Asiana flight crash at San Francisco

[tilnextime]

What a number of us have tried to argue on this forum is that a visual approach, especially at SFO, may not be as simple as it appears.

I have flown into SFO enough to know that it can be very challenging ie landing a large jet when being given inappropriate speed/height constraints.

Obviously, a visual into SFO, is not something that a majority of crew thinks is unreasonably dangerous. If that was the case, fine.

If the NTSB does a proper investigation, it will make their determinations based on standards, not how things "appear", level of "challenge" or what a "majority" think. It's not a matter of trying to identify possible pieces of cheese to line up in the manner most supportive of making it look as if the mishap was inevitable. If a given "piece of cheese" was in the pilot's hands to begin with (i.e. - prior knowledge of a having to perform a given approach) then it is how the pilot places that cheese in the stack that is the contributing factor, not the mere existence of the cheese.

An approach is either executable within in the established standards for the aircraft and crew or it isn't. A mishap investigation does not deal in "near the limits", but within published limits. If "near the limits" is unsafe, then the limit needs to be corrected. If a given limit hasn't been established taking normal operator frailty into account, then both the operator standards and the aircraft/maneuver limits need to be revisited. As my late uncle would say, "There is no such thing as almost pregnant". The flight was either assigned an approach that was within the standards for the type aircraft and crew, or it wasn't. That determination has to be made on sound technical info, not feelings or personal opinion. Otherwise, it would be impossible to determine what is a safe maneuver.

You can't ascribe causative or contributing factors until you establish the standards the aircraft and crew are held to, and the standards they failed to meet. In my investigating experience, there were no end of maneuvers that were based upon proper standards that were only challenged when a crew member failed to perform, or failed to admit that he couldn't perform, until after the mishap.

And yes, I investigated mishaps where the commander should not have assigned a given mission profile to a given crew or crewmember. However, the crew has mouths, and are not absolved unless they spoke up. Embarking on a mission or maneuver that is beyond what you think is your ability level cannot be absolved, unless there is a real, loaded gun at your head. "Job security" is not a real, loaded gun when the lives of others are at risk. It is simply the same concern for your pocket as that which drives the "management" you condemn.

I said it before and will say it again, until pilots' unions take job actions for higher standards, to include more hands on time, the existence of these safety threats are not the sole sin of management. In effect, while the trend in "management" thinking may be cutting costs, there is a trend in pilot thinking is cutting workload and responsibility - and thus, wanting simpler or mandatory automated approaches, for example, to make up for lesser hands on capabilities.

The thing is that you can't have it both ways. Without standards, there is chaos.

[John Farley]

tilnextime

Well said sir.

[nitefiter]

Brilliant, I hope the training depts are reading that one.

[BBK]

tilnextime

As you have quoted me may I respond. Firstly, I think your post is thought provoking and intelligently argued. I merely posted to give an operator's perspective. I don't have an agenda and I am not suggesting a methodology for conducting the accident investigation. That is not an area I have any experience in.

I hear what you are saying about standards etc and you may have a good point there. However, is it wrong to ask whether some of the manoeuvres pilots are asked to perform are "sensible". Perhaps I am seeing shades of grey where you are seeing rules as black and white. I will give you an example. My company, based on flight data monitoring, has sent representatives to ATC at some of the airports where there is a clear trend towards unstable approaches. Of course the buck stops with the commander and if he sees an unstable approach then whether PF/PNF he MUST ensure a go around is flown. However, is it really a good idea to have been given vectors that made such an event all but inevitable in the first place. Note these are not approaches that are illegal in any way. I believe not many companies have taken the initiative in this respect and it has been welcomed by the ATC departments concerned. Sadly, I can't take any credit here personally.

As for the unions taking a proactive approach, no pun intended, then I wish them well. In the UK BALPA are doing exactly that with respect to Flight Time Limitations or at least trying to. Regarding training syllabi I am not sure whether they can influence them that much. It is, I agree, an excellent idea. Apologies, to all, for the thread drift.

[bobcat4]

The landing checklist was not completed until 500 feet

If SOP in VMC says ”stable at 500 ft”, and assuming ”stable” includes ”landing checklist complete”, that seems a bit on the edge of being stable. Although we know they weren’t stable, they obviously thought they were (otherwise they would go around for a second attempt).

So at 500ft they completed the landing checklist and concluded they were stable. I guess they missed the point “engines spooled up” probably because they trusted the A/T.

Now, in VMC with the required “stable at 500”, the landing checklist MUST be completed at (or rather before) 500. If flaps cannot be fully extended (or flap 30 selected) at 180 knots, and they had to maintain 180 to 5, the workload seems to pile up. Landing configuration and landing checklist to be completed in less than 90 seconds. Looking at the data from FlightAware, the Asiana pilots delayed the slow-down from 180 at 5, shrinking the time to get stable.

[BOAC]

Originally Posted by tilnextime

An approach is either executable within in the established standards for the aircraft and crew or it isn't.

- first we need to establish these. Who does that? Manufacturer? PPrune? Regulator? Accident/incident records?

