seater had his head up his a** why didn't the captain have his head out of his? I have never had a jump seater making sure the check airman didn't screw up too. Where does the line stop?

[BOAC]

I have never had a jump seater making sure the check airman didn't screw up too - that is one of the main reasons there is a 'j/s' pilot! People being people makes it easy for a TC (yes, even a TC) to miss something vital. I saw one failure to set QNH for approach in my time - competent TC but high work load..

[PJ2]

bubbers44, BOAC;

The point is, (as I know you know), that things happen in airplanes...to vast majority of veteran crews with tons of experience, talent and a professional work ethic second to none, and to the few, of all ages, who have no idea what they don't know about airplanes and aviation and no business being in an airplane with others on board yet are there by circumstance.

It is better to be aware that truly astonishing events and actions, though rare, are possible even with the former kinds of crews described above.

Reading widely the accident reports helps bring a keener awareness into the cockpit, as does regularly communicated results of an airline's FOQA Program....heard it, seen it.

d105, thank you for taking my thought as intended - respectfully, as it was plain to see you were speaking from lots of command experience. A good site for such reports is JACDEC (http://www.jacdec.de/), the NTSB (http://www.ntsb.gov/aviation/aviation.htm) and the Boeing Statistical Summary of Jet Aircraft Accidents 1959 - 2009 (http://www.boeing.com/news/techissues/pdf/statsum.pdf) is a very good summary of this aspect of our work. I don't mean to overdo it, but it's just that there's a lot of good gen on the net.
PJ2

[Safety Concerns]

Both tragic and shocking

The autopilot effectively masked the approaching onset of the loss of control of the airplane

not on a boeing surely the report is wrong:eek::eek::eek::eek::eek:

[DozyWannabe]

NOTE : I'm pretty sure Safety Concerns has his tongue planted firmly in cheek, before any catcalling happens...

On a more serious note though, autopilots and flight management systems masking potential dangers is not peculiar to any model or manufacturer. Off the top of my head we have not only the aforementioned China Airlines 747SP incident, but also the fatal ATR icing accidents, where the AP kept the aircraft trimmed until it could no longer cope and handed a practically unflyable aircraft back to the pilots. That's before we get into the AA 757 at Cali and other losses of locational awareness.

The rule of thumb should always be that the AP is there so that you can disengage your hands from the controls while managing other aspects of the flight - one should *never* disengage one's brain from monitoring the situation, no matter how clever the automatics.

One of the reasons I've tried to keep my head down on this thread is that it is very much focused on the issue of automation. My contention has always been that automation is simply a tool, no matter how advanced it may be on any particular airframe - and as such cannot be responsible for good or for ill in it's own right. That said, any individual pilot who relies on the automation at the expense of their own situational awareness is making a rod for their own back at best and a noose for their own neck (and potentially hundreds of others) at worst. Any airline whose management knowingly reduces core competencies (i.e. How Aircraft Fly 101) prior to putting a newly-qualified pilot in the flight deck using modern automation as a reason for doing so should be considered criminally negligent.

Using Airbus as an example, the detractors on this forum like to use the words of the product evangelist at the time of the A320's launch as justification for their particular bugbear. But he was only one man with a specific job to do (i.e. talk up the accomplishments of the aircraft) - and the actual design of the system was thrashed out by hundreds of engineers, with many pilots either consulting or directly involved (one of whom was Gordon Corps, who I've never heard anyone on here say a bad word about). I would quite happily say that I don't think any of those involved would be anything less than horrified at the thought of automation being used as a sticking plaster to put pilots in the right seat before they were ready, and as a cost-saving measure even more so.

The Captain's persistent intention of getting the AP back on is frightening when you're only getting erroneous readings back.

I think it's a basic cognitive thing. I have no idea as to how much - if at all - this translates to the real world, but there is a perception that when solving a problem, it can be helpful to let "George" take it while a member of the crew performs troubleshooting. In this particular case the Captain doesn't make the mental connection that the autopilot can only work with the same erroneous readings that he's getting from his instruments - although the amount of stress that the situation exerted has to have been considerable.

Interestingly, this crash was fundamentally caused by the static ports being rendered inoperable by being covered with speed tape, but if you look at the CNN News link at the bottom of the Wikipedia page, you can see that the press at the time reported it as a computer fault. You can bet that many more people saw and internalised that erroneous information than the correct information that would have been released to much less fanfare several months down the line.

[A37575]

You are joking? You are seriously suggesting that after engaging the aircraft into a 1000fpm step-climb both pilots would not even notice an engine flaming out or the A/T not moving with reducing speed?

I believe him. In fact, if one could write a book on what amazing things instructors observe during simulator training I wonder how many people would think twice about flying as passengers.

Re the above comment "You are joking". One particular scene observed recently in the sim. Crew from north Asia on DME arc leading into an ILS. Left autothrottle developed a fault while both throttles at idle in descent to initial approach altitude for ILS. Normal flap extension followed by gear down at ILS capture on AP. Right autothrottle goes up to 80 percent N1 while left autothrottle stays at idle 30 percent N1.

Control wheel well over at 45 degree angle as AP attempts to hold LLZ and captain has both feet as normal on rudders since they are not AP controlled. Neither pilot twigs there is a serious problem rapidly developing. They are automatic dependant. After a short period the instructor has no choice but to point out a problem is developing and quick action is needed. The pilots were apparently quite blind to this.

The instructor points to the offset control wheel and immediately the captain looks in alarm at the perfectly serviceable closed left throttle and calls for the engine failure checklist.

As the F/O reaches down to find the QRH, the AP gives up in despair and disengages - allowing the aircraft to yaw and roll beyond 60 degrees. The nose drops to minus 10 degrees and soon after the GPWS sounds . The captain stares at the ADI dumbfoundly but takes no action to correct the throttles or to recover from the unusual attitude. His hands are still on his knees and he makes no effort to touch the controls as the aircraft continues to roll and dive.

The F/O meanwhile is heads down trying to find section 7 of the QRH that deals with engine failure checklists. He never sees what is going on in front of him because he has orders to read the engine failure checklist. And in that culture the copilot follows the captain's orders to death if necessary.

Never mind that within the next 30 seconds both engines will fail anyway as they bury themselves in the swamp below. As the aircraft goes into an ever tightening spiral dive from 1500 feet with both pilots on another planet, the instructor saves their face (literally) by pressing the freeze button on the instructor panel.

And you think the original contributor to this particular post is joking?

These things happen in the simulator a lot more than most pilots would ever know. Thank goodness for well maintained real aircraft where pure manual handling skills in IMC are rarely called for.:ugh:

There is no shortage of other similar scenes we see and these are often with very experienced captains at the controls. In every case it comes down to automatics dependancy to a stupid degree. The absence of manipulative skills is frankly bloody alarming.

[sevenstrokeroll]

I've seen too much myself...I am more in the loop when not using auto throttles.

Captain (me) on 737 climbing out of BWI with new copilot,I advise I am turning off the auto throttles and then call for climb power and start looking for called traffic as we continue the climb.

I look down a few seconds later and ask the copilot if he set climb power...he said the auto throttles did that.

I told him that I had disconnected the auto throttles and advised him of such shortly after takeoff.

he hadn't even looked down at the gauges to make sure that the engines were performing properly.

I felt badly too, but I had called for something and then got distracted looking for traffic...but the NFP is supposed to actually look at things like engine gauges!

Quite simply, I don't see the need for auto throttles. How amazingly hard is it to move throttles with your hand?

[DozyWannabe]

A37575:

I don't see that as automation dependency so much as an alarming lack of SA. You've got two visual cues - one of them right in front of you and neither of them particularly hard to spot, and when you spot the idle thrust lever, immediately assume an engine failure rather than see if moving the throttles manually resolves the issue.

That guy clearly needed some remedial training before he was put anywhere near the left seat again. Letting George take the strain is one thing, but paying so little attention to what George is doing is going to get you killed if something goes wrong.

[Tee Emm]

I don't see that as automation dependency so much as an alarming lack of SA.

On the other hand, it could be argued it is automatics dependancy that leads eventually to lack of situational awareness?

[DozyWannabe]

We don't know that though - it could be any number of things. The point I'm trying to make is that allowing SA to degrade that far by putting one's faith in the machine is a fundamental misuse of the automatics. Despite the press-fuelled hyperbole of the last 25 years, they are designed to assist the pilots, not replace them.

[d105]

A37575,

However much I find what you wrote above to be absolutely amazing it's beginning to dawn on me that I can't just keep shrugging those stories off as exaggerated or bar stories. I have no training experience except for single engine piston so personally I haven't seen anything like what you describe above. But I take your word for it.

I come from a country with a very traditional flying culture. We fly manually in good and in bad weather. Personally I try to maintain an acceptable level of raw data + no map display flying as well.

Up until quite recently I thought this was how most other pilots still fly. I too have heard the stories of total AFDS dependency in large airlines but it didn't quite dawn on me how far that would/could go until I started reading PPruNe...

[RAT 5]

One thing about using automatics wisely is it allows you to 'manage' the operation. Good buzz word, but true. However, monitoring the automatics control the a/c is only possible if you know what it should be doing; i.e. what pitch & power the automatics should be applying. If you see what you want to see you can relax. Just leaving George and his family to fly and assume it is correct is highly dangerous, but sadly the new generation do just that because they've not been given the grounding to be able to monitor sensibly. Is George doing what i would do? Yes, then I'll give him a biscuit after the flight. No, then I'll kick his backside, takeover and only let him play again after a big sorry. Or I'll say a big Oops!, sorry myself and push a different button. Following the magenta line to 'who knows where' on a VNAV PTH to 'who knows where' is the name of the game for many. Needles are something for junkies. On non-GPS a/c they are vital. LNAV can not be used for approaches. These GPS jockies don't know what map-shift is, but it could still happen. We tune Nav Aids as back up, but the youngsters don't monitor them. LNAV/VNAV is God. I found it amazing that in the Cali B757 crash they tried to blame Jeppeson and Boeing. Incredible. It was a major human screw up, up front. How much raw data, SBY instrument flying is included in command courses. I still say the pax expect us to be their insurance policy. They expect us to handle it when the computers screw up and go fizz pop. The technocrats answer to human failings is to invent more backup systems and redundancy. In dong so they have indeed made many things safer, but the training dept's have a duty to teach awareness and not dependancy. One airline I was with had a training captain whose rote in the sim was "fly the FD, fly the FD". It was always a little fun to demonstrate how this could stall the a/c or fly it into the ground. He didn't change, but a cadets thankfully got the message.

[PJ2]

d105;

Regarding your comment, "Up until quite recently I thought this was how most other pilots still fly. I too have heard the stories of total AFDS dependency in large airlines but it didn't quite dawn on me how far that would/could go until I started reading PPRuNe..."

I suspect most pilots of small and large commercial operations alike, still fly the way you describe - professionally, quietly competent and entirely unaware of the present dialogue - they're getting on with business.

Yet it is important to observe the trends...to see "which canaries in their little cages are still alive", as it were - to note and delineate the boundaries of what is acceptable and what is not. That is what most do here who take the time to offer views, thoughts and facts.

While standards, professionalism and the career/job itself have changed almost singularly through the hands of non-aviation people who, among many things they must deal with to keep their enterprise going in a neoliberal political economy must eke out a profit for speculative and sometimes-fickle shareholders who reward slowly and punish quickly, do so by taking money from the only "flexible" source left - their employees, their training budgets, and other "non-profit-making" parts of a commercial aviation operation such as staffing and resourcing flight safety departments and programs, etc where the effects of fiscal parsimony do not show themselves right away thus making them easy targets for bottom-line thinking and meetings with the CEO and COO. The effects and the results of such fiscal cutting to the bone only emerge years later, (as they are now), the connection with turns in accident rates long since disconnected from origins.

