Go Back  PPRuNe Forums > Ground & Other Ops Forums > Safety, CRM, QA & Emergency Response Planning
Reload this Page >

Airbus Official Urges Major Pilot Training Changes

Safety, CRM, QA & Emergency Response Planning A wide ranging forum for issues facing Aviation Professionals and Academics

Airbus Official Urges Major Pilot Training Changes

Old 16th Apr 2015, 18:04
  #81 (permalink)  
 
Join Date: Jun 2000
Location: last time I looked I was still here.
Posts: 4,509
'I can't see how a pilot could let skills decay to the point he lacks the confidence to disconnect at any time.'

Using the sim for a few minutes is not enough IMHO. It's a start, but not the end. In my apprentice years a visual manual approach was the norm and looked forward to. It was the company culture. Airfields with no approach aids, just let down procedures, held no trepidation. Night arrivals into NDB airfields with limited lighting; no hesitation. Visual approaches from downwind with an ILS waiting for the last minute confirmation you'd got it right; a joy. It was demonstrated by the captains and learnt at their knee; it was the culture.
In recent airlines the opposite is true. All that good stuff is frowned upon. The rapid expansion of many airlines, with crews over the horizon and at an extended arms length, the leash is kept tight to avoid any cowboys. It is not helped by a sparse local oversight at bases. Rigid SOP trained monkeys give some peace of mind to mission control, over the other horizon. It has been said that the increase in G/A's (unstable at landing gate) after the winter, as soon as CAVOK arrived, was startling. That is time & money to tight schedules and tough profits. Restrict such adventures is the quick fix answer. That's todays real world. However, if the culture had been different in the beginning (chicken & egg) perhaps the problem might have been procatively avoided; perhaps. The rate of expansion of many airlines has proved too fast for such philosophies. Sadly, therefore, I suspect the sea-change in culture will not happen because the task will be too formidable.
A few moments in the sim will be for nothing unless it is consolidated on the line on a daily basis. What I would prefer to see, as a step forward, is a massive improvement in the basic type rating training of many pilots: more in depth training of how the a/c flies and what it is capable of. Dilute the SOP's and allow the crew to use the a/c systems and capabilities to perform the task required. Professional choice & behaviour. There are various methods of completing a G/A once the a/c is safely transitioned from approach to climbing. What is required? It is a CAVOK day; ATC call a late G/A; you clean up to intermediate flaps, fly visual down-wind and land. Simple, easy, un-rushed. Some airlines have only 1 G/A profile. The whole 9 yards have to be accomplished and a full approach set up performed. Thus the crews do not know the various options available. They have not been trained in depth about the a/c systems & capabilities and are not allowed to use them. All this has nothing to do with more hand flying, it is to do with understanding and being able to control your steed as you see fit and as required. More manual flying with trained monkey SOP's will not provide too many gains. An increase in knowledge and understanding, and freedom of professional choice is the foundation that could well be the start of a worthwhile change. I wonder how much the management really trust their captains to make those judgements.
RAT 5 is offline  
Old 16th Apr 2015, 22:41
  #82 (permalink)  
 
Join Date: Sep 2014
Location: Canada
Posts: 1,257
In my apprentice years a visual manual approach was the norm and looked forward to. It was the company culture. Airfields with no approach aids, just let down procedures, held no trepidation. Night arrivals into NDB airfields with limited lighting; no hesitation. Visual approaches from downwind with an ILS waiting for the last minute confirmation you'd got it right; a joy. It was demonstrated by the captains and learnt at their knee; it was the culture.
We all like to reminisce about the "good ol' days"... but the fact is, fatal accident rates "back then" are also horrendous compared to today, despite the fact that the amount of air travel over the years have grown immensely.



Indeed, even accounting for Germanwings, aircraft accident rates today are at a historic low.

There is no doubt in my mind that the use of automation have had a large positive impact in aviation safety overall.
peekay4 is offline  
Old 17th Apr 2015, 06:35
  #83 (permalink)  
 
Join Date: May 2004
Location: Somewhere in nowhere
Age: 44
Posts: 4
AF447 is often used as the prime example by people that are expressing an opinion on the topic of deteriorating flying skills. The fact of the matter is that the aircraft was flown into the water by an overwhelmed first officer in the context of lost situational awareness, multiple system failures/indications, heavy buffeting from the weather (as well as the stall), zero external horizon with limited time to take meaningful corrective action that could save the aircraft.

