PDA

View Full Version : FAA Puts American Airlines Under Close Scrutiny


md-100
5th Jan 2010, 10:30
Sun, 03 Jan '10

Three Incidents In December Leads To Increased Oversight

Aero-News Network: The Aviation and Aerospace World's Daily/Real-Time News and Information Service (http://www.aero-news.net/index.cfm?ContentBlockID=673f33d1-0426-4f5b-938e-6d4c8714448c&)

After having three landing incidents in quick succession, American Airlines (AAL) is now under investigation by the FAA. Pilot training, fatigue, and weather-related landing procedures are all likely to receive an in-depth review in addition to increased FAA inspections.

"In situations where there may be several incidents involving a single carrier over a short period of time, FAA inspectors increase their oversight, which we're doing now, and conduct a review of those events to determine whether they might be indicative of a larger issue," FAA spokesman Lynn Lunsford said in a statement released Friday.

AAL's string of incidents started when an MD-80 landed erratically at CLT on Dec 13, leaving the runway and then striking the right wingtip. As ANN reported, there were no injuries though pilot fatigue was being investigated.

On December 22, AAL flight 331 overran the runway at KIN in Jamaica during heavy rains. The fuselage of the B737 broke into three pieces with reports that many of the 148 passengers were injured.

Another AAL MD-80 struck a wingtip while landing at AUS on Dec 24. No injuries were reported.

AAL spokesman Tim Wagner told The Wall Street Journal that the three incidents will be investigated separately. "We take each event as an individual event" he said.

As for the result of the FAA's investigation, Lunsford noted that "the FAA communicates its findings to the air carrier and assists in the development of the appropriate corrective action."

golfyankeesierra
5th Jan 2010, 11:08
Well, they were already under scrutiny by 411A; wonder which is worse...

411A
5th Jan 2010, 11:38
It's about time the FAA looked closely at AA flight ops.
Lets look at the latest accident at KIN.
Landed very long, wet runway (very recent TS), tailwind...and went smartly off the end

AA was very lucky folks did not die in this very serious accident.
AA has the largest hull loss record of any US trunk carrier, however...if you look back at their accident/incident record, each end every one of these incidents/accidents was directly rooted in pilot inability, or very poor thought process.
Couple these incidents/accidents at AA together with the massive fine the FAA imposed on AA in the recent past for maintenance violations, leads me to believe that AA needs continual detailed oversight until these problems are corrected.

aa73
5th Jan 2010, 12:24
Every major US airline has had a rash of accidents attributed to pilot error at some point of time. AA's just happens to be these last 10 years. Unfortunate, yes: but it will be corrected. I remember in the early 90s, USAir had the highest hull loss record of all US airlines. It was eventually corrected as well. AA will be no different.

sludge
5th Jan 2010, 12:30
This is the FAA justifying its existence. High profile, lots of ink, see how busy and concerned we are? Send more tax dollars. :rolleyes:

WindSheer
5th Jan 2010, 12:43
One would guess this is occurring due to commercial pressures??

PPL pilots in the UK face excess diversionary charges, which can sometimes lead to fatal 'press on' decisions being made.

I fear this is very much a similar case now in the commercial sector, with Captains facing an extreme grilling down if they have diverted during 50/50 conditions.
Things need to change industry wide....not just within AA!!

411A
5th Jan 2010, 15:14
Changes are a comin' at major airlines, you can be sure, in the USA.
One recent example, Southwest at RNO.
The airplane diverted to OAK due to a reported brief heavy snowfall and high winds on approach...and several pax were wondering why.
RNO is (or can be) a very challanging airport in the wintertime, and I suspect that Southwest learned their lesson well at Midway, when they slid off the end into a passing auto, and killed a child.
The FAA is clamping down, and I hope it continues.

Ocampo
5th Jan 2010, 15:55
I fear this is very much a similar case now in the commercial sector, with Captains facing an extreme grilling down if they have diverted during 50/50 conditions.

It's not just now...remember AA at Little Rock some years ago? In my opinion, the pressure is always there; we have to learn to fly with it. But then again, some so-called airlines do not "encourage" pilots to make the correct decision (of deviating) because you might get a call to visit the Chief of Pilots Office...

bearfoil
5th Jan 2010, 16:07
Then again, Mesa is having trouble with take-offs.

protectthehornet
5th Jan 2010, 20:38
I don't always agree with 411A...but I recognize that he is an experienced pilot.