I believe what we are observing here are trends at the edges of what we have been accustomed to believing over the past thirty years, are nominal (professional, safe) operations, certainly in the western world, (North America, Europe, Australia, New Zealand, some countries in South America). The fatal accident trends have borne this belief out but many saw the underlying seeds of change and began commenting early in the 1980s. The "neoliberal political economy - the set of notions de-regulating business to release it to free-market forces, the privatization of all services, including regulatory oversight, (SMS), which formerly came under government responsibility and which began in the US in the early 1970s and which effects and outcomes have been latent up until the mid-80's or so, have brought strains to our industry which are no longer invisible. The best summary of these trends and problems was given by Captain Sullenberger in his February 2009 address to Congress.

Many here "get it" and have already clearly stated that "automation" is not the problem - it is the belief by non-aviation people who have grown up in a software/microprocessor age and who manage the business of aviation but don't or can't manage "aviation" itself believing, for whatever reasons, that "these airplanes fly themselves", and that professional pilots are just expensive add-ons to the bottom line which can, and have, been cut.

An MBA from Harvard or whereever does not teach someone about aviation - it teaches one about business principles in which the discourse is "profit, loss, and cost control".

We long ago lost the third crew member and our defence, based upon flight safety, was dismissed as "union featherbedding" - an effective if not rhetorical technique which easily convinced non-aviation, anti-union designers, manufacturers and airline managements.

Now I know that that appears to be hyperbole - an exaggeration. Without the lengthy discussion which must flesh out the details of this unfortunate present trend, it probably is, but it is, above all else, true - it IS what has, and is happening to our profession. PPRuNe has done an admirable job of providing the place and collating the discourse by professionals who have not only observed these trends but are patient enough and well-written enough to have expressed these views making such trends and causes visible to those who are getting on with the job of being a professional commercial pilot.

Accident reports are about "what happened", but rarely about why. "Why" comes from the "philosophers" so to speak, and are as worth reading (here and elsewhere), as are formal accident reports.

Your incredulity is entirely understandable - the turns which the profession of "airline pilot" has suffered at the hands of those who only comprehend the cost of pilots but not the value, are a rude surprise to all veterans. The difficulty now emerging is for those who, despite the atrocious pay, the terrible working conditions and the constant dissing and lack of respect for what pilots do for aviation and for their companies, still choose to come into the profession, is that these changes are "normal", and, (like the Colgan First Officer), don't know what they dont' know about the profession and are not being mentored or taught by those who do know.

The fatal accident rate is turning around from its long reduction and beginning to show the results, not only in number, but in the puzzling "quality" of the causes such as stalling one's airplane...a dozen or more such accidents are now on the books when such causes in the 70's/80's were quite rare.

Nor am I suggesting that pilots run airlines...that actually would be as bad...:)

What I would envision within the severe limits of the present political economy is a hearty respect for what one's operations people are saying, first among those being the people at the pointy end of the airplane. SMS is about data, so staffing and resourcing programs which can tell non-aviation people just how close they're getting to the bone and where the risks and precursors to an accident are, is absolutely necessary notwithstanding that such programs and departments aren't traditional "profit-centers".

That, for me anyway, is what this and other extended conversations concerning the effects of "automation" is about. It is not about introduction of or disturbing reliance upon automation, it is about attitudes, priorities and values which have been inappropriately cheapened through a number of economic forces. If the "assisting tools" are used appropriately flight safety is enhanced, but not if done cheaply, with poor understanding or plugging it in and sitting back or not hitting the books and learning.

To emphasize this point, made by many already:

Automation is about cost, not safety, but safety can be enhanced through the full comprehension of such systems and the appropriate level of use of automation and knowing at all times what the automation is doing. However, a lazy reliance on things that are designed to fly the airplane better than we can but which don't think, has atrophied gut-sense and situational awareness so necessary for survival. In the vast majority of events, incidents and even accidents, to "blame" automation, is really to blame ourselves. I reserve judgement for those very rare circumstances in which "what's it doing now?" really means something such as the QF72 A330 incident near Learmonth.

PJ2

[DozyWannabe]

...but sadly the new generation do just that...
...but the youngsters don't monitor them...

Are you sure you're not unfairly tarring many younger pilots with that brush? I find it hard to believe that what you're describing is universal.

I found it amazing that in the Cali B757 crash they tried to blame Jeppeson and Boeing.
Firstly, that's lawyer SOP - always target the organisations with the biggest pockets. Sometimes fair, other times not.

Secondly, if I recall correctly there was a mismatch between the short-form waypoint identifiers as they appeared on the chart and as they were programmed into the FMS. Rozo (the correct waypoint) appeared on the chart as "R", but was only programmed into the FMS as "ROZO". What happened when they entered "R" was that the FMS gave them a listing of matching waypoints, of which the first was the Romeo waypoint near Bogota. The pilots (understandably, given the error on the chart) simply hit the enter key twice. The human error in that case was not following what George was doing immediately after that (a course change which included a significant left turn) - however in mitigation, the cockpit workload had increased significantly due to them accepting the straight-in approach offered by ATC. This doesn't excuse the flight crew for not having at least one of them monitoring what the aircraft was doing - a problem that has existed at least as far back as EAL401.

In short, the major factor in the accident was a lack of situational awareness on the part of the pilots, however the mismatch between the charts and FMS must be considered a significant hole in the cheese.

The technocrats answer to human failings is to invent more backup systems and redundancy. In dong so they have indeed made many things safer, but the training dept's have a duty to teach awareness and not dependancy.
(emphasis mine)

This is where I get uncomfortable with blaming automation directly. When it comes down to it, what we're looking at is a complex series of related issues - senior management who may not have an aviation background and therefore do not understand that as an industry it needs to be treated differently than many other businesses, training and operations departments who are under pressure from the management to keep costs down and yes, some pilots who have been poorly trained as a result of this.

But I'm convinced that the safety improvements that the automatics offer outweigh the problems caused by some in the airline industry misusing them, and it is something that needs to be addressed by the airline industry sa a whole. I'm also deeply saddened that some pilots view modern automatics with suspicion, largely because of press-fuelled speculation about computers one day replacing pilots in airline operation. I don't think there's an engineer alive who would want to take on that responsibility with the technology we have today, and I suspect the technology that will become available in my lifetime.

[EDIT : PJ2 has said what I'm saying far more eloquently - though I think automation is about cost *and* safety, at least from an engineering perspective. ]

[Gretchenfrage]

In the modern world the political nations are governed by separation of powers. There are the executives, the lawmakers and the judges. (i will not contest the new power, the media, having a big impact as well).

In aviation there was a comparable set up with the company executives, the aviation authorities and the genuine postholders of safety within the company or some NTSB. (the unions also playing a part in it).

Today the company executives pay all of them and practically own them at least politically through economical blackmail. Rising big powers in Asia are set up in dictatorship countries anyway. The unions are almost eliminated and the media is so incompetent, that their impact is controllable. The big manufacturers play along, as it serves them well. The outcome is blatantly readable on more and more accident reports, but only for insiders, as it is well masked by the mighty interest groups.

It will not change, as it works nicely on the bonus side. Victims don't matter as long as the numbers stay below an unfortunately very high public trigger.

So brace for more automatic induced and low training enhanced accidents. The warnings on threads like this will only serve as cover-up in a future rude awakening.

For example:
" ... there were some voices among the professionals, but they should have stressed their point much stronger at that time, after all they were in charge of the final safety. So in effect the professionals bare the main responsibility for the present catastrophe, as the engineers and CEOs couldn't have fully known the impact of their actions ."

[Willit Run]

Human Beings are inherently poor multi-taskers. We are also poor monitors. We are better monitors than multi-taskers. Automation can reduce the multi-tasking making the operation safer, but we need to train people to be better monitors so the automation can do whats its intended to do,make things safer; not complicate the situation. Complacency is similar to fatigue. You don't know you're introuble/tired until its often too late.
If humans are involved, mistakes will happen. Computers are programmed by humans.
As automation becomes more commonplace; we must always ask ourselves, is the automation doing what its supposed to do? We must always crosscheck the automation. Keep that scan going!

Edited on the advice of the next poster. Good catch, thats exactly right!?!:"+)

[aterpster]

Willit Run:

As automation becomes more commonplace; we must always ask ourselves, is the automation doing what its supposed to do? We must always crosscheck the automation? Keep that scan going!

You ended the next to last sentence with a question mark. I don't believe that is what you intended. (?) :)

Unless the crew fully understands the automation and all of its possible annunciations and nuances, they will not be successful (or valid) monitors in all circumstances.

The near loss of the 747 discussed earlier in this thread is a classic example. The autopilot gave a stark indication by attempting to do with aileron what needed to be accomplished with rudder, and the engine instrumentation gave indications, all of which were ignored.

I suppose the Boeing designers could have also been faulted for not having a full-time three axis autopilot on a modern design aircraft. (if my recollection is correct that it was a 747-400.)

[d105]

But where will it end when we start to blame manufactures for not dummy-proofing their aircraft?

[DozyWannabe]

I suppose the Boeing designers could have also been faulted for not having a full-time three axis autopilot on a modern design aircraft. (if my recollection is correct that it was a 747-400.)

Nope, it was a 747SP - manufactured in 1982, but definitely one of the older models.

[PBL]

I found it amazing that in the Cali B757 crash they tried to blame Jeppeson and Boeing. Incredible. It was a major human screw up, up front.

RAT 5, I am going to be somewhat more harsh than Dozy Wannabe, because the available information and analysis has been out there, in full view, on the WWW, for some 14 years now.

There is no "tried to". The named manufacturers were responsible for features of their products that demonstrably played a causal role in the accident. You may read about those features in the NSTB letter on the accident to the FAA. A Texas court later found those manufacturers partly responsible, because of the causal factors of the accident contributed by those features of their designs. The courts are, sometimes, able to follow careful causal arguments that don't rely on the gut reactions of casual observers, and pilots everywhere should be grateful for that.

Accident analysts look at all the factors. When you say "major human screw up", I take it you mean that the behavior of two pilots + controller was solely responsible for the accident, and that features of the nav database and FMC programming were not at all responsible. If that is what you are claiming, it is demonstrably mistaken: check out our causal analysis from 14 years ago at http://www.rvs.uni-bielefeld.de/publications/books/ComputerSafetyBook/Chapter_23_The_cali_accident.pdf

If, rather, you are claiming something milder, such as "yes, features of those products did play a role, but the human-behavioral components on the day were so much more important causal factors", then you owe us a definition of what makes a "more important" causal factor and what makes a "less important" causal factor, and a demonstration that, according to your criterion, the features of the automation provided by those companies was unimportant. And while you do that, you can explain your disagreement with the NTSB and the Texas court along with it. Good luck with that.

PBL

[aterpster]

DozyWannabe:

Nope, it was a 747SP - manufactured in 1982, but definitely one of the older models.

Thanks for the correction.

In that case, the supposed "short stop" F/E was apparently asleep at the switch, too.

[DozyWannabe]

In that case, the supposed "short stop" F/E was apparently asleep at the switch, too.

Actually no - he was well aware of their situation and focused on trying to get the number 4 engine restarted as ordered by the captain, in fact they called the relief FE up to assist him. This was just prior to the upset.

What he did not do was inform the captain that he'd ordered a restart of the engine when the aircraft was 11,000 feet higher than the maximum relight altitude - but I'm not sure if that was either a memory item for the FE, or in the QRH.

[DozyWannabe]

While standards, professionalism and the career/job itself have changed almost singularly through the hands of non-aviation people who, among many things they must deal with to keep their enterprise going in a neoliberal political economy must eke out a profit for speculative and sometimes-fickle shareholders who reward slowly and punish quickly, do so by taking money from the only "flexible" source left ... where the effects of fiscal parsimony do not show themselves right away thus making them easy targets for bottom-line thinking and meetings with the CEO and COO.

