Asiana flight crash at San Francisco

[sccutler]

Originally Posted by cat1

So, you are assuming it was just an unstable approach.....being the most likely explanation because of your racial prejudices and the make up of thr crew?

That is a shameful and unfortunate accusation.

One who does not recognize the difference between differences based upon race, and differences based upon culture, might perhaps try to better educate oneself before spouting off.

[jportzer]

Why on earth would the pilot not order an immediate evacuation as soon as the plane came to rest? They crashed. Planes that have crashed tend to catch fire. So get off the plane!!!

You must not be a pilot? Evacuations can and do cause injuries; you don't evacuate until you complete an assessment and verify that evacuation is a safer course of action. It would have taken a moment or two for the flight crew to realize the amount of damage to the rear section of the fuselage - by all accounts, things were much better up front. Yes it was a very hard landing but until they have gathered at least the basics they aren't going to magically know the extent of the injuries and damage.

Perhaps the following B777 evacuation checklists will help:
http://www.aaib.gov.uk/cms_resources...%20App%20B.pdf

As you can see the Boeing version of this doesn't have the evacuate call (step 5) until after a few other steps are completed, which will make the evaucation safer for all involved.

[pfhjvb0]

Freehills said:

"Should we compare accident rates from western Europe and Asia? Anybody?"

IATA: 2006-2012 data.

North Asia (which includes Korea) has had lower accident rates than Europe

Europe 0.32 0.29 0.42 0.45 0.45 0.00 0.15
North Asia 0.00 0.88 0.00 0.00 0.34 0.00 0.00
Industry 0.65 0.75 0.81 0.71 0.61 0.37 0.20
IATA Member Airlines 0.48 0.68 0.52 0.62 0.25 0.41 0.00

Actually, Asiana is considered part of the Asia Pacific region. North Asia is essentially China, the Chinese-controlled city states such as Hong Kong, and North Korea. See IATA - North Asia

Asia Pacific IATA accident rates are notably higher than Europe:
Asia Pacific 0.67 2.76 0.58 0.86 0.80 0.25 0.48

[jportzer]

That crash would never have happened with an automated cockpit, and the move towards getting pilots out of airliners just took a huge leap. The day is coming, and like the horseless carriage opponents of yesteryear it would behoove us to accept and adapt to that fact.

I agree, I cannot understand why so many people on this thread seem to believe the problem is too much automation - to me it seems like not enough, particularly for the minimum-training regime that many people are saying is common for Asian airlines. Had the ILS been available, and an ILS-coupled autopilot been used, would the accident have occurred? No? What precision GPS approaches were available that could have been used, and why weren't they? That's just as much a question as the apparent lack of hand-flying ability.

So to me it would be perfectly reasonable for Asiana's bean-counters and risk assessors, to decide an appropriate response to this crash is NOT to increase training for hand-flying, but rather require ILS or other precision approaches for all operations. Isn't this something that could reasonably be attained in the next few years if made a priority by operators and airports? For wide-body operations at least.

Related to that, was the ILS available for 28R and could Asiana have landed there? Could an airline state to ATC that they require a precision approach, even in CAVOK conditions, if they had a policy against visual approaches? I realize there are complications as it relates to separation with parallel runways.

[bklooste]

crew used axes after failure

San Franciso Plane Crash Asiana Cabin Manager Reveals What Happened

Looks like a well handled evacuation , despite issues..

Also it looks like the company will come down on the pilots like a ton of bricks ( CEO already said the plane was fine) yet the organization trained them... I wouldnt be surprised to see some suicide.

re racism vs culture - I dont think its racism , racism is what you get when you work in an Asian country especially if you are black! A lot of the Australians here have worked in Asia , and there is more respect for some countries/ airlines eg Singapore and Cathay than others. KLM as has been mentioned had huge cultural issues which were a big talking point at the time .

[jugofpropwash]

No, I'm not a pilot - but I don't believe that waiting around patiently in a crashed plane waiting for it to catch fire is a prudent idea.

Granted - perhaps there needs to be some sort of command half way between "we'll wait here for them to bring the stairs" and "get out now, the plane is on fire" - in short, a command for a controlled, orderly exit.