American has rarely impressed me with piloting skills. Some exceptions might be pilots acquired through mergers like Air cal or TWA or such. American has claimed that Lindbergh was one of their own...through a very big stretch of the imagination.

I know at least one pilot hired At American who lied in his log book...oh well.

I've seen something that can only be called over confidence...thinking that something can be ''salvaged'' but the skills are rusty or lacking.

With new planes, higher automation, it seems to me that hand flying skills are gone. straight down a runway, pull up...autopilot on...hand fly about 2 mins.

automation to the outer marker and hand fly to the runway...2 or 3 minutes.

So, on a 5 hour flight, you get 5 minutes of real flying. Maybe 20 hand flying hours a year for a 1000 hour year. Think about that.

Dragging wing tips, overshooting a runway...this all sounds like an old pro who has lost his touch on the stick.

There is a difference between these AA crashes and the USAIR stuff in the 90's.

All of the AA stuff is happening in the same part of flying/phase of flight...landing.

All of the USAIR stuff was pretty unique accidents...aborted takeoff after V1 at LGA...rudder trim out of wack or is it whack?

cleared to land at LAX with another plane on the runway at night...obvious ATC error.

Microburst/windshear at CLT ( I was there watching the smoke come up)...atc involved with that one too, though the pilots were flaky on this too.

And the rudder deal at PIT...'nuff said on that one.

I hadn't heard about southwest diverting from Reno to Oakland...I know both places like the back of my hand...odd stuff.

aa73
6th Jan 2010, 00:10
Protectthehornet... all airlines go through rashes of minor stuff like wingtip scrapes, over-rotations/over-flares, etc. The determining factors are whether they make the news or not. I have plenty of friends at all airlines, and trust me when I say that AA is NOT unique to these incidents. If I showed you all the stuff my SWA buds send me, you'd be astonished. They never make the news, that's why you don't hear about it.

We could go back and forth on pilot error stuff. Truth is, all airlines go through it. DAL had two very high profile pilot error accidents before they changed. So did UAL... So did CAL. My point is that, while not defending AA's recent mishaps, they all happen to all airlines. The AA pilot group is as professional as any other out there, including the training. It's just the luck of the draw, which is an unfortunate fact given the thousands of flights flown every day with no incidents.

protectthehornet
6th Jan 2010, 00:31
aa73

its fine...I understand...but remember please that the american boys aren't MORE professional than anyone else...as, yes...more NO.


and southwest has gotten away with murder...better PR firms!

backseatjock
6th Jan 2010, 08:08
Nicely said WindSheer and agree with comments in full.

411A
6th Jan 2010, 08:37
Alaska MD80, jackscrew

This one was directly a pilot problem.
Why, you may ask, when the stab jackscrew failed?
It completely failed after the crew attempted many maneuvers, both flaps up and flaps down, to check handling.

Instead of....diverting to LAX or LGB when the problem was first noticed, with minimal configuration changes.
This screwing around with the airplane instead of diverting in quick time defied all common sense logic, thereby contributing directly to the accident.

TIMA9X
6th Jan 2010, 13:10
As a daily reader of PPRuNe for years now I felt it necessary to comment regarding this thread in defense of AA.

Agree with the following
by aa73: Every major US airline has had a rash of accidents attributed to pilot error at some point of time. AA's just happens to be these last 10 years. Unfortunate, yes: but it will be corrected. I remember in the early 90s, USAir had the highest hull loss record of all US airlines.In America the 5 major airlines are huge compared to many other parts of the world, it's economy of scale that come into play here, compare the number of rotations in the USA and here lies the answer. AA have had a bad run of late but the AA Kingston, Jamaica 737 off runway crash is still not conclusive, and to be fair to all pilots on this forum, it is my view we should wait for the official reports.