Example: Earlier someone somewhere said "250kts at 10,000' at 30 miles is on 'on the profile'. Which profile? Yes, I know it is fine, but who decides?

Regarding SFO/ATC - if, as it appears, they 'expect' and accept a speed reduction at 7 miles from 180 as posted here. why not say "180 to 7"?

[tilnextime]

BBK

If the standards are not appropriate for getting the job done safely, then by all means, correct the standards. However, before one reaches a conclusion that all airports must have instrumented G/S facilities, one has to ask why. If the answer is to compensate for low proficiency crews, or to compensate for possible crew inattention, then the question is whether or not it is "sensible" to continue increasing automation reliance to address a crew problem arising from automation reliance. Now, that's a question I pose, not an answer.

One significant human factors problem I have dealt with as a safety officer, flight instructor, crew member, mishap investigator and unit commander is that pilots do not want to be measured "pass/fail" on set flying standards. We want the "system" to rate us based on what my first CO called the "Ace Factor" - the guy who is smoothest on the controls is the best, or the guy with the most hours, etc, etc. But for 99.9% of the flying that is done, we don't need "best", we need "fully competent", not just in stick and rudder wiggling, but planning and judgement. It's not a contest. Every maneuver has parameters - the "plus and minus" factor. Keeping it all in 'the green arc" is competency, everything else is testerone driven overkill.

[MPN11]

Originally Posted by BOAC

Regarding SFO/ATC - if, as it appears, they 'expect' and accept a speed reduction at 7 miles from 180 as posted here. why not say "180 to 7"?

It appears from earlier posts that is "accepted" that an early speed reduction from 180/5 to 180/7 is OK. Would the Asiana crew have that perspective? Especially when the PF is 'under training'? What would be the IP's opinion if the deceleration started 2 miles earlier than instructed? Would the PF not try to comply with 180/5?

[tilnextime]

BOAC:

- first we need to establish these. Who does that? Manufacturer? PPRuNe? Regulator? Accident/incident records?

With the exception of PPRuNe each of the entities you mention, along with others, has a specific role. I exclude PPRuNe, simply because it has no formal accountability. Rather, I would suggest that formal, accountable, pilot input is needed.

I read post after post about pilot fatigue, yet not one post about ATC controller fatigue. A crew of 4 flew from point A to point B, knowing before takeoff that they would most likely have to make a visual approach. Meanwhile, the controller handling them was dealing with a long string of aircraft with no idea of whether any one of them had lifted a single finger to prepare for the PUBLISHED peculiarities of the airport.

Pilots are not the final word on what an aircraft can safely do. I have seen and or investigated far too many mishaps where well intentioned pilots made maneuver parameter or operating procedure decisions based on their cockpit view, and not the engineering principles used in the design and manufacture of the aircraft, for example. In one case, a "home brew" operating procedure that made the operation of a given helo "easier" for the pilots caused structural stresses that were, after repeated conduct, catastrophic.

Aviation is not just about flying. If it were, it would be so much simpler.

[BOAC]

All valid questions, MPN, and it is to be hoped the NTSB might look at this. I recall LGW accepting an early decel from "160 to 4", but I feel it would be much easier if firm gates were imposed ie 180 to7/160 to 5 etc, and promulgated and not 'nudge nudge wink wink' as it appears at the moment.

[Alexander de Meerkat]

In a sea of woeful ignorance in the last 112 pages, there have been one or two nuggets of gold - principally from 777 drivers who actually know what they are talking about. No doubt much will be made as to how the crew got into the situation they did and that will be a debate for another time. The key thing that stands out to any professional pilot is that the crew in question ignored the stabilised approach criteria and were allegedly descending at nearly twice the normal rate of descent below 500'. For any professional pilot that would necessitate an immediate go-around and living to fight another day. In any normal airline it would be a non-event and the most the crew could expect would be a question as to why they waited so long to initiate the go-around. Sadly in this case the worst excesses of Korean culture (and there are many very positive elements to this great nation) may have come into play. At the exact moment a regular western First Officer would have been shouting his head off to go-around, it may be that an over-polite and deferential group of three other pilots chose to watch a dangerously decaying situation become irrecoverable before the call was made to go-around.

How you defeat such a cultural problem, whereby it is rude and unacceptable to question someone in authority, is hard to say. Perhaps the answer lies with the Koreans themselves who surely must begin to question why they have had so many accidents over the years despite being such highly educated and intelligent people. The two other questions I would be asking is what happens to someone who does question authority in a decaying situation, and do military pilots receive preferential treatment in terms of advancing up the seniority list? Do pilots who have questioned authority get 'sorted' at some later stage in sim checks or experience general rejection by the rest of the pilot community? The sub-question I would then ask would be how the ex-military pilots conduct themselves in terms of CRM (I am ex-military myself by the way). To an extent these are all taboo areas in a cultural sense but they are so critical to understanding what went wrong here.

[BBK]

tilnextime

I'm not sure that your comments regarding the ace factor reflect a typical civilian airline. Maybe on a squadron perhaps? In fact it sounds like a lot of what you are describing might be more applicable to military aviation.