Like any accident the final impact was a culmination of a chain of events with multiple lost opportunities and should be considered to be a total system failure that started with the temporary inconsistency in airspeed indications. Apart from this the final report puts heavy emphasis on persistent errors made by the crew (inappropriate control inputs that destabilized the flight path, failure to recognise approach to stall, etc., etc.).

Personally for me this this accident was all about complete loss of situational awareness and nothing to do with whether or not the pilots could hand fly the aircraft or not. Again my personal view is that the weight of emphasis placed on crew error by the final report is clearly unfair.

I wonder how well some of us would have dealt with the same chain of events. The luck of the 228 passengers and crew ran out that dark night over the South Atlantic.

Last edited by fastmover123; 17th Apr 2015 at 08:01.
fastmover123 is offline  
Old 17th Apr 2015, 06:38
  #84 (permalink)  
 
Join Date: Jun 2000
Location: last time I looked I was still here.
Posts: 4,509
"We all like to reminisce about the "good ol' days"... but the fact is, fatal accident rates "back then" are also horrendous compared to today, despite the fact that the amount of air travel over the years have grown immensely."

I'm talking of the years 1980's before VNAV/LNAV a/c. If I read your Boeing graph correctly it looks very level for that period; and my memory confirms it. What did not happen too often was a crew, on approach, spearing in with a completely serviceable a/c.
RAT 5 is offline  
Old 17th Apr 2015, 06:45
  #85 (permalink)  
 
Join Date: Jan 2001
Location: In the boot of my car!
Posts: 6,000
There is no doubt in my mind that the use of automation have had a large positive impact in aviation safety overall.
All due respect but that is the statement of the year Obviously improving automation and navigation systems will make flying safer

Is this a debate on whether you should hand fly or use automation? I don't think so?

its a debate on whether crews rely too much on automation and in that process are ignoring basic handling skills.

with aeroplanes everything is fine while its fine its when things go wrong that a pilots skills have to be second to none.

To keep those skills honed a pilot needs to handfly the aircraft as much as possible when conditions allow.
Are you saying that keeping those skills honed somehow will increase the accident rates ?

If you are you might as well take that argument to a conclusion and go fully automated replacing the in that case the joke of a name like PILOT for a computer monitoring nerd " This is your CMN speaking" Not pilot /Captain

Last edited by Pace; 17th Apr 2015 at 07:08.
Pace is offline  
Old 17th Apr 2015, 07:12
  #86 (permalink)  
 
Join Date: Sep 2014
Location: Canada
Posts: 1,257
Are you saying that keeping those skills honed somehow will increase the accident rates ?
I am saying that despite pilots today being perceived to have less skill, we are in reality still much safer compared to the "golden-years" era some here like to reminisce about. And automatics is a large part of the reason why.

A ton of those "highly skilled" pilots from the 1960s and 70s bought the farm, at an accident rate which would be completely unacceptable today! Some just choose not to remember it that way.
peekay4 is offline  
Old 17th Apr 2015, 07:33
  #87 (permalink)  
 
Join Date: Jan 2001
Location: In the boot of my car!
Posts: 6,000
I am saying that despite pilots today being perceived to have less skill, we are in reality still much safer compared to the "golden-years" era some here like to reminisce about. And automatics is a large part of the reason why
But that is an obvious statement!! In the 50s racing a racing car was a death sentence with open wheels, sitting in a fuel bucket, no seat belts, crash structures etc.

Hardly a comparison to modern racing cars with all their driver aids and safety features.

But you still need a racing driver with the best skills and those skills will be needed when things go wrong with the automatics and take my word for it they do go wrong auto pilots included

With the advance of systems and automations there are still accidents where the pilots could have recovered the situation by solid piloting skills but didn't statistics or no statistics

The same goes for GA where the more automation the more needless accidents occur but I add that in GA those skills were never really learnt and some of the pilots used automation to cover a lack of skills a dangerous route to take

so not really sure what you are arguing ???
Pace is offline  
Old 17th Apr 2015, 09:08
  #88 (permalink)  
 
Join Date: Jan 2010
Location: Marlow (mostly)
Posts: 239
Manual practice on the line at the right time

The dilemma for airlines of how to balance the need for pilots to maintain manual proficiency while still using available automation at the appropriate time and in the right place is addressed in a bit of detail at Approach management | PICMA . org, in the second half of the discussion.
slast is offline  
Old 17th Apr 2015, 09:59
  #89 (permalink)  
 
Join Date: Nov 2010
Location: USA
Posts: 96
From the Wall Street Journal article:

“We do a lot of checking” of the same required maneuvers and emergency procedures each year, Mr. Nelson said, “but we don’t do much teaching.”
In today's "technology world" many humans including many pilots are only "learning" to be button pushers.