The Dallas Morning News also reported that the runway at Norman Manley International Airport in Kingston, Jamaica was not grooved like many runways in the US, making it easier for water pool on the runway and increase the likelihood of hydroplaning. The main runway (12/30) is 8900 feet (2713 meters) long, with an instrument landing system available (ILS) only on runway 12. Because of the weather conditions at the time of the landing (rainstorm), and the lack of an ILS option on runway 30, it is likely that the crew had no choice but to land using the ILS equipped runway which meant either landing with a tail wind or landing at an alternate airport.It is possible that commercial pressures may apply in this case.

I agree with WindSheer
I fear this is very much a similar case now in the commercial sector, with Captains facing an extreme grilling down if they have diverted during 50/50 conditions.
Things need to change industry wide....not just within AA!!Having said that, 411A also has some very valid points, and I always respect his point of view, re the Alaska MD80, jackscrew disaster. But I must point out that it was not long ago that the FAA had also focused heavily on Southwest and United maintenance procedures.

I am not American but I do respect the American pilots and their crews, they may do things slightly differently to say Europe but at the end of the day all the American airlines manage to get to through most 24 hour periods without incident.

fdr
7th Jan 2010, 02:26
This one was directly a pilot problem.
Why, you may ask, when the stab jackscrew failed?
It completely failed after the crew attempted many maneuvers, both flaps up and flaps down, to check handling. -411A circa dark ages

So, since we warm fuzzied the aerospace scene with CRM etc, we set up the process for activity beyond the checklist. While there are times tht this is appropriate, there are as many or more when that is not the case.

I do not argue that 411's doesn't have some point on this one, I just suspect it is a case of "shootin' the messenger". Is it not of interest that the system set this program up, through the auspices of AC120-51 (orig, A thru E), JAA-OPS TGL No5 & 6 etc. The line between correct action and correct inaction is blurred, as the basic premise of CRM is to give better outcomes, which often is misrepresented as being the perfect outcome. As a lowly 22,000 pilot and air safety investigator off and on for the last 30 years, and a past CRM facilitator, I find myself rather reserved in the benefits of teaching perfect outcomes in a dynamic environment. (well, when I taught...). As a pilot, I think that an adeqate solution that removes my airborne problem to a ground management problem is usually a good thing, satisficing my requirements of minimal disruption to my operation, minimum interference by external intrusions and hopefully least risk to me, which is what I get paid for by the passenger.

I have watched too many relatively competent crews loose SA while running evermore complicted scenarios of their own making, (both in simulations, and in investigations after whacky outcomes occur in 12" to 1' scale, HiDef etc), to be overly enamoured with our current philosophy in opertions and training.

Operationally, the driver (which is professionally where we have allowed ouselves to be re-categoried as... thanks AI...) is torn between the basic tasks as written, and the desires of the passenger (percieved) and the companies overt and subjective demands. Out of all that, he/she is damned if he does (AS261), and damned if he doesn't (SR111).

It is far easier to be perfect after reading the crash comics.

I recall a small vignette from a prior life; where in beginning of the '80's, flying with tin hat on, I had a jammed elevator system in flight. Oddly the cause was immediately obvious and the emergency checklist was considered to be catastrophically detrimental to the actual problem (shifting hyd boost on > off)... jury rigged a solution, and landed. The subsequent investigation found I had acted correctly in light of my diagnosis being correct, "but the NNCL (NATOPS) had not been actioned as written". A week later the writer of the comments (an experimental TP...) had a similar jam of his ailerons, and followed the same actions I had for the same reasons. No comments made on his report.... Easy to have the solution on the ground, and after the reports published.

Crew disregarding an approved procedure is a matter of some gravity, yet, notwithstanding the vignette above, external agencies such as Maintenance and Ops Control suggesting the crew don't action a fire checklist, MEL etc, is hardly sound risk management, but crosses the line to being interference in the operation. If the PIC is compelled by his knowledge to act in some way, then he will take responsibility for that at all times without question. The dispatcher or maintainer don't. And by law, nor does the FAA ATC controller. After a crew have actioned the NNCL and the situation has not improved will always call for careful and thoughtful consideration of the relative risks involved, classic decision making under conditions of uncertainty.

As far as Alaskan goes, I think the greatest learning point missed was the fabrication of records, and the absolutely disgraceful treatment of the insider who blew the whistle to the FAA and was pilloried for his efforts at attempting to preserve life, while of course little substantive action was taken by the regulator to do same. :mad:. As is often the case, the company fails to see that their behavior when it goes feral due to commercial factors makes them the enemy of both the pasenger and the company itself, whereas the whistle blower is perceived to be the threat.