This is something that is most acute in the West, where short-term thinking on the part of executives and shareholders trumps all. Everything is focused on the next quarter's results and longer-term trends are effectively ignored. Contrast this with the old Japanese/Asian model (which admittedly has it's own problems), where longer-term trends are given equal importance.

The "neoliberal political economy - the set of notions de-regulating business to release it to free-market forces, the privatization of all services, including regulatory oversight, (SMS), which formerly came under government responsibility and which began in the US in the early 1970s and which effects and outcomes have been latent up until the mid-80's or so...

Well, what we have in the UK and the US are government or civil service organisations, where the investigating agencies are nominally independent from the regulators - as far as I'm aware that is still the case. Where this model has severe limitations is that the investigative agencies can issue recommendations, but the regulators are under no pressure to implement them - the exceptions to this being incidents and accidents where either the incidents themselves or the potential for reoccurence are so horrifying that something has to be done, or at least seen to be done - which is a whole other can of worms.

Many here "get it" and have already clearly stated that "automation" is not the problem - it is the belief by non-aviation people who have grown up in a software/microprocessor age and who manage the business of aviation but don't or can't manage "aviation" itself believing, for whatever reasons, that "these airplanes fly themselves", and that professional pilots are just expensive add-ons to the bottom line which can, and have, been cut.

You have no idea how much it heartens me to hear that coming from a pilot.

I bang on about this a lot, but the whole "plane can fly itself, will pilots become obsolete?" spiel has only ever come from - and been propagated by - the press (and maybe a couple of sales guys who took things a little too far have said the former). Given the short shrift so many on here ascribe to the accuracy of reporting in aviation matters it really bothers me that some pilots on here seem to think that journalists are on to something in that case.

An MBA from Harvard or whereever does not teach someone about aviation - it teaches one about business principles in which the discourse is "profit, loss, and cost control".

Precisely - but what it does get you, even more so when backed up by business and/or political connections, is a route straight to a senior management or executive level position without the arduous task of having to learn aspects of the industry in which you're working.

We long ago lost the third crew member and our defence, based upon flight safety, was dismissed as "union featherbedding" - an effective if not rhetorical technique which easily convinced non-aviation, anti-union designers, manufacturers and airline managements.

I want to jump in here to point out that the first 2-crew airliners (the BAC One-Eleven, Douglas DC-9 and Boeing 737) arrived nearly two decades before modern automation as we know it became widespread - a fact that seems to pass a lot of people by.

The difficulty now emerging is for those who, despite the atrocious pay, the terrible working conditions and the constant dissing and lack of respect for what pilots do for aviation and for their companies, still choose to come into the profession, is that these changes are "normal", and, (like the Colgan First Officer), don't know what they dont' know about the profession and are not being mentored or taught by those who do know.

The other problem being that a lot of these "lower-rung" positions seem to largely be offered by organisations like Colgan, which are run so close to margins anyway that corners are likelier to be cut in ops and training.

What I would envision within the severe limits of the present political economy is a hearty respect for what one's operations people are saying, first among those being the people at the pointy end of the airplane. SMS is about data, so staffing and resourcing programs which can tell non-aviation people just how close they're getting to the bone and where the risks and precursors to an accident are, is absolutely necessary notwithstanding that such programs and departments aren't traditional "profit-centers".

And here we get into my particular bugbear - one that's not restricted to aviation, and comes very close to your views on "neoliberal" business practice. When the Greeks defined "hubris" they were exceptionally prescient - pretty much every tragic event up to the present day has involved people either individually or collectively "believing their own hype". In the case of modern business (not all, but many), you have a managerial and executive class that answer to no-one bar themselves or the shareholders. Individuals among them, because of the rise of the MBA, have never known anything different - and they (the finance industry in particular) have their own press, which has spent the last 30 years declaring them the best of the best, and inculculating the belief that they can do no wrong. In business and in politics, the top people were quite happy to say to the people below them "If you don't tell me what I want to hear, I'll just have to get me someone that will".

It got to the point about two decades ago where some middle managers started aping them, professing not to care about how the employees they were responsible for did their work, only that it was done. Some of those took it so far as refusing to listen to employees who tried to explain that their targets were unrealistic, and took it out on those employees when the wheels came off as predicted.

The operations people are therefore facing the unenviable task of telling management things they don't want to hear, and if they are unlucky, reporting to the kind of manager who will threaten their job if they don't come up with something more palatable to the board. In an extended recession, the situation is even more perilous and it becomes a question of who is willing to stick their neck out and potentially risk everything for the sake of pointing out something is unsafe when there is a very real chance that they will simply be sacked and replaced by someone who is willing to do things the way the board wants, no matter how unsafe.

Then, even when the midden hits the windmill, the executives - who by and large negotiate their own contracts with clauses to deal with such things - simply take their six or seven figure golden parachute and leave the company. Most of them will find employment again in short order, and even if they don't the value of their severance pay alone will allow them to make investments that will keep them in a very comfortable lifestyle to the end of their days. Again, contrast this with the Japanese/Asian model where executives must take cuts in pay if the company fails to perform.

In short "Too Big To Fail" isn't new, it isn't just about banks and it's just about as corrosive a system as can be designed.

Now, I realise I've gone way over the bounds of an "Automation" topic, but I had to get this off my chest - thanks to PJ2 for providing the springboard.

[aterpster]

PBL:

RAT 5, I am going to be somewhat more harsh than Dozy Wannabe, because the available information and analysis has been out there, in full view, on the WWW, for some 14 years now.

There is no "tried to". The named manufacturers were responsible for features of their products that demonstrably played a causal role in the accident. You may read about those features in the NSTB letter on the accident to the FAA. A Texas court later found those manufacturers partly responsible, because of the causal factors of the accident contributed by those features of their designs. The courts are, sometimes, able to follow careful causal arguments that don't rely on the gut reactions of casual observers, and pilots everywhere should be grateful for that.

That Texas jury was hoodwinked by deep-pocket-seeking plaintiffs' litigators.

Further, the NTSB has less than a sterling record of objective and competent analysis. And, Cali was not their investigation in any case.

My throat is sore from saying it over and over. :) :

These guys were trying to fly a non-LNAV VOR/DME combination arrival/IAP and throwing in an attempt at a short-cut in a non-radar environment. It would have been so easy to:

1. Gone to heading mode and raw data to sort things out.
(I believe AAL policy at the time required that one of them be in raw data mode.) (They finally did go to heading mode, but far too late.)

2. Have stopped a wild, excursion out of protected airspace while continuing a wild-a**ed descent with spoilers extended.

3. The course change was unwarranted on its face.

[DozyWannabe]

aterpster:

That's why the chart/FMS mismatch was considered a "contributing" cause rather than a "primary" cause (most of which were variations on loss of SA on the part of the crew).

Nevertheless, the mismatch did contribute and was proven to contribute. I don't mean to sound harsh, but the fact that you feel the flight crew should have done things fundamentally differently doesn't alter the situation.

One thing that we're skipping in this case, which relates to what PJ2 was saying regarding management pressure was that the whole reason they agreed to expediting the arrival was because they were concerned about delays - go back far enough and this crash was another example of the dangers of "Get-there-itis".

[aterpster]

DozyWannabe:

aterpster:

That's why the chart/FMS mismatch was considered a "contributing" cause rather than a "primary" cause (most of which were variations on loss of SA on the part of the crew).

Nevertheless, the mismatch did contribute and was proven to contribute. I don't mean to sound harsh, but the fact that you feel the flight crew should have done things fundamentally differently doesn't alter the situation.

No, I don't see you as being too harsh. You and I have a different view of the facts. :)

The four probable causes makes it quite clear the Colombia investigators reached the same operational conclusion that I did. I have no quarrel with their probable causes. But, in the area of contributing factors I believe they were expecting more of 1995 FMS databases (i.e., be the "same" as the paper chart) than was realistic. In fact, the same circustances exist today in many, if not most FMSes: select either direct-to a downstream waypoint or VTF, and "poof" the legs page no longer resembles the paper (or electronic) chart.

Also, as discussed in the report, no doubt that "get there itis" really had this crew hooked. Nonetheless, had they been sufficiently sharp to have simply reverted to raw data, they could have been quickly back on track and had sufficient distance remaining to the airport that a speed-brake assisted descent would have gotten them down in time for a landing on Runway 19.

From the Colombian report:

3.2 Probable Cause


Aeronautica Civil determines that the probable causes of this accident were:

1. The flightcrew's failure to adequately plan and execute the approach to runway 19 at SKCL and their inadequate use of automation.

2. Failure of the flightcrew to discontinue the approach into Cali, despite numerous cues alerting them of the inadvisability of continuing the approach.

3. The lack of situational awareness of the flightcrew regarding vertical navigation, proximity to terrain, and the relative location of critical radio aids.

4. Failure of the flightcrew to revert to basic radio navigation at the time when the FMS-assisted navigation became confusing and demanded an excessive workload in a critieal phase of the flight.


3.3 Contributing Factors


Contributing to the cause of the accident were:

1. The flightcrew's ongoing efforts to expedite their approach and landing in order to avoid potential delays.

2. The flightcrew's execution of the GPWS escape maneuver while the speedbrakes remained deployed.

3. FMS logic that dropped all intermediate fixes from the display(s) in the event of execution of a direct routing.

4. FMS-generated navigational information that used a different naming convention from that published in navigational charts.

[bubbers44]

I flew into San Salvador the night of the Cali crash and my FO was concerned about our VMC circling approach. There was no moon so even in clear conditions you could not see terrain. I flew the Cali flight the next several months after checking out in the 757. I always flew the whole ILS approach after what happened. One night they cleared me for the 19 straight in approach and I declined it because so many had died doing it even though I could see the runway with calm winds. It just didn't seem right to do an easy approach that had proved fatal to so many. Situational awareness was their problem. They went heads down and forgot to fly the airplane. They turned east because of the Bogota R outer marker showed up first on the FMC. I don't know why they didn't notice the left 90 degree turn putting them into the mountains. Yes it was pilot error that was helped by computer generated data. It should have never happened with a crew paying attention to the store.

[sevenstrokeroll]

Bubbers 44 has hit the nail on the head....it is discipline. It is knowing that their is a ''shortcut" and not taking it because there can be a fatal flaw when things aren't perfect.

Going to an airport you've never been to? Take the full ILS and don't just head for the nearest airport.

Ask for a radar point out even on a beautiful day.

And, a real killer, ask for progressive taxi instructions if the airport has construction or something odd going on.

And I think the biggest one...if the weather is bad, always wait 15 minutes. If it gets worse, you are vindicated. If it gets better, you were just waiting for a better ride for the passengers.

[PBL]

terpster,

it is likely we are not going to agree on the Cali analysis, but at least we could agree on what the criteria for a proper causal analysis are, and hammer out the differences.

That Texas jury was hoodwinked by deep-pocket-seeking plaintiffs' litigators.

Further, the NTSB has less than a sterling record of objective and competent analysis. And, Cali was not their investigation in any case.

My throat is sore from saying it over and over. :


That is all ad hominem stuff.

Further, I refute your suggestion that the people I know at the NTSB who were involved with that analysis are neither competent nor objective. The mystery of the left turn was solved by the NTSB investigator who found the non-volatile FMC memory amongst the rubble (that is the same guy who solved the TWA800 case, BTW. Just to have those two on the resume is a good lifetime's work, in my opinion). The human factors work was undertaken by the same investigator who introduced the questions about biorhythms and alertness physiology for the very first time in any accident investigation into the 1985 China Air Lines upset over the Pacific, which I regard as a milestone in accident investigation.

However, we both know that that ad hominem commentary is not your real argument. You real argument is more careful. Our argument is contained in the paper I referenced.