Many humans believe Smart phones make them smart when the opposite is true.
airman1900 is offline  
Old 17th Apr 2015, 11:25
  #90 (permalink)  
 
Join Date: Jul 2003
Location: An Island Province
Posts: 1,116
Most areas of industry recognise that there is an emergent safety issue in modern operations, but not always agreeing as to what the issue is or mitigation required.
The Airbus paper calls for action within current training activities, particularly those associated with regulatory requirements and guidelines for operators – training/checking. Manufacturers have little influence in these areas, but for initial type training they can make significant contributions.
Airbus has taken the initiative, well done. A new aircraft type provides opportunity for change, no doubt balanced by thoughts of common type ratings which other manufacturers might rigorously adhere to – avoiding the expense of changing training.

A key item is the need to move from the regulatory based checking to teaching. Such a change is in the hands of regulators, local authorities, and operators; thus the paper provides a mild rebuke to these agencies for the apparent lack of action.
The industry is unlikely to find more time for training, nor money, thus any change requires alternatives for the efficient use current of resources. I agree that the industry needs to teach, but teach what and how. We have to move away from training for checking, and check again and again; pilots need to develop expertise relevant to current operations.

This does not necessarily mean more hand flying; there is little data indicating that the industry has a hand flying problem in normal operations – it’s incredibly safe. However, there does appear to be a problem associated with automation, and reactions to failure or unplanned situations – non normal operations. The latter point relates to managing automation and assessing situations with sufficient understanding to act appropriately. It will be difficult for pilots to absorb all of the knowledge required for these activities during type training, nor have sufficient experience to apply knowledge, thus there is a need to develop these skills over time, with operational guidance, teaching, and practice. An apprenticeship, continued learning, and knowing how and what has to be learned.

Hand flying can contribute to this, but only within the limitations of the need to use automation. We cannot escape the need for technology and automation in modern operations, thus training change and teaching must be automation focussed, together with situation awareness, and the reaction to and control of non/abnormal situations.

Well done Airbus for leadership, but how do we to get others areas of the industry to act.
alf5071h is offline  
Old 17th Apr 2015, 11:37
  #91 (permalink)  
BRE
 
Join Date: Mar 2007
Location: Europe
Posts: 213
@ peekay

Any idea what the definition of accident is for either of these organisations? The Boing numbers are on average 2x higher than ASN's and also show more fluctuation. BAAA (whoever that is) must be counting completely different events...
BRE is offline  
Old 17th Apr 2015, 14:16
  #92 (permalink)  
 
Join Date: Jan 2010
Location: Marlow (mostly)
Posts: 239
Different data sets

BRE:

Yes these are different data sets but convey the same basic message.

Boeing: If taken from the standard Boeing charts, their 2013 numbers would be "The accident statistics presented in this summary are confined to worldwide commercial jet airplanes that are heavier than 60,000 pounds maximum gross weight. Within that set of airplanes, there are two groups excluded:
1) Airplanes manufactured in the Commonwealth of Independent States (CIS) or the Union of Soviet Socialist Republics (USSR)
are excluded because of the lack of operational data; and
2) Commercial airplanes operated in military service. (However, if a military-owned commercial jet transport is used for civilian
commercial service, those data will be included in this summary.)

"Aviation Safety Network" : covers accidents and safety issues with regards to airliners, military transport planes and corporate jets."

BAAA (Bureau of Aircraft Accidents Archives)
[It] is considered as an accident any event where aircraft suffered such damage that it is not in a position to be used anymore and that it is removed from service. In any case, aircraft should be considered as damaged beyond economical repair.

Is considered as an accident any event involving an aircraft certified to carry at least 6 people, crew included. Are not considered helicopters, balloons, hot air balloons, airships, gliders, fighters, and all other aircraft which does not correspond to the criterions mentioned here above.