All depends on your point of view.

411A, I think the issue of operation risk management is far more complex than it appears on the surface, and that simplistic observations are often just that, simplistic.

happy new year.:)

Warm Springs

bubbers44
28th Jun 2011, 19:57
So the NTSB didn't understand he was doing a cat 2 approach and had to manually land?????? If it was an MD80, I have lost autopilot and autothrottle below 300 ft on a cat 1 and recovered and landed with no go around with the same airline. MD80's are full of surprises. Hopefully we don't replace them with the A320 as rumored. Then we have worse problems. At least the MD80 will do exactly what you want, even if it's wrong.

beardy
28th Jun 2011, 21:42
bubbers44,

1st line of the report:

Both pilots indicated that they had set up the navigation systems and briefed to fly an ILS Category III instrument approach to an autoland to runway 36C

Last line of the report:

Factors contributing to the incident were: The Captain's decision to execute a Category III autoland approach on a runway without rollout guidance contrary to company Flight Manual guidance, the Captain's decision to deviate from the Category III approach and continue it manually contrary to the company Operating Manual, and the First Officer's failure to call for a go-around when the approach became unstable.

So I guess that NTSB were correct in assuming he wasn't executing a Cat2 approach.

Hopefully we don't replace them with the A320 as rumored. Then we have worse problems.

What an odd statement. I assume that the second sentence should include the conditional 'would.' I am not sure what condition of the A320 itself would lead to worse problems than the operational indiscipline highlighted in this report.

bubbers44
28th Jun 2011, 23:48
I'm still trying to find the accident report because it seems they could have at 1400 rvr done a cat 2 approach. Maybe briefing a cat 3 approach by AAL procedures would require a rebrief for cat 2 but the FAA doesn't require it. 1200 rvr is the normal cat 2 minimums. I think I could do a 10 second rebrief on changing to a cat 2 approach and I don't think any pilot would initiate a cat 3 with no required roll out visibility.

beardy
29th Jun 2011, 07:27
Notwithstanding a sloppy report, briefing and executing a Cat 3 approach to a non Cat3 runway (no auto rollout) and not briefing and not executing a go-around when patently unstable shows a less than rigorous attitude to procedures and safety.

Both NTSB and the crew refer to outer marker, the Post Incident ILS readings make clear that there is no outer marker. NTSB refer to AP disconnect at less than 300', the Captain says 'at' 300.' NTSB refer to the aircraft being 'fully configured' leaving the reader to assume what that configuration is.

IGh
4th Jul 2011, 00:42
observation just above, Re' the landing-ARC at CLT 13Dec09 :"... shoddy... slow investigation that the NTSB should be ashamed of ..."
The NTSB's main offense springs from their attempt to formulate a "probable cause" statement: The Ops Group Report (52-pages, see docket) includes most of the omitted information. The writer of the Ops Group Rpt provided a better quality document, and provided information that contradicted the NTSB's "final". That Ops Group Report was completed in March 2010 [fifteen months prior to the NTSB's release of their "final" P.C.].

Advice: Pilots should disregard the NTSB's final -- instead get the Ops Group Report, that clarifies most of the uncertainties of the Board's P.C. "final".

Advice for the Board: Lacking resources for careful consideration, you shouldn't even try to formulate a P.C. [Most of the time your official P.C. leads your reader off-course; with protecting the USA's airline, regulator, and manufacturer as NTSB's highest priority.] Another err in the NTSB's "full narrative" (web), was your omission of any FDR analysis; there are lessons for pilots in the FDR-data after the human disconnected the Autoflight: Loc-dev goes from bad to worse with the human's "visual" over-control while in low-visibility near the runway (look at "bank angle" plot during last three seconds of flight). Docket offers only the "preliminary" time-history plots of selected data, docket contained NO FDR ANALYSIS. Each of three FDR-plots mistakes the MD80's "Air-Ground Sensor" (NLG strut compression) as instead "weight-on-wheels": the Abnormal Rwy Contact occurs about 4-seconds EARLIER than the time-line labeled "B" (right end of green shaded interval).