Had the plane caught fire even just a minute or two faster, I suspect I wouldn't be the only one asking why the pilot didn't order an evacuation immediately.

[vieuphoria]

Automation won't save you

All these people sproking automation as the saving grace, just remember there is only so much a computer can do!

If everything was automated, what would happen to aircraft still in the air around KSFO after this crash?

Do you close an airport on CAVOK day because you don't have ILS?

How do automated systems handle flameouts/engine failures/FOD blown tires ETC....

Fair enough SOMETIMES pilots make mistakes, but I don't think we are any safer relying in automation.

It all comes down to AVIATE, NAVIAGATE, COMMUNICATE.

While this accident appears to be a result of human error, lets also remember the crews of US1549 and QF32 in which ALL PASSENGERS were saved as the result of crews doing what they are trained to do!

[ChrisJ800]

Any speculation as to what caused the post-crash fire?

hmmm, the starboard engine and ruptured fuel lines that broke off and ended almost adjacent to the starboard fuz. Pics here on this thread.

[Brian Abraham]

Just an observation, but find the parallels with Turkish Airlines flight 1951 of interest.

Co-pilot new to type
High and fast initially
Stalled with insufficient height to recover

That crash would never have happened with an automated cockpit, and the move towards getting pilots out of airliners just took a huge leap..... but we are at, or close to the threshold where the overall reliability of an autopilot exceeds that of a human, and airlines and their accountants shareholders and lawyers will react accordingly

In your dreams, you'll find yourself one of the extremely few takers.

let's see your computer do a Circling NDB to minimums in a thunderstorm with a failed AHRS

[BBK]

NTSB Press Conference

Does anyone have a link to the NTSB Press Conference on the 8th. I believe it was held at 1130 am local.

Make a nice change from reading that pilots should monitor airspeed and other things we pros never knew.

[ianwood]

Originally Posted by femanvate

That crash would never have happened with an automated cockpit, and the move towards getting pilots out of airliners just took a huge leap. The day is coming, and like the horseless carriage opponents of yesteryear it would behoove us to accept and adapt to that fact.

Originally Posted by jportzer

I agree, I cannot understand why so many people on this thread seem to believe the problem is too much automation - to me it seems like not enough, particularly for the minimum-training regime that many people are saying is common for Asian airlines. Had the ILS been available, and an ILS-coupled autopilot been used, would the accident have occurred? No? What precision GPS approaches were available that could have been used, and why weren't they? That's just as much a question as the apparent lack of hand-flying ability.

As lowly SLF, it seems the aviation industry is at an inflection point. This crash might be an argument both for and against more automation. While it improves safety, reliability and ease of commercial flying, it appears to come with a cost. It would appear that some pilots get less training, forget or don't ever fully learn the basics of flying and get away with it until something like this happens. I don't think turning on ILS at every runway is the answer.

I personally don't ever want to get on a plane with a push button pilot. It leaves a giant hole in a very important piece of cheese. You can automate the hell out of planes and it's ultimately a good thing, but when something goes wrong, the pilot and the plane need to be able to "go back to basics". If the plane or the pilot can't resort to a fundamentally "manual"or "analog" mode of flying, we will see this again and again no matter how good the technology.

[MountainBear]

I agree, I cannot understand why so many people on this thread seem to believe the problem is too much automation

Because the current problem is too much automation just like in the long run the problem is too little automation. While the completely automated plane will happen one day the system is by no means ready for that to happen this decade, let alone this year or this month. There is a tremendous amount of work that needs to be accomplished first from regulations, to procedures, to testing, etc etc. Moreover, such planes will be thoroughly tested in GA or cargo operations long before they ever make it to commercial operations.

In the meantime, over-reliance on automation is a real problem right now and there is widespread (albeit not total) agreement on that point. There is much less agreement, however, as to what is to be done about such automation dependency.

If the plane or the pilot can't resort to a fundamentally "manual"or "analog" mode of flying, we will see this again and again no matter how good the technology.

This is a commonly expressed fallacy. It is erroneous because it evaluates the success of automation in light of an regulatory framework and an operating environment that was not designed for fully automated flight. The entire system has to change to accommodate fully automated flight and when it does this concern will disappear.