In military aviation, only are considered accidents involving aircraft intended for troop transport, reconnaissance, surveillance and logistical support, provided they are able to carry at least six people.
slast is offline  
Old 17th Apr 2015, 14:17
  #93 (permalink)  
 
Join Date: Aug 2005
Location: fl
Posts: 2,561
The Boeing line on the graph shows over 40 crashes per million departures in 1960 dropping to 4 in 1974. I joined the airlines in 1979 and we had primitive autopilots that would hold a pitch attitude, altitude hold, and couple a basic ils approach.

In 40 years that 1974 data has barely changed. All the automation we now fly with is handy but we are starting to become automation dependent and that is not good. Pilots need to be able to use this automation at the level they choose, not be mandated by SOP's to turn it on at 300 feet and leave it on until 300 feet on landing. If they do they will soon lose their basic flying skills. Then they will be automation dependent.
bubbers44 is offline  
Old 17th Apr 2015, 14:42
  #94 (permalink)  
 
Join Date: Jul 2001
Location: Australia
Posts: 4,942
---- was poor instrument flying ability in the first place.
Centaurus,
Absolutely agree, and I can't think of any substitute for hand flying to maintain a decent instrument scan, the heart of any instrument flying. Looking at the panel on A/P doesn't cut it, and never did.

peekay4,
Still waiting for you to tell me the names of the airlines using strengthened biz-jets for unusual attitude and recovery training.
LeadSled is offline  
Old 17th Apr 2015, 15:13
  #95 (permalink)  
 
Join Date: Sep 2014
Location: Canada
Posts: 1,257
Any idea what the definition of accident is for either of these organisations? The Boing numbers are on average 2x higher than ASN's and also show more fluctuation. BAAA (whoever that is) must be counting completely different events...
Slast has it mostly covered. The only thing I'll add is that the ASN numbers only consider civil aircraft capable of carrying 14 or more passengers.

In summary:

- Boeing stats are the most restrictive, only considers civil aircraft > 60,000 lbs MTOW
- ASN stats include civil aircraft carrying 14 or more passengers
- BAAA stats further include civil and military aircraft with 6 or more total seats

In 40 years that 1974 data has barely changed.
Don't let the graph-scale fool you!! The Boeing accident data after mid-1970s looks stable only because the prior accident rate was so bad they skewed the chart!

But if we could "zoom in" we would see that the accident rate has continued to dramatically improve over the years. E.g., in the 1990s the accident rate still hovered around 1 per million departures, while fast forward to 2010 the rate has dropped to around 0.4 per million departures. That's a ~ 60% reduction just from the 90s!

(Consider that we now have 30+ million commercial departures per year!)
peekay4 is offline  
Old 17th Apr 2015, 15:15
  #96 (permalink)  
 
Join Date: Feb 2008
Location: In the Old Folks' Home
Posts: 396
Personally for me this this accident was all about complete loss of situational awareness and nothing to do with whether or not the pilots could hand fly the aircraft or not.
Yes, it was about loss of Situational Awareness (SA), but hand flying and SA are mutually interdependent. If you don't understand the situation, you can't adequately fly the airplane and manual flying helps you to maintain SA. Apparently the newer generation of pilots treats flying the airplane like playing a video game and they don't become a part of the airplane. It really shocked me to read the part where one of AF447 pilots commented to the others that they were going to crash. He had gained partial SA but about five minutes too late.

Early in the discussion of AF447, a non-pilot, who is a software designer commented that the auto-trim feature that had assisted the PF in pulling up into the fatal stall is such a good thing because it relieves the pilot of the task of trimming. The software designer did not understand that trimming is not an arduous task and actually is something that helps pilots maintain SA. Auto pilots and auto-trim are great for relieving the tedium of long hours of cruise but flying by pushing buttons does not promote SA. I contend it is just the opposite.
Smilin_Ed is offline  
Old 17th Apr 2015, 15:28
  #97 (permalink)  
 
Join Date: Sep 2014
Location: Canada
Posts: 1,257
Still waiting for you to tell me the names of the airlines using strengthened biz-jets for unusual attitude and recovery training.
I don't know about usual attitude training (Piper_Driver wasn't talking specifically about that) -- but e.g., Singapore Airlines via their training subsidiary SFC operates five Citation Mustangs with modified cockpits for their cadet program. They previously operated highly modified Learjets for a number of years.
peekay4 is offline  
Old 17th Apr 2015, 16:46
  #98 (permalink)  
 
