Go Back  Airline Pilot Central Forums > Airline Pilot Forums > Foreign
Emirates EK521 preliminary report released >

Emirates EK521 preliminary report released

Notices
Foreign Airlines that hire U.S. pilots

Emirates EK521 preliminary report released

Old 09-21-2016, 11:26 AM
  #21  
Gets Weekends Off
 
Joined APC: Nov 2009
Posts: 5,169
Default

Originally Posted by tailwheel48 View Post
Reliance on automation, in and of itself, is not a bad thing. If I was in the back of a Cathay B777 executing an engine out missed approach in Hong Kong, I would hope that the automation was working properly, and that the crew was using it. Quite frankly it's foolish to expect a pilot to hand-fly a complex arrival into a busy international airport after a long haul flight. All you're going to accomplish, is to overload the monitoring pilot.

For the handful of situations where automation has contributed to an accident or incident, there are likely thousands of 'situations' where automation has prevented something bad from happening.

For the most part, widebody long-haul crews may get to fly two or three legs a month. Trying to pretend that you're as proficient at hand flying as a guppy driver who flies forty legs a month into low risk airports is a fools errand.
As a professional pilot, it's expected you can do both. Fully utilize the tools of the airplane, and hand fly when they fail. Reliance on automation is a serious problem, as is not understanding how it works.
Grumble is offline  
Old 09-26-2016, 02:46 AM
  #22  
Gets Weekends Off
 
Skyone's Avatar
 
Joined APC: Nov 2005
Position: B777 Left
Posts: 736
Default

Originally Posted by Adlerdriver View Post
Come on. Blaming a company policy for that accident is a bit ridiculous. What if ATC told them to go around or the company policy for stable approaches wasn't met. Are you going to blame those things for the same type of accident? Any go around shouldn't be inherently dangerous, it's a routine maneuver.
They didn't fly the airplane. The reason for the go around (whether you think it's valid or not) has no bearing on the outcome. That's all on the crew.
Routine maneuver? When, may I ask, was the last time you did a go-around AFTER touchdown? Not touchdown after performing the go-around. In your last several proficiency checks, did you practice this maneuver or touch and go's? In twenty years of flying the 777/767/757 at three different airlines, not once have I been required to perform a g/a AFTER touchdown....20 ft., yes, resulting in a touchdown.

Secondly, you seem to have a lack of understanding of Dr. James Reason's Swiss Cheese model. A lot of negative training involved, also.

I'm not defending or blaming the crew, only that factors involved in this event are not evident at first blush. And basic training for children of the magenta line is not there. Finally, an airline that at one time was contemplating that all landings on a certain type of 4 engine aircraft be autolands......well tell me that this isn't one more lined up hole in the Swiss cheese with regard to relying on only automation.
Skyone is offline  
Old 09-26-2016, 06:31 AM
  #23  
Gets Weekends Off
 
Adlerdriver's Avatar
 
Joined APC: Jul 2007
Position: 767 Captain
Posts: 3,987
Default

Originally Posted by Skyone View Post
Routine maneuver? When, may I ask, was the last time you did a go-around AFTER touchdown? Not touchdown after performing the go-around. In your last several proficiency checks, did you practice this maneuver or touch and go's? In twenty years of flying the 777/767/757 at three different airlines, not once have I been required to perform a g/a AFTER touchdown....20 ft., yes, resulting in a touchdown.
July 2016......Jan 2016 before that......July 2015........Jan 2015...... and so on. We've been practicing bounded landing recoveries followed by a go-around at FedEx since 2009, especially in the MD-11. That training was adopted by the other fleets, so I regularly do it in the 777 recurrent training/proficiency checks as well. So.... yes, a routine maneuver. Maybe at some point in those 20 years, you could have suggested a change to the training? Do those airlines not solicit feedback on the training programs or have an method to allow changes to be suggested?

Originally Posted by Skyone View Post
Secondly, you seem to have a lack of understanding of Dr. James Reason's Swiss Cheese model. A lot of negative training involved, also.
Of course I'm familiar with the swiss cheese model. While somewhat dated, that concept is still valid in helping to explain why a particular accident happened. It does very little in terms of accident prevention. I suggest you (and maybe your company) consider some of the more modern safety concepts being advanced by people like Dr. Tony Kern. In general terms, this is a focus on error control on an individual level rather than attempting to create systemic barriers to error that can still eventually result in lining up the holes.