BTW, the conclusions to which the NTSB came are not necessarily the same conclusions to which the Colombian investigators came; as you say, it was Colombia's show. And citing the list of causes as listed in the accident report neglects the unfortunately pervasive fact that the causal reasoning in most accident reports is nowhere near rigorous (read: there are mistakes in that reasoning in at least half of the reports we have analysed). You are not arguing with the report when you are arguing with me: you are arguing with our work, not theirs.

You have a list of causal factors which make your throat sore :) I do not disagree that those were causal factors. What I would like to know is what criteria you are using to rule out the other demonstrable causal factors. You are picking and choosing and I want to know how and why.

Further, I want to know what exactly is wrong with the causal analysis we presented a decade and a half ago in that paper. I don't see any mistakes in it. If you can't find mistakes in it, they you haven't refuted it adequately just by saying you disagree.

PBL

[Tee Emm]

Nonetheless, had they been sufficiently sharp to have simply reverted to raw data, they could have been quickly back on track In my experience as a simulator instructor for many years, it is still common to see crews so locked on the MAP and magenta line, that even with raw data up via RMI VOR/ADF needles in front of them, the raw data information meant nothing to them - because they didn't know how to check the RMI readings quickly to get fixes from a bearing or radial.

A quick 30 second switch-over to the VOR/ILS HSI mode - coupled with an RMI reading and a DME reading, would have proved instantly where they were in relation to the MSA. But we regularly saw crews so rusty on basic radio navigation aid position fixing techniques, that it became readily apparent they did not know - or had forgotten - how to interpret their basic nav instruments.

So whatever raw data they had available, and in theory were supposed to be monitoring, meant nothing.

[aterpster]

PBL:

Further, I want to know what exactly is wrong with the causal analysis we presented a decade and a half ago in that paper. I don't see any mistakes in it. If you can't find mistakes in it, they you haven't refuted it adequately just by saying you disagree.

Link me to your paper and I will read it carefully.

My issue is squarely with relating their misuse of the FMS, auotflight, and navdata becoming in any way a liability for anyone other than American Airlines, and only in the context of their agents' inept operation of the flight. The analysis of the FMS was fine to point out limitations of which the crew should have been aware. But, some of those limitations still exist today; hopefully all crews understand them better.

I was a "pioneer pilot" on the 767 having started training in 1983 and flew it unil April, 1986. We reverted to raw data periodically as a matter of staying proficient in all of the options.

[aterpster]

PBL:

Just after the Colombians made a preliminary press release, I opined about the accident on my web site. A few months later I wrote an aritcle about the accident for the Air Line Pilot Magazine:

Here is the opinion part of my posting on my web site some 15 years ago:

MY OBSERVATIONS AND COMMENTS ABOUT THE ACCIDENT:

The instrument approach from the north at Cali, and its essentially identical terminal arrival procedure, proceed via a dogleg approach course. This approach course begins at Tulua VOR and proceeds down a canyon that just has room for standard instrument flight rules (IFR) terminal and approach procedural airspace. The majority of the instrument approach procedure can be flown at as low as 5,000 feet, msl, in a canyon with terrain rising steeply to 13,000 feet to the east, and over 6,000 feet to the west.

The non-radar services provided by the Cali ATC controller were completely correct and in accordance with accepted international standards. Further, the controller knew that it was essential that AAL 965 begin the approach at its mandated beginning: the Tulua VOR. This is evidenced by his repeated requests for a Tulua mandatory position report.

Although it would have been helpful had the flightcrew had intimate familiarity with the terrain along the approach course, such knowledge was not essential to the safe use of the approach procedure. Instead, the crew should have been conditioned to know that an approach without an ATC-provided radar vector must be flown in its entirety. In this case, that meant starting the approach at Tulua VOR without exception.

However, pilots trained in the United States, and who generally fly in the United States have, as a group, been lulled into generally thinking in terms of instrument approaches in a radar-driven ATC environment. Plus, to move traffic, the FAA itself encourages shortcuts with a wink and an approving nod, so to speak. Air carrier simulator and ground school training deals with radar vectors to the approach's final approach course as a matter of routine.

Further complicating the mix are the area nav systems on modern airline aircraft, which make it easy and tempting to always cut the corner, and go: direct, direct. This is fine in a radar enroute environment, but it killed the crew and passengers of AAL 965. Imaginary TERPs containment areas prior to the final approach segment are routinely breached during non-radar operations within the United States. This is because of inadequate training and understanding by pilots, ATC controllers and, today, most of FAA's management, about the essential requirements to fly the entire instrument approach procedure with absolute compliance. Usually, the transgression is forgiven, because there are no rock walls in the area beyond the protected airspace. But, the rocks can exist, as they do north of Cali Airport. As the United States pushes forward with GPS approach procedures, we will see more approaches hugging the canyon walls, so to speak. Yet, the FAA seems oblivious to the problem.

ALPA's Charting and Instrument Procedures Committee (CHIPs) has been urging the FAA for over three years to publish first-rate, instructional and directive information about all the critical nuances of proper flying of the full instrument approach and IFR departure procedures. These efforts have gone nowhere with a unresponsive FAA. Further, the CHIPs Committee forced the FAA to issue a legal interpretation that, excepting a radar vector, an instrument approach must begin at the appropriate feeder fix or initial approach fix (IAF). But, the FAA refuses to publish this requirement in the Aeronautical Information Manual, much less widely disseminate comprehensive guidance to the aviation community. Because of lack of FAA leadership, it is probably a rare air carrier recurrent training program that addresses these IAF issues at all.

Further, the FAA, in a rush to develop 500 new GPS instrument approach procedures, is creating deadly gaps in these new instrument procedures by violating their own criteria. Instead of always tying the beginning and end points of these new instrument approach procedures to the published enroute airway structure, they are leaving deliberate gaps for pilots and air traffic controllers to try to work through. Also, these approach procedures are often designed to encourage shortcuts around required segments, because of lack of flexible design of individual approach procedures.

The FAA is now pointing the finger at American Airlines' training when, instead, they should be pointing the finger at themselves. Like the cancer that had grown in the ATC system that resulted in the TWA 514 crash at Dulles Airport on December 1, 1974, a similar FAA-induced infection can be seen in the recent crash at Cali.

The crew obviously lacked recent, good training on the essential requirement to begin the approach at the IAF. Instead, they took full "advantage" of their modern glass-cockpit, area nav system, and simply punched in the approach fix nearest the airport which, in this case, was a stepdown fix in the final approach segment of the approach. This confusion and lack of essential understanding was compounded by this stepdown fix (ROZO) being the name of the arrival procedure. Thus, the crew established a flight track that, although it converged with the instrument approach criteria's mandated protected airspace, it was outside of that minimal airspace, which resulted in the airplane literally scrapping the canyon wall.

In conclusion, I submit that more professional flight crews than not would have made a similar fatal mistake had they been in this situation. This can be directly laid at the door of an unresponsive, disjointed FAA.

[aterpster]

Tee Emm:

So whatever raw data they had available, and in theory were supposed to be monitoring, meant nothing.

If so that was grossly inexcusable. This was not an RNAV IAP. They had no need to use RMIs. They had just passed the IAF (ULQ) and when confusion ensued, a switch to raw data VOR/DME and selecting the 202 Radial would have been all they needed to do. There was no need to use RMIs or do fix positioning.

This is the VOR/DME chart they were using:

http://tinyurl.com/4jcxrld

[sevenstrokeroll]

when I first read about flying , thinking in three dimensions was part of the equation...up/down, left right/ forward back....but we must add time into that process, so, let's say 4 dimensions.

I think modern automation reduces the mind's capability to think in 4 dimensions. Whether really good pilots know it or not, they are working all the time in 4 dimensions.

Using RMIs is sort of tough and your must project your position upon a chart for terrain reference/safe altitudes.

Are our minds becoming weaker? I think so. Or if not weaker, more clouded by gadgets instead of using the imagination to know where we are in 4 dimensions.


When I was first learning to fly in California, USA...a number of radar mistakes lead airplanes into the sides of mountains. when I first flew to one of these airports, I insisted on flying the entire procedure as published...it added quite a bit of time to the flight, but I knew where I was every step of the way...right down to minimums.

Anyone here think flying is easy? Well, it isn't...if you aren't firing on all 8 cylinders of the mind, you aren't doing your job.

[aterpster]

sevenstrokeroll:

When I was first learning to fly in California, USA...a number of radar mistakes lead airplanes into the sides of mountains. when I first flew to one of these airports, I insisted on flying the entire procedure as published...it added quite a bit of time to the flight, but I knew where I was every step of the way...right down to minimums.

EGPWS has greatly reduced CFIT from errant radar vectors. Not everyone has EGPWS/TAWS, although with the handheld VFR GPS devices that have terrain alerting they should.

Moving map MVA charts would be great, too. In fact, pilots with geo-referenced Jepp charts have just that, but not in the U.S. The U.S. filed an exception with ICAO to radar charts.

[DozyWannabe]

This is the VOR/DME chart they were using:

http://tinyurl.com/4jcxrld

I'm pretty sure they were actually using this one (note Rozo identified as R):

http://www.rvs.uni-bielefeld.de/publications/Incidents/GRAFIK/ComAndRep/Cali/c2.gif

Regardless of how you feel about the unwillingness or inability to go raw data when they realised they were confused, the fact also remains that if they had entered "ROZO" into the FMS they would also most likely have come in for a safe landing.

[aterpster]

DozyWannabe:

I'm pretty sure they were actually using this one (note Rozo identified as R):

http://www.rvs.uni-bielefeld.de/publ...ep/Cali/c2.gif (http://www.rvs.uni-bielefeld.de/publications/Incidents/GRAFIK/ComAndRep/Cali/c2.gif)

Regardless of how you feel about the unwillingness or inability to go raw data when they realised they were confused, the fact also remains that if they had entered "ROZO" into the FMS they would also most likely have come in for a safe landing.

They were indeed filed via airways (W3) to the Cali VOR, followed by the ILS to Runway 01, as you have linked. But, all the discussions with approach control were about instead flying the Rwy 19 procedure, which I linked. (Both charts have ROZO on them).

It's not a matter of how I "feel;" they were obligated to fly the VOR/DME arrival/IAP to Runway 19 once they accepted the clearance. You feel they would have come in for a safe landing had they managed to jury rig the IAP and skirt the margins of the procedure. But, they screwed it up royally. As soon as things got screwed up they had an obligation to either abandoned the VOR/DME Rwy 19, and rejoin the airway at altitude to proceed for the ILS Rwy 1 or, if they were savvy, do the simple task of recovering the Runway 19 approach via raw data. That was a very simple thing to do in a 757.

As to skirting the procedure by going direct to ROZO, they would have been violating this statement on their AAL flight plan (as well as the Colombia AIP procedure):

From the accident report:

AA did however, provide the flightcrew with written terrain information on the flightplan. This noted that: "Critical terrain exists during the descent--Strict adherence to STAR necessary for terrain clearance." The evidence suggests that the flightcrew did not take this information into consideration during the descent into Cali.

Further, the accident report fails to mention whether the accident aircraft was an original FMS (no GPS sensor, just IRUs with DME/DME/VOR update). Because of the date of manufacture of the aircraft, it probably was a first generation FMS. That alone should have prohibit use of LNAV below en route altitudes because of lack of robust DME geometry, which can and has caused serious map shifts in Latin America during those years. Also, although I am uncertain, I suspect Colombia was not WGS84 compliant on the date of the accident.

[fdr]

"If the world were perfect, it wouldn't be"
"Yogi" Berra

Ataps': did you update your observations with the hindsight of the evidence that was rather skillfully derived by the AAI's?

(Peter is correct, the investigation techniques were advanced by the efforts of those involved in this investigation).