Join Date: Jul 2008
Location: USA
Posts: 21
AF447 and Automation, also in the BEA report you will find mention of the C Chord, static position error and an incorrect down vsi accompanied by an incorrect loss of indicated altitude. All of these are additional factors helping to lead to disorientation. In that same report you will find reference to other operators experiencing the same undesired aircraft state with varying outcomes. For some it was a nonevent while it was a bit of a ride for others. It was not that long ago that jets were routinely hand flown to cruise, allowed to get on speed and in trim, and then given over to the unsophisticated, by today's standard, autopilot.

Then there is the issue of the Flight Director bars. How many times have you yourself or someone you are flying with followed the flight director bars when they should have been disregarded. You know, the mode was not appropriate. Manage the automation, and workload. We return to a discussion that was started about 20 years ago. Use the appropriate level of automation.

Now look at the BEA final report page 96. Then correlate your understanding of that graph with the timelines elsewhere that show the AOA. Pitch commanded up when it should have been down ...

A strong and well practiced sense of energy management would have helped these pilots to maintain pitch and power while sorting out the issue, wherever and however that sense was attained. And to have guided them in the right direction as to which elements of the automation were erroneous.

If the automation never took a vacation or lied there would be no concern about automation dependency.
noalign is offline  
Old 18th Apr 2015, 07:16
  #99 (permalink)  
 
Join Date: May 2005
Location: middle of nowhere
Posts: 348
It's a matter of definition.

What does 'hand flying' really mean? Most think of simply AP off and then cling to the FD (which in some designs is the same computer signal). Todays pilots are great in FD hand flying. Most manual handling sim sessions hastily introduced have the one engine ILS manual approach in their program then followed by a raw data EO ILS. The whole trick is that the pitch, power and crab values have been established on the previous FD approach and memorised. Big deal with no value!

My definition of hand flying in this context is the one of failure of automation, be that AP/FD, or input data affecting a lot of presented data. In most accidents the very basic information, attitude and altitude, were still correct. Secondary data like speed, vertical speed, trends, FD signals, FMS signals etc. were partly or grossly wrong.

Hand flying a complex aircraft with misguided information therefore means the ability to establish a safe flight path with very basic information.

It's exactly that kind of skill that is lacking. I test all my fellow pilots in the cockpit randomly by asking in any kind of flight situation what pitch, bank and power would be required, fast, right now .......
A frightening number of buddies does not have a fast enough or survivable clue!

The argument that the FO on AF447 was not lacking handflying skill, or that did not matter, is tragically wrong. He was definitely confused and lost situational awareness and the conclusion that even better hf-skills would not have helped him is almost cynically wrong.
Follow my definition of what hand flying all contains, it is the very nature of confusion and loss of situational awareness that defines absence of experience and skill that does not allow such pilots to get out of a situation where automatics go bezerk.
Gretchenfrage is offline  
Old 18th Apr 2015, 08:15
  #100 (permalink)  
 
Join Date: Sep 2014
Location: Canada
Posts: 1,257
In most accidents the very basic information, attitude and altitude, were still correct. Secondary data like speed, vertical speed, trends, FD signals, FMS signals etc. were partly or grossly wrong.
In most accidents involving commercial airliners, there's nothing wrong at all with the airplane, or with the sensors, or with any indications, or with any automation.

Boeing estimates that up to 80% of all crashes are caused solely by human error. The vast majority of these are pilot errors, but this category also include errors made by ATC, and errors made by mechanics.

So the largest safety initiatives are focused to reduce human error, e.g., by: 1) improving the pilot decision making process; 2) using tools & automation to help prevent mistakes.

Improving manual handling in response to possible mechanical issues, while important, would only factor in a diminishing minority of accidents.

This wasn't always the case. Per Boeing data, in early days of aviation only 20% of crashes were caused by human error, while 80% were caused by machine causes.

Today, the situation has reversed. 80% of accidents are now caused by human errors, while only 20% of crashes involved any mechanical issue.
peekay4 is offline  

Thread Tools
Search this Thread

Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service - Do Not Sell My Personal Information

Copyright © 2018 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.