Originally Posted by Skyone View Post
I'm not defending or blaming the crew, only that factors involved in this event are not evident at first blush. And basic training for children of the magenta line is not there. Finally, an airline that at one time was contemplating that all landings on a certain type of 4 engine aircraft be autolands......well tell me that this isn't one more lined up hole in the Swiss cheese with regard to relying on only automation.
I suppose one of the "factors" involved that you refer to is that a professional 777 crew wasn't trained to push the power up (and make sure it stays there) for a go-around. Isn't that equivalent to training an NFL quarterback to throw a football or a Navy Seal to load his weapon? I guess there might be a problem with the hiring/screening process if a major international airline has to train their pilots to execute a basic, entry level skill like a go-around.

As long as airlines are going to allow various hiring practices that result in low-time, low-experience pilots flying modern airliners while mandating maximum use of automation, there will continue to be accidents. Creating "children of the magenta" and then having to go back and try to fix their shortcomings isn't a good plan.
Adlerdriver is offline  
Old 09-26-2016, 07:31 AM
  #24  
Gets Weekends Off
 
Skyone's Avatar
 
Joined APC: Nov 2005
Position: B777 Left
Posts: 736
Default

Originally Posted by Adlerdriver View Post
July 2016......Jan 2016 before that......July 2015........Jan 2015...... and so on. We've been practicing bounded landing recoveries followed by a go-around at FedEx since 2009, especially in the MD-11. That training was adopted by the other fleets, so I regularly do it in the 777 recurrent training/proficiency checks as well. So.... yes, a routine maneuver. Maybe at some point in those 20 years, you could have suggested a change to the training? Do those airlines not solicit feedback on the training programs or have an method to allow changes to be suggested?

Of course I'm familiar with the swiss cheese model. While somewhat dated, that concept is still valid in helping to explain why a particular accident happened. It does very little in terms of accident prevention. I suggest you (and maybe your company) consider some of the more modern safety concepts being advanced by people like Dr. Tony Kern. In general terms, this is a focus on error control on an individual level rather than attempting to create systemic barriers to error that can still eventually result in lining up the holes.

I suppose one of the "factors" involved that you refer to is that a professional 777 crew wasn't trained to push the power up (and make sure it stays there) for a go-around. Isn't that equivalent to training an NFL quarterback to throw a football or a Navy Seal to load his weapon? I guess there might be a problem with the hiring/screening process if a major international airline has to train their pilots to execute a basic, entry level skill like a go-around.

As long as airlines are going to allow various hiring practices that result in low-time, low-experience pilots flying modern airliners while mandating maximum use of automation, there will continue to be accidents. Creating "children of the magenta" and then having to go back and try to fix their shortcomings isn't a good plan.
Actually, I agree 90% with what you say. Airlines that fly the MD11 have learned from issues AFTER accidents with rejected landings. I got my type rating on the a/c in '95, and we didn't practice rejected landings. At that point there had been no evidence of a threat with a rejected landing. Mainly because most aircraft pilots had flown previously had not the level of automation sophistication.

I disagree with your assessment of the Swiss cheese. It is still very relevant in that factors beyond the day have major contributory effects on accidents. Better training, company policies, fatigue, SMS, just culture, perceived threats from management can all lead to the holes lining up. As pilots what can we do about it? Manage our environment; threat and error, leadership, decision making, communication, s.a., knowledge, etc.

But remember to take in the startle effect. Sometimes seconds are at sake and the startle effect takes more seconds to realign the cognitive functions. That's why GPWS, TCAS, V1 cuts, mins, etc., are practiced year in and year out to condition out the startle effect. But when certain maneuvers are not practiced, and there are not regulatory requirements to do so, well we see what happens. Evidence based training is finally making it's way forward. Already handling sims practicing rejected landings are being implemented. The new way forward in training is eye tracking. Evidence has proven with highly automated cockpits that the airspeed indicator is almost ignored. Why so many stalling accidents recently with the highly automated cockpits?