Would love to operate in a world that humans don't suffer from human constraints and where machines don't break down, and where computers don't suffer BSOD's. But that is going to have to be in another universe. In this one, we have "the germans in charge of law and order, the british in charge of cooking (and car electrics), the french in charge of personal hygiene... etc, and the Italians in charge of maintenance..." (apologies to all).

AA965 [oops... 587 was bigfoot....or how we learn about certification issues after the event, thanks zee] had 2 nominally performing humans (3 if the ATCO is included...) acting in a standard routine and didn't need much for the round bit to go pear shaped. What may be considered rash or wilful disregard later is often hard to distinguish pre accident from being expedient, and "applied initiative" at work.

Expectancy was a significant issue with the FMS-NDB issue, and it occurred at a tragically inopportune time with a crew that were at a low arousal state, in high risk conditions. That the crew didn't recognise that they had multiple SA errors starting to compound again confirms that they were human. (as was the software developer who released the NDB that was not consistent with ARINC standards... the same for the people who considered that the QMS of the NDB was adequate...).

" The future ain't what it used to be "
Yogi

If Peter's Risks Digest methodology is not to your liking, perhaps you may be interested in looking at risk as considered by Erik Holnagel (if there is some discomfort with Reason's simple model of causation), or at least consider SA matters in general as well described by Mica Endsley. As contributory factors, I believe if anything that the AA[965] investigation could have followed further down the path of causation, to highlight opportunities for reinforcing systemic weaknesses. (Vaughn or McDonald can also give some insight to how technology retains flaws, against the best intentions of those entrusted with it's implementation).

After Annex 13, and all other processes are complete, it is always the intent of this industry to avoid repetitive failings. Civil tort law has different priorities, and does not in itself lend to avoidance of repetition. If it did, my computer screen wouldn't regularly go to BSOD.

"It's deja vu all over again"
Yogi

regards.


FDR

[PBL]

Terpster, Dozy, fdr,

there is such a lot to say that it has taken me some time to think about how to reduce it to a few sentences. Let me first say that I am glad you all take these issues seriously. I think that Cali could provide a good test example for thoughts and tropes about automation and human operation.

First, terpster asked for references to our work on Cali. Here an extended answer. The paper AG RVS - Analysing the Cali Accident With a WB-Graph (http://www.rvs.uni-bielefeld.de/publications/Reports/caliWB.html) was presented at the first Human Error and Systems Development workshop, organised by Chris Johnson in Glasgow in March 1997. It is now called Human Error, Safety, and Systems Development and has had seven meetings. It is now due for the 8th, but I haven't heard. Chris also started a workshop on Accident Analysis, of which the first two meetings, in Glasgow and Virginia (organised by John Knight and Michael Holloway) were superb, but then petered out. The problem seemed to be that everyone wanted to come and talk, but no one would/could submit a paper (we got three submissions the next year!). That is, everyone is interested in accidents (witness the explosion of threads here on recent accidents) but few actually want to work on them, that is, submit an analysis in public and subject it to open criticism. There is no open forum for accident analysis, not even in the context of the premier system safety conferences. One should ask why this is. I was on the Editorial Board of the UK IMechE Journal of Risk and Reliability for a number of years, responsible for system safety and accident analysis, and we got not one submission. Even the people I worked closely with on the Bieleschweig Workshops submitted not one paper on accident analysis during the entire time. Bieleschweig was all just talk and slides (except for the papers I produced) - a good example of the PowerPoint syndrome.

The paper was incorporated into the first draft of a Causal System Analysis text in 1991:
http://www.rvs.uni-bielefeld.de/publications/books/ComputerSafetyBook/Chapter_23_The_cali_accident.pdf.

This text will not appear as it is. It has been split (and extended) into a text on system safety for computer-based systems, which is in draft form, and there will be a separate text on WBA. The original turned out to be simply too hard to teach from. It includes some logic, and proof of correctness of an accident analysis (according to explicit criteria) and it turns out that no one who is interested in accident analysis has the background in logic to be able to read this, let alone apply it themselves. Even after ten years. So I have given up this approach to teaching.

The last paper is AG RVS - Comments on Confusing Conversation at Cali (http://www.rvs.uni-bielefeld.de/publications/Reports/cali-comment.html), which will be incorporated in extended form in an article on safety-critical communication protocols in the Handbook of Technical Communication, ed. Gibbon and Mehler, Mouton de Gruyter, Berlin, to appear 2011.

Terpster, you will see if you look at these carefully that we knew about your work when you wrote it! Thanks for including some of it again above.

Second, I would like to emphasise and expand the view proposed by fdr. There is a context in which the human pilots played out their fatal game. Terpster says earlier here "the pilots did this and this wrong" and puts responsibility solely there, with the human pilots. But in his older writing which he quotes, he acknowledges the context (namely, what is "common acceptable behavior" in US operations) and puts the responsibility for encouraging/allowing that context to develop solely in the hands of the US regulatory authority (FAA). I want to say: thank you for making my point for me!

Third, let me say explicitly what terpster repeatedly points out: it is true that the pilots did not follow good, safe, TERPS-consistent procedure. But then, as terpster keeps pointing out, this is endemic in US operations. Not only that, but there are parts of the world (dare one say, Colombia?) which are not necessarily regulated by TERPS (indeed, one might say anywhere outside the US). Pilots cannot be expected to know all the approach-design criteria in use all over the world, just as those pilots regularly flying across German airspace cannot be expected to have read and understood German air law (first of all, it's in German; second, even if you know German, German-legal-speak is a different language with some - and I emphasis some - syntax in common).

There is no point of disagreement with the fact that the Cali pilots did not follow advisable, safe procedure. But I disagree strongly that that is the only factor (even terpster must back away from that claim, as his indictment of the FAA shows). I would even doubt that it is the most important factor, given that that kind of behavior is pervasive, as terpster points out, and most people behaving like that don't crash. There is a line of thinking about explanation, which I shall call "contrastive explanation" after the late Peter Lipton (Inference to the Best Explanation, 2nd edition, Routledge, London, 2004), which proposes that explanatory causal factors are those factors which were different in the (in this case) fatal case from how they are in all the non-fatal cases. If we are to explain contrastively (and Lipton gives deep arguments why we do and should, which I have not completely digested), then the contrast is not in how these pilots accepted a clearance and tried to clarify what that clearance was, but in the to them misleading naming conventions and the misleading "affirmative"s uttered by the controller when he knew that "negative" was the correct procedural response. (Terpster, if we are to lay responsibility on pilots for not following defined procedure, why not on the controller for not following defined procedure? The only answer could be: the controller is there to ensure separation only. But of course that is not his only role. There was only one aircraft around, so it cannot be. He is also issuing approach clearances, distributing critical information, and, one would forlornly hope, trying to ensure the approach is more or less followed).

Fourth, just as the pilots obtained misleading information through their miscommunication with the controller, the pilots also obtained misleading information from the nav database, whose detailed and sometimes whimsical design they were not completely familiar with (and, indeed, it takes a computer expert to be completely familiar with such things. That's my day job). Now, one may want to argue who is responsible for that? The pilots who were misled and "should have known better", or the DB designer who should have thought through the safety consequences of the design decision (routinely: hazard analysis, risk analysis, elimination and mitigation. I would bet you that no hazard analysis as we system safety people teach it was performed on that DB design before it was used)? The answer, surely, is that assigning responsibility is a different question from determining causality. As a causal factor, it is irrefutably there. Similarly with the FMS. Let me also say that the manufacturer, Honeywell, is very concerned with such questions, not particular as a consequence of the Cali accident and the adverse court decision but because they have some very smart people there who take such things very seriously indeed.

There is lots more to say, but let me (almost) quit here. The final thing is that it is futile to continue to put the majority of the responsibility on people not following procedure to the letter. They never do. People working their roles in complex systems optimise their roles according to criteria local to them (e.g., "I can get my job done faster with less fuss, and have more time to think about the *important* things, that is, what I consider important"). This is a pervasive phenomenon which has been identified independently in two noteworthy works and is probably about as permanent a feature of human operations in complex systems as there is. You cannot wish it away by saying "people should have followed procedure to the letter", because they never, or almost never, do. The phenomenon was identified first by Jens Rasmussen and called "migration to the boundary" (in his Accimaps paper from 1997). It was also rediscovered, ostensibly independently, by Scott Snook in his work on the Iraq Black Hawk friendly-fire shootdown, where he called it "practical drift".

So in some sense terpsters admonition that people should be sticking to procedure is tilting at windmills, if you take it at face value. The only way to change the human operators' habits is to create a context which does not allow them the latitude to "optimise" their work to the point at which safety is diminished.

Such contexts could be created at the carrier by, for example, instituting a rule that all approaches are to be flown as published. Then up go the fuel bills! Up go the flight times! Controllers in busy airspace such as NY, SF and LA are terminally aggravated! In short, the whole way in which a major carrier uses airspace and gets along with the rest of the system is radically changed. Won't work.

In contrast, fixing a DB design or an FMS design is easy.

Terpster says some of that design is still with us. I have my opinions on that, and I am working hard in standards circles to see that things evolve for the better.

PBL

[fdr]

Hi Peter, atapster,

"There are known knowns;
there are things we know we know.
We also know there are known unknowns;
that is to say we know there are some things we do not know.
But there are also unknown unknowns,
the ones we don't know we don't know".

Sec Def, "Rambo" Rumsfeld.


Jens and Eriks work as well as your own, Peter, inherently acknowledge that the systems are at the best non linear and at worst are stochastic. Certainly, Eriks resonance model's assume underlying stochastic system performance. These models are far more realistic albeit less "tidy" in application, than the near linear Reason Model. (I may be rather unfair on the limitations of artistry rather than the concept of James Reason's popular model). The potential for an error from any source to not only act directly but to result in unintended behaviour of the system is high. In operations, the sight of a trained crew all being head down dealing with a minor problem, incorporating their CRM training... while heading toward terrain, thunderstorms, other traffic or alligators (old days anyway...), or again being head down while taxiing on a crowded ramp while checking the CMC to confirm that a PA is needed to be made to the passengers for a "hard landing", or... having a 1kt overspeed due to environmental changes and pulling the aircraft into the vertical (well... 17 degrees nose high etc at FL420).

Migration to the boundary is a characteristic of the feed forward from successful outcomes where no incident occurred, under the conditions of human economy of effort, and environmental and operational stress vectors. Over time, the nett outcome will naturally trend to and beyond the regulatory boundaries unless monitoring of the actual performance leads to intervention and return to nominal design. Humans adapt pretty well, and the ability to find new methods to economise on the procedures is remarkable, inventiveness probably being as important as opposing thumbs in evolution (sorry creationists).

[This stands apart from the gross wilful violation of say having a child in the seat inadvertently nudging the APFD mode into CWS and the crew not having either the self preservation instincts to be near the controls, or the SA to understand what the mode of the AP was... (history does repeat, with a recent tail strike by a passenger flying an RPT jet...where do you start the remediation training for that? Put the pax in the sim?)].

Economy drives operation, design (mars surveyor... O rings... ), operational deviation, and risk acceptance away from nominal (assuming that nominal is a valid status anyway), as the result of all interactions that occur. Hardly a surprise, the second law of thermodynamics states that a closed system will tend to increased entropy. Human systems appear to also suffer divergence from nominal unless action is taken to restore performance. The background stochastic signal of performance mixing with variations from nominal performance just give some apparent arbitrary misfortune to the time and place of the event.

while hoping to use either Monte Carlo simulation or neural networking to improve modelling of off nominal performance, have been more effective at getting system crashes than modelling aircraft crashes.... main outcome has been less confidence in most of the existing mnemonics such as CRM and AQP, SMS etc, and more focussed on the basic loss of SA that exists at the core of almost all events.