If I have given the impression that SFO ATC are acting unprofessionally then I certainly did not mean to. It is a very busy airport but I cannot help thinking whether some of the procedures are just pushing things in a direction that is not conducive to flight safety.

BOAC/MPN

I do hope that the local procedures are looked at in detail. It is a grey area I accept that but if Human Factors are an issue then by definition it is complex. I asked an HF psychologist why the stats were so rigorous in her field of work. She replied that because we are so "woolly" you need to some serious number crunching to try and distil the real effects being observed.

[Fox3WheresMyBanana]

There is an extradition treaty between S. Korea and the USA.

[TOGA!]

Here you go Vasco. Happy? Same argument stands. Flaps 25, 180/5. Flaps 30, slow to ref + 5 from 14-1500', easy.

from 777 Ops Manual:

[-200ER]
FLAP LIMIT
1 - 240K
5 - 220K
15 - 200K
20 - 190K
25 - 180K
30 - 170K
CENTER FORWARD PANEL

[pax britanica]

After 112 often interesting sometimes bizarre pages on this topic I thought what does this incident mean to me as a user of the product that aviation serves up-i.e. a passenger .
It seems reasonable that in exchange for my ticket money airlines are supposed to offer me fast, safe transportation from A to B. Implicit in that contractual arrangement, to my mind, is that they should make every possible effort to ensure the flight is safe by :-
1 Having properly maintained and equipped aircraft for the role.
2 Having properly trained flight crew capable of maintain safe operation in all reasonable circumstances.
3 Not fly to any airfield that they think is beyond the capabilities of 1 or 2.
Clearly in this case Asiana dropped the ball, they must know all about SFO and its ATC peculiarities (on one trip there I was astonished to find ATC allowed us to take off ‘the wrong way-i.e. against the landing stream. No doubt all coordinated etc but something I could never imagine happening at LHR or FRA). Asiana must have been aware that visual approaches are far from unknown at SFO and that as the ILS was inoperative and had been for some time so why didn’t they plan for it.
Leaving aside all the speculation about converting from Airbus to Boeing ( surely even a stressed tired pilot can remember what kind of plane he was flying (not like it was A321 vs an A319 was it) , tiredness. Fast speaking US ATC, A/T Modes, and a score of other rather thin excuses it comes back to the fact when assessing those three elements that concern me as passenger
1- Nothing was wrong with the aircraft other than a few tricky issues with the Auto throttle modes
2- A lot was wrong with the crews overall competence for the task- I just cannot readily accept most of the ‘excuses or reasons’. To me as a customer a crew having to fly a visual approach on a CAVOK day because the navaids are inoperative is surely a fundamental, realistic and reasonable expectation of a competent crew
3- A few quirks for the airport approach process that can distract or hurry the crew .
It looks to me like an absolute field day for US lawyers and for Asiana a massive legal and compensation bill , insurance complications and horrendous negative publicity in one of their major markets.

And personally I would not fly Asiana any time soon even if some people would suggest that made me racist.

[tilnextime]

BBK

I'm not sure that your comments regarding the ace factor reflect a typical civilian airline. Maybe on a squadron perhaps? In fact it sounds like a lot of what you are describing might be more applicable to military aviation.

Then why all the discussion of hours, hours in type, hours as a check pilot, "I can handle SFO, but a lot of pilots can't", etc.

You are either qualified to and perform to the standard or you are not. There is a task to do, there are the conditions within which the task is done, and you either do it to standards or you don't. Just as there is no such state as "almost pregnant", there is no state of "more pregnant".

If the procedure at SFO are unsafe, then the response from the pilot community would be "WE can't handle it safely". Saying "SOME can't" is Ace factor.

[nitpicker330]

TOGA ----the prudent Airline professional wouldn't (doesn't ) fly at the limit speeds for each flap setting so holding 180 kts F25 wouldn't be done.....most of us hold about 10 kts margin to allow for speed variations. Limits are limits..

Besides flap limits if you are going to hold 180 kts to 5 nm's then in Asiana's case they needed to lose 43 kts, configure final flap and then spool up. ALL IN ABOUT 2 nm and around 50 seconds. A little de stabilizing I would think and totally unnecessary. The fact the PM was new to the 777 and hand flying a visual approach made it harder to do. No excuses for the TC in the right seat though

[CONF iture]

Originally Posted by suninmyeyes

The biggest clue that things were about to go wrong was most likely the word "HOLD " at the top left of the primary flight display that indicates to those in the know that the autothrottle although engaged will not increase the thrust. Had the word at the top left been SPD (Speed) the accident would not have happened.

In this eventuality, would it be enough or necessary to deselect the second FD to get back to SPD ?

[bobcat4]

Besides flap limits if you are going to hold 180 kts to 5 nm's then in Asiana's case they needed to lose 43 kts, configure final flap and then spool up. ALL IN ABOUT 2 nm and around 50 seconds.

nitpicker330, do you say ATC put them unstable?

I also wonder why ATC want to rush approaches (and possibly make them unstable) when it can lead to a go-around which adds another aircraft to the queue…

[nitpicker330]

ATC did require 180 kts to 5 nm. It's up to the Pilots whether or not they accept this in the prevailing conditions and taking into account their own experience levels etc.