So finally, if you have never flown outside a FAA, EASA type agency, one might be surprised at the affect of culture on all aspects of the operation and pilots' actions, decisions, and overall thinking towards their "airmanship". I guarantee that when it comes to decision making, you and your colleagues very seldom take into consideration what FedEx management might have to say afterwards or the result on your career. The captain of the Mangalore 737 accident, when landing long, had these factors to deal with. Two warning letters with regards to hard landings previously, one more and fired, and any go-around resulted in an explanation to the chairman. Result was a long landing on a contaminated runway, to an overrun down a slope killing many. EK requires a safety report for ANY g/a for any reason. At FedEx, you probably have a deidentified FOQA. EK does not deidentify the data and requests explanations on "pinged" events. So for new captains, many times the mahogany table has a bigger affect on thinking than the mahogany casket.

Professional Standards committe? A number of us hAve tried to get something in place over the years. Met with deaf ears, because management wants control over ALL aspects from safety, training, standards, operations etc.

So, overall, we agree. But your airline does not have culture interfering with all aspects of operations. What concerns me is most accidents end up being pilot error. Unsophisticated managements see accidents rates drop because of their embrace of automation. Now if we could eliminate most pilot interaction......

Sorry for the length, but trying to explain why corporate culture plays such an important aspect in decision making. One cannot compare sophisticated unionized airlines like FedEx, DAL, AAL, UAL, BA, QANTAS, etc., to heavy handed management run airlines that the rest of the pilot world has to contend. But then again, I still have my pension intact here. Sorry, just had to throw that in.

Last edited by Skyone; 09-26-2016 at 08:15 AM.
Skyone is offline  
Old 09-27-2016, 06:05 AM
  #25  
Gets Weekends Off
 
Adlerdriver's Avatar
 
Joined APC: Jul 2007
Position: 767 Captain
Posts: 3,987
Default

OK Sky,
I get the management and the corporate culture issues are far more of a factor than we might appreciate in the US.

This whole line of discussion started when someone attempted to lay the blame for this accident at the feet of the management policy regarding long landing automation. My point was that a simple go-around decision (whatever the reason) isn't (shouldn't be) on par with a V1 cut or reject in terms of risk. Automation or not, startled/unexpected or not, if I pilot doesn't have the basic muscle memory and natural instinct to physically push the power up because he wants his airplane to climb......that's not management's fault (other than maybe hiring criteria).

I'm not saying the cheese model is irrelevant. I am saying that when it's discussed these days, it's usually with caveats that it has limitations.

Eye tracking detecting lack of airspeed cross-check doesn't surprise me in the least. Hence my continued emphasis on manual throttle use whenever appropriate (if company procedures allow).


Originally Posted by Skyone View Post
But then again, I still have my pension intact here. Sorry, just had to throw that in.
As do I (and our UPS brothers) thankfully.
Adlerdriver is offline  
Old 10-01-2016, 11:36 PM
  #26  
Gets Weekends Off
 
Joined APC: Nov 2006
Position: Early Retiree SWA
Posts: 354
Default

Well said, Skyone!

Kap
Kapitanleutnant is offline  
Old 10-09-2016, 12:56 AM
  #27  
Gets Weekends Off
 
Joined APC: Jan 2013
Position: Q400, B-737
Posts: 324
Default

I'm sorry, gentlemen.

I'm late to this conversation, but after running through this thread, I went back and read the preliminary accident report.
After all the back-and-forth over basic airmanship and appropriate levels of automation, I came back to the data point in the report that says that:
Between the initiation of a go-around/"missed approach" it was FIFTEEN SECONDS before power was applied!!!
Did I miss something?!?!
I'm speechless, frankly. I may fly a piddling regional turboprop w/ 10,000hp and only 76 pax, but I still FLY IT!
I understand that EK proceedures are automation heavy, but really?!?!
What kind of atrophy developed for the PIC to forget basic pitch/power/thrust-required relationships?
When I first started down this (cursed) career path, my mentor mentioned something he had an issue with:
At TWA, they were very focused on crew using the highest level of automation to gain maximum fuel savings. Spike observed that those crewmembers following the company dictates most closely were the ones most likely to bust their proficiency checks. (He hand-flew as much as he could.)
It's terribly unfortunate that EK follows the same mindset.
I just finished my annual check this evening, and both of my instructors over the last two days thanked us for making their lives so easy.
My normal SOP is to hand-fly to level-off, and below 10,000'. (We have no autothrottles, we have no restrictions). We do the entire US west coast, so that includes SFO, LAX, SEA and YVR. I will revert to automation if conditions demand, (temporarily).
Yes, it can be challenging, and I encourage my FO's to do the same thing, to gain experience and self-confidence. I realize that heavies only flying a few legs a month is different, (observed that in the USAF, (ick!)), but we're paid for our bad days, not our good ones. (As a previous poster eloquently observed).
I'm sure that the crew involved were professionals, but dammit,(!), we're also supposed to be "pilots", at some basic, fundamental level. It seems like they lost that.
Let the arrows fly.