"Therefore, failure-tolerant management systems basically depend on the continuous and efficient communication of corporate and individual values and intentions. One of the major risk problems
related to the introduction of information technology in centralised systems may therefore be the temptation of rational, scientifically minded experts to design large systems in which centralised data banks with factual information are the basis for communication between decision makers, and, unwillingly, disturb the communication of values and intentions which is necessary for error recovery".
Rasmussen, RNL, 1985 Risk and Information Processing

From designing jumbos to justifying invasions, somethings don't change in a hurry.

"You've got to be very careful if you don't know where you're going, because you might not get there".
Yogi...

[alf5071h]

fdr, PBL, et al, whilst it is interesting to seek a high level (academic) understanding of the human / system interface, it is also necessary to provide practical guidance for operators – a bit more than "You've got to be very careful … ” ;)

Humans have lived with ‘automation’ for a long time, generally adapting to the new circumstances and situations which arise. However, in aviation either due to the extent, or rapid development and implementation of technology in a complex operating environment, it appears that we in part failing to adapt and cope.

“More recent and contemporary utopias are almost invariably dystopias. This is interesting and in its way surprising development. It suggest that we have found technology more of a burden and a threat than a liberation and help as was promised.” A.C. Grayling, Utopia. ‘Ideas that matter’.

You provide a hint to solutions focusing on the loss of situation awareness. I would look further into this asking if an appropriate awareness was ever attained – having a plan; as Yogi said ”… if you don't know where you're going, ..."
I see these aspects as shortfalls in mental preparation, poor strategic decision making, which generates and updates the mental model. Perhaps there is a reluctance to do this (assuming that it was done pre technology), because of the existence of technology; ‘it’ provides, or will provide answers otherwise generated via a mental model – so now we have don’t think, just use the EFIS, FMS, autopilot.

A possible contributing problem is that the EFIS, FMS, or autos are not good devices for establishing or maintaining the mental model.
Is this due to a design issue or the way in which we have been taught to construct our mental model (planning) when using technology? We are unlikely to improve the design in the near future – at least in a meaningful flight-safety timescale, thus we might only have the human to work with.

A solution? Perhaps a way forward might require greater understanding of the technology we use, the situations where it is / is not of benefit, and how and when to use this information. Not just more training, practical training and continual knowledge-building focussed on the critical use areas of technology.

[rubik101]

Pardon me but shouldn't it be Bogy, as in; an imaginary evil being or spirit; goblin; anything one especially and often needlesly fears?

Rather than Bogie; undercarriage having pairs of wheels that swivel so that curves can be negotiated. An axle holding the undercarriage wheels.

[BOAC]

Alternative spelling, R101.

[HazelNuts39]

The alternative spelling is bogey?

[DozyWannabe]

‘it’ provides, or will provide answers otherwise generated via a mental model – so now we have don’t think, just use the EFIS, FMS, autopilot.

That was never the intent though - automatics are there to provide a backstop, not do your thinking for you. In an auto-flight environment you need to think *and* use the EFIS, FMS and autopilot.

[fdr]

Alf,

"Always go to other people's funerals, otherwise they won't go to yours."
Yogi.

Sorry if the malapropism is lost in translation from Mr Berra's comment of "you have got to be careful..."

The application of FMS, ND's etc changed the problems, as you say, for various reasons, not merely as a response to loss of SA, there being a slip between reality and the mental model of the operator, however in most incidents and accidents there is evidence of a loss of SA other than some structural failure events.

Advanced applied technology such as APFD/display systems and increased automation resulted in new ways of getting hurt, as humans didn't act as the designer predicted. Nor do the systems always. Removing the human from the control loop has also resulted in new failure modes, from increased detachment of the cognitive capacity of the individual to the system, acting to reduce SA in some occasions. The new opportunity for errors and slips change failure modes, such as MK's B742F OPT issues at Halifax. If humans are considered to be excessively prone to failure, then for a real mess just add computers. For all the frailty of the human in respect to system failure modes, they also remain the best last hope of intervention when conditions are not exactly as programmed by the... human in the design or automation processes.

Human conditioning is not immune from similar problems, an excessive expectation on the benefits of CRM, or a mis application of the concepts can just as easily result in target fixation, poor workload management and losses of SA. It is embarrassing to see a crew CRM a problem to perfection and then not implement a solution... just as it is pretty depressing to see a crew dealing with all the processes of CRM, EMC, and similar programs while running out of fuel in a holding pattern, or flying away from an airfield while on fire etc... The salient point remains in most cases, that any processes that reduce SA are not conducive to good health... including company SOP's, say, that cause a saturation of the crew at a critical time of operation, such as EK's pre departure routines, or KAL's pedantic application of pre takeoff reminders... or large legacy carriers operating an advanced cockpit variant in the same manner as a legacy system to "standardise".

"...And I for winking at your discords too
Have lost a brace of kinsmen: all are punish'd".
Wm Shakespeare, (1564 -1616) Romeo & Juliet, Prince, Scene III

Management are accountable for the unintended consequences of ill considered process changes, both legally and morally.

My views may be considered to be depressing, or not. Human failure tends to be the result of (IMHO) loss of SA in the main, even a large percentage of violations result from the individual achieving a state of loss of SA due to the extent of the deviation undertaken. I would think that it is a point of some hope for improved safety as it follows that processes, procedures, practices and design of systems that promote SA are going to generally be advantageous, but more specifically, the capacity of the individual to be trained in heightened SA awareness, and to be at the very least alerted to the precursors and indicators of SA loss is a practical program. This is not warm fuzzy stuff; the effect of simulating conditions where a student ends up with a substantial slip in their SA model can be fairly traumatic.

System safety does improve to some extent with the application of emerging technology that is developed as a result of the detailed analysis of system behaviour. The system demands also evolve and so the desired performance level to achieve an acceptable risk level is also an ever changing target.

PBL; thanks as always for adding qualitatively to these discussions on this forum.

[aterpster]

PBL:

Terpster, Dozy, fdr.

there is such a lot to say that it has taken me some time to think about how to reduce it to a few sentences. Let me first say that I am glad you all take these issues seriously. I think that Cali could provide a good test example for thoughts and tropes about automation and human operation.


We indeed have a communications issue between us. You are a scientist. I am not. I gave up reading the first paper you cite/linked because it delves far beyond my education, skills, and experiences. My primary life’s work was as an airline pilot. That was and is a job that does not require any formal post-high-school education whatsoever. Granted, many of the pilots hired by my airline (and other carriers) in the 1960s had undergraduate degrees, because the airline strongly desired (and usually could command) that level of formal education.


Those degrees did not have to be aviation-related, and typically were not (and are not to this day). The successful completion of an undergraduate degree provided a measure of assurance to the airline H.R. gurus that the candidate would likely be both disciplined and sufficiently intellectually alert to avoid difficulty with the equipment ground school courses over his (and later, or her) career. Nonetheless, because of expansion pressures some pilots were hired who had no college education, but who had a substantial amount of varied flying time. These college-deprived pilots typically did quite well throughout their airline careers. What the successful candidates had in common, were those difficult-to-define set of human skills necessary to be a good planner, evaluator, and able to continuously think ahead of the aircraft, so to speak, while being able to sequence tasks in the sense of appearing to “be able to juggle several balls at the same time.”


My education as an accountant, combined with my pre-airline stint as an instrument flight and ground instructor, plus a compelling interest, caused me to become an instrument procedures analyst, but as an adjunct (avocation) to my career as an airline pilot. I learned about the design concepts (and sometimes lack thereof) and construction methods used to build TERPs instrument approach and departure procedures. This is not an area of interest intended for the line pilot, nor should it be. PANS-OPS are not all that much different than TERPs, except for circle-to-land criteria. After all, a given airplane needs to operate at an acceptable target level of safety while IMC, whether in TERPs or PANS-OPS airspace. In fact, with the advent of performance-based navigation, the nominal differences between TERPs and PANS-OPS will eventually disappear. This is already true today with RNP AR instrument approach procedures.

As to my taking the FAA to task in 1996, I felt then they were doing a poor job of promulgating and explaining RNAV procedures. They have improved that mission greatly in the intervening 16 years, although the inevitable conflicts in the U.S. between the “900 pound gorilla” (FAA ATC) and the remainder of the agency continues as it always has.


I stand by my writings about Cali I penned in 1996. And, my comments in Post #128 of this thread are consistent with my view of the Cali accident. I do, however, take exception to contributing factors 3 and 4 more strongly today than I did in 1996. Those factors continue unabated to this day. But wary, informed, cautious pilots are able to resolve the seeming discrepancies quite nicely (the EDDS ILS 07 in your country recently discussed on this forum is a good example). I don’t worry for a moment, though, that the pilot proficient in both RNAV and ground-based instrument procedures would have any problem flying that ILS approach in any modern aircraft. What I do worry about on a systemic basis is the ever greater possibility of a repeat of “We’d like direct Rozo” (as in “we’d like direct to the middle marker.”) I don’t foresee a repeat of the wild excursion of AAL 965 into the mountains miles removed from protected airspace; instead I see some member of today’s “direct-to” crowd eventually shaving off a hilltop not far off to the side of the modern RNAV containment areas.


Finally, you dwell on the communications difficulties between the captain of AAL 965 and the Cali controller. My pragmatic view of that one is, and always has been, “Hey captain, knock off your doomed attempt to turn a Colombian controller into an FAA controller, and get back on your unabridged flight plan.”

In no way do I denigrate your work. It is extremely important for this industry. But, your mission is to open the avionics device and examine/critique the fundamental work of the designers. Indeed, that is quite necessary. My primary mission is to get the device to work safely in spite of its blemishes.

[alf5071h]

DW, my phrase might have better been constructed as “there is an attitude of ‘don’t think just use’ …”, my concern is that some people have not been taught or do not understand the intent of automation.

“In an auto-flight environment you need to think *and* use the EFIS, FMS and autopilot.” I agree, but how is this accomplished, how does the industry ensure that the intent of design and certification is realised in operation, how is this gap, full of assumptions, to be filled?

fdr, as depressing as your views might appear to some, I agree with you that awareness and its cognitive components are at the root of many human problems in auto operations. I often quote Orasanu and Martin who for me simplify the problem to that of either not understanding the situation, or with understanding, choosing an incorrect course of action. Thus situation awareness is central to current problems as is the accumulation of the various forms of knowledge.
If we wish to use Kern’s analogy of Airmanship, where knowledge are the pillars/walls of our building, and SA the roof, then for issues of automation and technology many pilots are living in the wrong house, not by choice, but due to lack of appropriate resource (training knowledge – what and how) and proficiency in assembling the resource, practice, practice, practice.

[RAT 5]

There's a lot of over complicating a simple case here. I am a simple guy and like to keep things easy. In many airlines they have an analysis system when problems crop up; to help the crew find their way through the maze of info, options etc. and to avoid rushing into the first option that pops into your head. In Cali they were perfectly on VNAV descent into a STAR for an ILS. It was dark, the PF had not been there before, they had briefed only the northerly ILS, they were FL200, 60nm out at 280-300kts. They were asked "if they'd like a straight in NPA to the southerly". Consider the facts from a simple airmanship point of view. Which part of NO THANKS don't you understand. All the 'causal contributions' came afterwards. The root cause was not mismanagement of the FMS, and Jeppe mistakes and Boeing mistakes with no auto-stow speed brake etc.etc. It was the overwhelming desire to try and save 10mins and change from a very safe Plan A to a very dodgy unprepared Plan B. That to my simple mind is an airmanship problem. That is the root cause of many prangs. Trying to dress it up afterwards as something else is not just. It smells like a mistake, sounds like a mistake and flew like a mistake. 10 minutes late is better than 20 years early. This will sadly not be the last such demonstration of rushing into a black smoking hole somewhere in the aviation world in the future. Such sad events have a habit of repeating themselves. Even if Jeppe had programed the R point correctly I judge the chances of a G/A to have been high. Either that or the chances of sliding off the end.... It should have been a non-starter. This is not easy hindsight, it is looking at the peformance parameters, the crew experience, the time of day and the type of approach in Plan B. "NO THANKS ATC. Kind of you to offer, but...." I stand by my opinion. We can agree to disagree. Fine.