[clayne]

This is so typical to see some people respond with "well, I guess this means we need more automation!" which is an everythings-a-nail solution. Fly the airplane, go around. It seems neither were done and it doesn't appear like there's anything broken to fix here other than the mindsets, training, and magenta lines. Why cater to the lowest common denominator as an excuse for lack of quality in mindset and approach?

[TRF4EVR]

Some peckerwood's gotta take the beast up, and some peckerwood's gotta...land the son-of-a-bitch. And that peckerwood's called a...'pilot'

...You're mighty right.

For the remedials: The notion that any "system" will replace the human mind in the short-term future is a pure fantasy. The virtue of the human mind is that can process information in a parallel fashion...computers are and will likely always be a linear processor. Let the computers loose on GA aircraft first, see how often "The System" fails, then come back and impress us with how "pilots" have become obsolete. This accident strikes me as the perfect example of how taking human beings OUT of the equation leads to disasater. You can't program Jugdment, it's a human trait. And when you head in that direction, well...

[neville_nobody]

To the people who are spruiking automation thats is most certainly a fool errand.

QF 32/BA 38/US1549 are incidents where automation would not have done a damn thing, yet humans managed to walk some 1000 people collectively from pending death.

This accident assuming it is not a a medical issue at a critical moment will probably end up being some sort of cultural problem ultimately. Someone didnt want to say or do something at the right time to save face.

If you cannot land any plane in those sort of conditions you shouldnt be flying.

[benh57]

So many excuses being thrown around, but the bottom line here is that if a healthy airliner on a sunny day, with 11k of runway crashes, something needs to be done. A quarter billion dollars and 300 lives were at stake.

Really? Well, that's debatable. From Robert Reich today:

"transportation analysts at MIT put the odds of death from air travel at two-tenths of 1 percent per 1 million departures, which they say makes flying safer than any other form of transportation, including escalators."

I say we tackle escalators first. Then cars. Then trains.

[jportzer]

QF 32/BA 38/US1549 are incidents where automation would not have done a damn thing, yet humans managed to walk some 1000 people collectively from pending death.

So wait, just because all accidents can't be avoided with increased automation, means it's a fool's errand? Do you skip wearing a seatbelt just because seatbelts can't protect you against death in all accidents? No one is arguing that automation can protect against ALL failures, we're just saying that there are SOME failures not yet protected against, that reasonably could be.

I agree with previous comments that the pilot of the future may be seen as more of a "systems administrator" - system administrators still need to take action in abnormal situations, but it will come to the point where hand flying skills are not the most critical aspect of this.

I fully agree that we are decades away from full automation. But accidents like this one will form part of the case for increased automation, not against it, in the long run.

[Capt Kremin]

The 777 is already heavily automated. The automation exists to reduce the workload in stressful situations.

Apart from fatigue issues and the need to correct a vertical offset in the profile, this sort of approach in the conditions that prevailed at the time, should not be especially taxing to any professional pilot.

All other considerations aside, I find this accident to be inexplicable. Like the AF447 accident, the Lion Air and the Turkish airlines accidents, there seems to something very insidious creeping into the world aviation scene.

At least the pilots survived this one, so they can be asked what happened. I am very intrigued as to what they may say.

[Cows getting bigger]

I think I said something about automation earlier. It appears to me that many problems occur in a 'semi-automated' mode whereby the crew loose SA regarding the division of responsibility between the pilot and the computer - "who's doing what?"

[TRF4EVR]

The logical fallacy here is that because there has been increased automation in cockpits, and accident statistics have dropped, the accident statistics must have dropped due to the increased automation. That's farsical. Correlation does not imply Causation, as even the most window-licking of students of statistics will tell you.

There's no reason that pilots flying trans-continental aircraft with hundreds of people on board shouldn't have the best of ALL possible worlds. Automation. Training. Experience. An "A" in Automation or Training doesn't render a "D" in Experience irrelevant. The automation is there to help you. The training is there to prepare you. The experience is there so that you have the basics down enough to pay attention to the other two. Why are we so willing to accept that any of us can do without any one of the three? That strikes me as the real hubris.