Essentially, I'm with Alderdriver on this.
N19906 is offline  
Old 10-09-2016, 08:29 AM
  #28  
Gets Weekends Off
 
Joined APC: Dec 2007
Position: Window seat
Posts: 5,157
Default

IMO weak pilots flock to automation. Everyone needs to use it as appropriate. Sometimes it's not the answer but IMO the weaker guys are often slower to revert to manual flying. Instead they continue struggling with the automation when 10, 20, or 30 seconds of manual flying would stabilize the situation/automation confusion and allow resetting the automation to an appropriate mode. Instead they hesitate in the 'what's it doing now?' phase unsure of the automation and themselves.

And the flip side is they sometimes revert to manual flying, thereby increasing their workload, when there's a simple correction available with the automation.
Sliceback is offline  
Old 10-09-2016, 06:27 PM
  #29  
Gets Weekends Off
 
Joined APC: Jan 2013
Position: Q400, B-737
Posts: 324
Default

Now, for a more temperate comment.
(My apologizies, the initial reading of that report got me quite spun up last night.)

Will this play out the same way as Asiana at SFO?
I assume it was a professional, well-trained crew. Was there some assumption of auto-throttle modes that was flawed? Some automation that was assumed to be there, but wasn't? I'd consider a go-around once you've set the aircraft down, (WOW switches made), to be an extremely uncommon event.
We treat skips or high bounces as a missed approach, and default to our standard calls. (Our airplane is a bit fussy, and a challenge to land smoothly, so we wouldn't even bother trying to salvage a poor attempt.)
As for going around once you've set down because the airplane starts *****ing at you, (judgement vs. the computer?), ...
All I can assume, being ignorant of the 777's automation laws, is that they expected the throttles to be there, and they weren't, (disarmed/ deactivated by touchdown), and the time lost in recognizing the situation, coupled with the spool-up time, accounts for those fifteen seconds.

That's a looong time to be bleeding energy.

As for comments regarding EK's managment culture, I really wouldn't want to be working under that philosophy. Doing a "carpet dance" to justify your missed approach? Appalling.
We have a one-paragraph stand-alone statement in our FOM which explicitly states that the company supports a no-fault missed approach policy. If you go-around, for any reason, you will never be asked to justify your actions. Safety first.
N19906 is offline  
Old 12-03-2016, 11:53 AM
  #30  
Gets Weekends Off
 
joepilot's Avatar
 
Joined APC: Jul 2008
Position: 747 Captain (Ret,)
Posts: 800
Default

If I remember correctly, go-arounds in the 777 begin with pressing the TOGA button, (on the throttles) which both rotates the aircraft ( with the autopilot engaged) and gives you go-around power. Also, if I remember correctly, touchdown disconnects the autothrottles. I can fully understand why the crew was surprised that the autothrottles did not advance as they expected if they had never practiced a go around after touchdown.

Joe
joepilot is offline  
Related Topics
Thread
Thread Starter
Forum
Replies
Last Post
vagabond
Technical
8
09-09-2008 08:16 PM
C212135
Military
2
08-09-2008 05:35 AM
Typhoonpilot
Foreign
20
11-15-2007 12:51 PM
wannabepilot
Major
32
09-22-2007 01:53 PM
ERJ135
Regional
36
03-08-2007 07:35 PM

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On


Thread Tools
Search this Thread
Your Privacy Choices