[fdr]

Rat 5: congratulations, I get you don't make mistakes. Thats a relief, now we only need to sort out every other pilot.

I am unfortunately a lesser evolved species, being merely "human" (my wife may disagree on that classification on occasions) as are the other pilots that crash aircraft, design lousy procedures, build poor computers, and run whack-a-doodle governments around this polluted and poisoned rock we call home.

Searching around in a 5 mtr hole in the ground with smoking wreckage and the remains of bodies nearby, looking for answers why all pilots aren't fool proof takes the work that I do, and which I heavily rely on the research and tools developed by people such as PBL, Holnagel, Rasmussen, Johnson, Endsley, Mauro, Klein, Sumwalt... and co to understand why the dad guys aren't making the same "OBVIOUS" decision that you do. Wearing Biohazard garments takes this discussion out of the realm of being "academic" [BOACs, Atapsters posit].

Being merely mortal, when my passengers are on board, the only thing I know is that I make mistakes, the designer makes mistakes, my crew makes mistakes, and the world is a dynamic place. My job is to attempt to catch those errors to the best of my ability, or avoid those areas of risk that I can reasonably do so, while being urged by all and sundry to expedite, save time, money etc... While reasonably proficient at doing so, I also understand that prior performance is a poor predictor of future performance, and StuF happens; that make it interesting to understand what issues are being [re] invented out there to try and derail my [any your] polished performances.

Individual performance in almost all if not all tasks have temporal qualities, (what I like about the resonance concepts...), an example is the Capt who inexplicably continues to ask for a higher altitude than the aircraft performance will permit (on a standards evaluation flight...) yet after being labelled a "BOZO"... then proceeds to a few years later save 800 people when a well respected B744 operator (competitor airline) taxies across his runway on a MTOW departure in a fully loaded B744. Not one out of the numerous other crew in his aircraft, the other aircraft or ATC recognised the critical condition, but the "BOZO" does a reject and saves everyone...

Most people if brought back from the dead and given the last 16 seconds again would probably make different decisions... (the "Omega III" principle from "Galaxy Quest").

:)

[aterpster]

fdr:

Wearing Biohazard garments takes this discussion out of the realm of being "academic" [BOACs, Atapsters posit].

Being one of the ALPA accident investigators on the landmark TWA Flight 514 crash in 1974 resulted in me becoming more academic.:)

[sevenstrokeroll]

aterpster

was that the crash at Dulles/near dulles? IF SO, that crash contributed greatly to my understanding of the pilot responsbility during approach phase and to have a good idea about terrain.

[galaxy flyer]

That's it--Mt Weatherly, VA.

GF

[PBL]

It is astonishing to me - but really not, because I experience it a lot - that we come directly from talking about the causes of the Cali accident to relating the individual experiences and resumes of the discussants.

So let me briefly discuss here the tropes that have been raised, and then let's please get back to talking about the important point: automation and human interaction with it.

First, academics is my day job. I also do a bunch of other things. I run a small engineering company. I also have experience with aviation accident analysis that few if any other people on this WWW site have (no, not walking around wreckage fields, although I have handled this and that broken part sitting in a state transport accident investigation facility from time to time). Let's not put people into shoeboxes and assume we know everything about what they do and what they can do from their day job.

Let's get on to the tropes. Easy ones first.

There's a lot of over complicating a simple case here. I am a simple guy and like to keep things easy. ..... Consider the facts from a simple airmanship point of view. ..... All the 'causal contributions' came afterwards. The root cause was not mismanagement of the FMS, and Jeppe mistakes and Boeing mistakes with no auto-stow speed brake etc.etc. .... That to my simple mind is an airmanship problem. That is the root cause of many prangs.

So, from a "simple" point of view (and I agree wholeheartedly with the adjective!): (a) there was a failure of airmanship; (b) other stuff is secondary; (c) failure of airmanship is thus "root cause".

That is the classic trope "blame the pilot if we can".

(Just to be clear here, I am rephrasing "came afterwards" as "secondary", because of course these other things did not literally come afterwards, they were installed in the airplane before the flight and executed their function during the incident, not "afterwards".)

No one working in aviation human factors accepts such a line of argument. The reason is part (b). Why is other stuff "secondary"? What criteria are being applied to conclude that other stuff is "secondary"? That these phenomena do not contribute to "root causes" (plural please, not singular)? Because they are, demonstrably, causal factors.

It's easy to write a note on a WWW forum and say "I don't agree; I am a simple man, and I say the pilots screwed up and you can ignore everything else." One can just as easily write a note on a WWW forum and say "I don't agree; I am a simple man and I say 2+2=5 and you can ignore everything else." And the two statements are about equivalent in worth, whatever that is.

To terpster I would say: sorry that you don't think you can understand the Cali analysis. The problem lies certainly in our presentation, because I can assure you from experience that people with widely differing backgrounds understand very quickly how to perform rigorous causal analyses. A 6-hour day can suffice, although two is more usual.

But I would equally point out that it is naive to dismiss technical results because one cannot understand them. Aerodynamics is complicated engineering and math, which you cannot learn in a couple of days of practice, as you can causal analysis. But you believe the figures in your FCOM. So what's different there?

What is different is that everyone and her dog thinks they can perform causal analysis, and thinks their preferred opinion is immune to criticism. The first is wrong for most people (they don't have the necessary discipline and training, just like with aerodynamics); the second is wrong for everybody, always. No analysis is immune to criticism. Even though the analysis is rigorous, and only open to observation that one has made a mistake, which may then be corrected ("look, you have here 2+2=5; but 2+2=4!" "You're right! Here, we changed it"), the two following questions of a causal analysis are always valid: (i) why did you stop there?, and (ii) why did you summarise these facts like that / why did you formulate this phenomenon/these phenomena in that way? One can only answer "because X" and then X is -always- open to discussion.

There is a similar phenomenon with the puzzle known nowadays as the "Monty Hall Problem" (look it up in Wikipedia). It has to do with estimating probabilities during a guessing game (one guesses to win - or lose - some prize). The probabilities may be derived, by either frequentist methods or Bayesian reasoning, and you can even establish them by playing the game over and over again and looking at the results (Bayesians can even guess random probabilities at first and then use Bayesian updating on the results of these game plays to reach the conclusion that others reach by ab initio reasoning).

However, the probabilities are, mostly, incorrectly estimated by most people at first. The odd phenomenon is that people stick with that incorrect estimate, even in the face of both proof and experiment that it is wrong. And not inexperienced people, either, but sometimes even people with PhDs in math.

Because of this phenomenon, the problem has also been well studied by cognitive psychologists. It is an amazing phenomenon. There is a very good book about it, The Monty Hall Problem, by Jason Rosenhouse, Oxford University Press 2009, which I read last week.

Unlike the Monty Hall game, causal reasoning occurs at some level in everyone's life all the time. This probably explains why more people are able to give their causal judgements as "simple" people, and stick with them, ignoring argument which demonstrates that their view is unsubstantiated.

Causal analysis ain't like voting. It isn't a matter of opinion. "Simple" means "naive", and naivety is about as welcome in causal analysis as it is in aerodynamics.

Back to the trope (a); (b); (c) above. In my experience, the only expert people who seriously propose it are very senior lawyers in compensation disputes. No scientific expert on automation and human factors whom I know (and I really do know most of them professionally on some level) supports such a view.

PBL

[S76Heavy]

@ PBL:

there is however, a chain of events leading up to an accident.

Removal of any of those links prevents the accident.

So reminding ourselves that airmanship is our last resort even when the rest of the design and circumstances conspire against us, is not a bad thing in my opinion.

Fact is, that without the decision to change a good plan, none of the other factors would have been able to cause the accident.
They probably would have been part of another chain of events, leading to another accident or incident, but not this one.

So for us at the pointy end of whatever we fly, design of aircraft, equipment and airspace are a given that we cannot influence, only our behaviour and how we work with what we are handed on the day.

In this case, as in may other cases, the crew made poor choices allowing the rest of the holes in the swiss cheese to line up. And as human beings, we need awareness of why and how we are prone to making poor choices. Because we will always have to deal with stuff designed by people without the full picture of what it means to work with it. We are always the last line of defence.

[SR71]

It is surely necessary to understand PBL's and fdr's papers/commentary/reasoning if one is to advance the safety of the "universe" we, as commercial airline pilots, operate in.

Nothing other than a rigorous analyses of causal factors will result in a coherent strategy that is able to reduce incidents of a similar nature occuring in the future. It doesn't surprise me (rather it deperesses me) that PBL's team have found numerous mistakes in the logic contained within a significant proportion of aircraft accident reports.

However, inspite of my pretensions towards academia - PhD in Aerodynamics - I'm not sure it is necessary for me, to, for instance, retain an understanding of all 59 causal factors in the case of AA965 to help me do my job better in future when faced with similar circumstances.

But then my aim in reading the report is not quite the same.

My guess is most pilots will (excepting, perhaps, those who take an interest in contributing to discussions such as this) employ the same "economy of effort" they apply to executing complex tasks whilst airborne to their understanding of complex accident reports. To this end, they will endeavour to distil only one or two significant and salient ideas from the text to inform their modus operandi in future.

Whilst intending no disrespect, I see RAT's post along these lines. There is nothing rigorous about his commentary. The post serves as an admonition, to others and, perhaps, himself, that deviating from published procedures in the vicinity of high ground, in an unfamiliar location at 400mph comes associated with a degree of risk.

I used to know a pilot who started everyday in the flightdeck by asking the question:

"What don't we know about today that is going to kill us?"

:\

[PBL]

there is however, a chain of events leading up to an accident.

Actually not. There is a network (not a chain) of states and events (not of events alone) "leading up to" an accident. Rephrasing, the chain of causal factors leads back from the accident (when one considers causal factors of causal factors and so on iteratively), and the structure so derived forms a directed graph.


Removal of any of those links prevents the accident.


Indeed so. Removal of any of the necessary causal factors of an accident avoids the accident. That follows directly from the definition of necessary causal factor.

So reminding ourselves that airmanship is our last resort even when the rest of the design and circumstances conspire against us, is not a bad thing in my opinion.

That is an argument for the prioritisation of crew's competent handling of the situation over those factors which remain constant throughout the playing out of the accident event - states of the runway, states of the navaids, state of the weather, *but not* design of the automation. The "but not" comes from the observation that it is the *behavior of* automated devices during the playing-out that is most immediately causally involved: the design is the cause (causal factors of) the behavior, and the behavior in turn (some of) the causal factors of the accident. That is true for pilots also, BTW. The behavior of the pilots in the Cali case was causal; the training, experience and expectations were (essentially) causal of their behavior (modulo questions of psychological causality and free will). That is why terpster's putting the finger on the US piloting environment (as he did in his quote from his 1996-7 work) is essentially as accurate a partition of the particular set of causal factors he is interested in as the behavior during the playing-out.

It does not by itself allow one to prioritise the behavior of the automation over the behavior of the crew as a set of causal factors.


So for us at the pointy end of whatever we fly, design of aircraft, equipment and airspace are a given that we cannot influence, only our behaviour and how we work with what we are handed on the day.

Of course a professional pilot focuses on what heshe, as a professional pilot, can do to avoid.... Just as an avionics system safety specialist will focus on what heshe, as a professional system safety specialist, can do to avoid .... and an ATCO will focus on what heshe, as a professional, can do to avoid...

But an accident analysis takes all factors into account. An objective prioritisation of some takes application of an explicit criterion (such as that above) and in my experience such criteria are always open to question by others. But making the criteria explicit allows the discussion to proceed as to whether the criteria are appropriate (serve such-and-such purposes, or not).

People here might be surprised at how often this crew-behavior versus kit-behavior argument gets played out, often behind closed doors and for very high stakes indeed. To my mind, the arguments are evolving from "the crew screwed up; end of story" to "the crew played their role in an environment set by the behavior of the avionics, ATC, weather, and so on. How did each of these classes of factors contribute to the outcome? How did, for example, the avionics provide a benign or a hostile environment in which the crew could execute (what they understood to be) their tasks?"

It wouldn't surprise me if in fifty years we didn't think along the mirror image. That is, the designers of the kit can in peace, quiet and with all the time in the world think how their kit can best operate with help from the pilots, and when the pilots attempt to achieve what they think is task A but which has different outcome B, or to monitor what they think is task A but which turns out to be task B, the design is faulted because the pilots are operating under heavy resource constraints with the usual human weaknesses, and the designers should have accounted for it. (That won't happen until the execution of complex functions is as well understood and standardised as the T-display of essential flight data, but this understanding and standardisation is very likely to come about within fifty years, I think.)

PBL

[John Farley]

Chaps

Let us call those who say "it was pilot error - pure and simple" group A

Let us call those who say "it was a complex accident that happened because of all sorts of factors" Group B

May I suggest the best thing we can do in the SHORT term to reduce accidents is to brief Group A's position to as may pilots as possible.

However in the LONG term the only way pilot's performance will improve is through applying lessons learned from Group B.

I think A and B both have a lot to contribute to safety.

JF

[safetypee]

Here, here, JF.

And as James Reason put it:-
“We need to find some workable middle ground that acknowledges both the psychological and the contextual influences on human performance, as well as the interactions between active failures and the latent conditions that serve, on rare occasions, to breach the system’s defences.”

Who moved my Swiss Cheese? (www.asasi.org/2004_PPTs/Young%20et%20al_Human%20Factors_ISASI04_PPT.pdf)

[S76Heavy]

Agree, JF.

@ PBL: I'm by no means saying that all we should look at is how pilots can prevent poor choices which can lead to accidents. I am fully aware of what bad ergonomics for example can do to someone's performance, or other more complex "chains" find their origin in how the human tries to cope with a design, but as a pilot all I can do is what is in my limited scope of control.
As rightly identified by John Farley.

However, I'd be more than delighted to be able to pass my experience on to those designing the stuff that pilots will use and in fact, have already done so on a few occasions.

So to me it is not black and white, just long term and short term.
And learning how to survive short term so I can benefit from the long term
developments made by others.

[bubbers44]

The Cali and other incidents discussed here could have been prevented by pilot intervention and not allowing automation to do what the pilot didn't intend to do. When I declined the Cali straight in approach after the accident it wasn't because it was unsafe. It just didn't seem right to show how easy it was. A lot of people died doing that approach because the pilots let automation take them out of the loop. Once you are out of the loop you are along for the ride. I think the new 250 hr wonders are out of the loop. I think if they can push the right buttons they will succeed. If not, watch the news.

[Jane-DoH]

I'm of the belief that almost anything can be taken to excess, and automation is just another example.

[SR71]

Flight International Jan 18-24 has an article on the levelling off of aircraft accident statistics and cites work done/being done by FAA Human Factors expert, Dr Kathy Abbott.

She suggests inadequate crew knowledge of automated systems was a factor in 40% of accidents.

EXCLUSIVE - Qantas QF32 flight from the cockpit | Aerospace Insight | The Royal Aeronautical Society (http://www.aerosocietychannel.com/aerospace-insight/2010/12/exclusive-qantas-qf32-flight-from-the-cockpit/) includes an interview with David Evans, QF32 Check Captain.

I'm not sure we're learning the lessons with respect to automation as it took a crew of 5 over 2 hrs to get this jet back on the ground as a result of having to deal with 43, some contradictory, ECAM messages.

In the interview he is asked whether he believes the result would have been the same had there only been 2 crew onboard?

I think he is generous in his response, suggesting that it "may" have taken longer but the end result would have been the same....

How much longer?

(5/2)x2 hrs?

I'm curious as to whether or not there is an optimum amount of information that needs to be presented to a pilot to do his job and whether advanced flight decks are merely saturating pilots with information on the assumption that "more is better"?

Is there an argument for increasing minimum crew complements again because of this complexity?

[RAT 5]

Greetings, to my mind this was not an automation issue. Bubbers44 says, perhaps rather sarcastically, he didn't want to show how easy the NPA was. Indeed any NPA can be easy if planned, prepared and briefed properly. However, they are a disaster waiting to happen when attempted without proper preparation. The worldwide CFIT stats bear that out. Situational awareness goes out the window as you slide ever more behind the a/c. You are a passenger on a sled ride. Sometimes you are saved by busting out of cloud, in an unexpected place, breath a huge sigh of relief, wipe the seat and spray some airfreshner, then land. No-one except you and the FDM monitor is any the wiser. Another statistic avoided. On the other hand...
An a/c at altitude has an amount of energy; This has to be reduced drastically to land safely. Our god given computer used to do this; now we trust VNAV/LNAV, but hopefully monitor it to make sure it is doing what we would have done ourselves. It is a great reducer of workload, when used properly. In this case the a/c was too high on energy; it was too high, too fast and too short of time to reduce the energy sufficienly to make a safe approach and landing. That would have been true for a needles & dial a/c or an EFIS a/c. It's physics and aerodynamics. Situational awarness told them how far they were from the rwy, the altimeter told them their height and the ASI told them their speed. Added all together they had too much energy. Without proper planning they did not have enough time to set up safely. Why they made a decision to defy energy I have no idea. Were there causal factors at play, e.g. behind schedule, tight on FTL's, commercial pressure, poor training culture etc. etc. I do not know. Some guys can push the boundaries and so learn, by mistakes or by successes; some guys stay in the middle of the box and are considered stodgy, but safe. The hairs on the back of my neck are a good barometer. I've pushed boundaries in many things, but usually with an escape route. When I've got close to the edge, unitentionally (= mistake) it's scary. Knowingly going into a cul-de-sac usually alerts the hairy sensors and I back off. I just cannot see this having, as a root cause, a mis-managed automation conclusion. In daylight, airfield in sight, perhaps it was worth a go. At night, PF on 1st visit, IMC, high on energy, NPA, it just seems a very odd human decision to even atempt such a manouevre. A couple of holes had lined up when they started to consider a straight in. By saying "NO Thanks" the 3rd hole stayed closed and the cheese was back in the fridge, at least for this scenario. By saying OK the holes started to line up 1 by 1.
From the spectrum of posts I suspect this thread will rotate and continue without reaching a concensus. I'll read with interest, hopefully learn, but my tuppenyworth is spent.

[rogerg]

think the new 250 hr wonders are out of the loop. I think if they can push the right buttons they will succeed.

Maybe the 250hr wonder would realise that what he was being asked to do is beyond his experience level and would have declined the offer, as this crew should have done.

[PBL]

Well, the same question that arises on a half-dozen other threads seems to have chosen different venues for its emergence. A bit like thunderstorm cells, isn't it? One matures and dies within a fraction of an hour but the squall line persists. This thread seems to have died over people's positions on the Cali accident.

Let's see if we can resurrect the discussion.

Terpster wants to prioritise the crew's violation of the good arrival/approach practice, as laid down in the design assumptions for the arrival/approach (say, similar to TERPS). RAT and bubbers want more or less to join that view.

I argue for consideration of all the factors, and consider that view proposed also by fdr, SR71, and Dozy.

Even though we were arguing, these views are not necessarily in contradiction.

Alf pointed out that, besides the comprehensive analysis (which he termed "academic" - surely he knows better than that!), there is a need to provide guidance to the front-seaters. My view on that: yes.

I pointed out that there is a trope:
(a) there was a failure of airmanship; (b) other stuff is secondary; (c) failure of airmanship is thus "root cause".


"Airmanship", or "adherence to TERPS-like approach procedures", or what-have-you. There was a failure of two organs sitting behind the eyes in the front seat; we have agreed. That failure was, as I have argued, facilitated in a way that those who have not experienced suchlike will not necessarily readily understand.

Let me offer an extreme example. Suppose it is well known that, at one specific crossing in the middle of town, the traffic lights show red for "go" and green for "stop". Everybody knows it. They know to take it into account. Nevertheless, would we be surprised if more accidents take place at that intersection than elsewhere in town? I would suggest: we would not be surprised.

Now, suppose we look at causes. We have a specific accident in mind. Mr. Jones went through a green light and collide with Mrs. Smith. One might argue: (i) Jones knows about the light and the convention; (ii) he willfully went through a green light without sufficient care and attention; (iii) he hit Smith's car; (iv) his insurance pays. One might also argue (I) as (i) above; (II) the city council knows about the light, that this is a counterintuitive affordance (in Don Norman's use of the word), and the result of counterintuitive affordances, namely that people are induced to do the "wrong" thing, even though they might theoretically "know better"; (III) the council's insurance pays.

There are two issues here. One is causality, the other is responsibility (phrased here as responsibility for compensation, as it often is in real life). I have tried previously to separate the two, with limited success. Let me point out here that in terms of causality we cannot ignore any of (i), (ii), (iii) or (II). Ideally, AngloSaxon common law (for example) suggests in general that responsibility follows causality, but in practice (that is, in the actual legal environments of countries following this paradigm) other factors mitigate.

Automation provides affordances in Norman's sense, lots and lots of them, for example behavior of nav data bases in unanalysed situations (two navaids within reception range with identical ID and FREQ). And so do variant design of approach procedures (naming an approach after an end point, for example). Causal analysts cannot ignore those, as I have tried to argue.

There is a criterion that will prioritise crew behavior over other causal factors, as terpster, RAT and bubbers wish to do, namely that other factors, such as DB design, FMS design, approach design, are temporally prior to the playing-out of the accident event, and the behavior of the participants during that playing-out should be prioritised over the temporally-prior factors. The reason for this causal priority can be seen to lie in the phenomenonen that the participants are expected to avail themselves of free choice, and choices are available to them whose consequences are not the accident scenario. Furthermore, those choices (which they unfortunately did not make) are worth making for more than the reason that they would have avoided *this specific* accident.

Examples of these choices are: (A) follow the full approach procedure diligently, including the arrival; (B) don't descend below MSA unless following the arrival/approach diligently; (c) in the sim, practice cleaning up the airplane upon GPWS warning.

Difficulties with the application of this criterion are that, although the design of the DB/FMS/airplane automation in general temporally precedes the playing-out of the accident event, that design is a design for certain specific future behavior, and that future (at design time) behavior is part of the behavior during the playing-out of the accident event. Forethought is explicitly required of the designers: possible future behavior must be analysed and hazards avoided or mitigated.

Similarly, crew behavior, with its MttB characteristics, is heavily mediated by the flying environment in which they have operated (which is the basis for terpster's 1996 critique of the US regulators, which he reprinted here).

So in fact it is not so easy to discriminate on the basis of the temporal situation of factors. Because we look with different-colored glasses. For the crew, we look at behavior-at-the-time and fail to look so closely at what in the past has afforded their current behavior (rose-colored glasses). For the kit designers, we look at the decisions they made in advance, and not so much at the behavior-at-the-time of the kit (lilac-colored glasses). There are other participants, past and present: controller, approach designers, navaid-placement designers, etc - let me stop for the moment with rose- and lilac-colored.

I propose we should look with clear-colored glasses. This is where I think there is lots of material to continue the discussion. For I don't think this is all about a few variously-experienced people blabbering in their free time on an internet forum. I think it is a major research topic for those interested in reducing complex-system accidents in general and commercial-aviation accidents in particular. SR71's citation of Kathy Abbott's view is pertinent.

PBL