See "docket", Ops Gp Rpt Operations - Specialist Report to the IIC (http://dms.ntsb.gov/pubdms/search/document.cfm?docID=327814&docketID=48678&mkey=75163)
-- pg 26, Figure 7: Right inboard trailing edge flap damage.
-- pg 41, Figure 13 Left edge of 36C approach end; tire ruts [photo of ruts in grass outboard of the concrete but INBOARD of the Rwy Edge Lights].
-- pg 42, Figure 14 Left main landing gear covered in dirt


Aside from the ARC [Abnormal Rwy Contact], the Ops Grp Rpt suggest this case as "USOS" taxonomy:UNDERSHOOT/OVERSHOOT (USOS)
A touchdown off the runway surface.
Usage Notes:
• An undershoot/overshoot of the runway occurs in close proximity to the runway and also includes offside touchdowns and any occurrence where the landing gear touches off the runway surface.
• Off-airport ... excluded ...
• ... landing phase.
= = = = = = = = ==

"… During post incident interviews, both pilots stated that they selected flaps 40 as a final flap setting." ["docket", Ops Gp Rpt, page 47]

"… chart depicts the bank angle limit versus the pitch limit (in degrees) …" ["docket", Ops Gp Rpt, page 48.]


"… dispatcher also wrote 'CREW WAS CAT 3 ACFT [aircraft] WAS CAT 3 … and further wrote “RECAP: CAPT, ACFT AND AIRPORT WERE ALL CAT 3.' …" ["docket", Ops Gp Rpt, page 51.]

= == = == = = = = =

The NTSB's "final" P.C. repeatedly states that those mishap-pilots were flying a Cat-3 (low vis approach). However, the Captain's own write-up in the Mx Logbook contradicts that investigator-assertion:"… logbook page showed the following December 13, 2009 second line entry (page 10532110) for flight 1402:

'On Cat III [Captain had striked-through the "III" then entered] II (2) Wingtip touch on landing ... [Captain’s Initials]'"
["docket", Ops Gp Rpt, page 25, Figure 5: image Captain' entry into the Mx Logbook].
Since the Rwy 36C-Loc did not provide rollout guidance (note #4 on approach plate), that ILS-Loc was NOT authorized for AA's MD80 Cat-3 Autoland: On final at 300' the pilot disconnected the autopilot, and did a manual landing. All the NTSB's paragraphs discussing Cat-3 Approaches lead the reader off-course [since both Autoland and Cat-3 were prohibited and the pilot did neither].

In several respects this ARC-, USOS- mishap [13Dec09 at CLT] is similar to the AA low-vis USOS- event at DEN/ 21Nov04 DEN05IA027 (http://www.ntsb.gov/aviationquery/brief.aspx?ev_id=20041122X01849&key=1)

sleeper
4th Jul 2011, 07:23
@Beardy
executing a Cat 3 approach to a non Cat3 runway (no auto rollout)


Not all Cat III require auto roll out, ie cat IIIa

Abbey Road
4th Jul 2011, 09:14
Since the Rwy 36C-Loc did not provide rollout guidance (note #4 on approach plate), that ILS-Loc could NOT be used for a Cat-3 AutolandI have been thinking the same as you, sleeper, - Cat 3 Autoland and Cat 3 Autoland with rollout are separate things. It is perfectly possible to do a Cat 3 Autoland (or Cat 2 Autoland) and disconnect the autopilot after nosewheel touchdown! :rolleyes:

IGh
4th Jul 2011, 15:39
Observation regarding Cat-3 ILS (low visibility approaches), from just above:"I have been thinking ... Cat 3 Autoland and Cat 3 Autoland with rollout are separate things. It is perfectly possible to do a Cat 3 Autoland (or Cat 2 Autoland) and disconnect the autopilot after nosewheel touchdown!"
The Ops Group Rpt cites the various conflicting restrictions [see docket, Ops Gpt Rpt, pgs 37 & 38] facing those mishap-pilots, when considering a low-visibility ILS approach, with a fail-passive autoland system, to a rwy without rollout guidance. The conflicting rules are mind boggling.

This statement sort of gets to the PROBLEM: neither the mishap-pilots, nor the investigators, seemed capable of sorting-out the various contradictions, cobbled together, to get CERTIFICATION of an MD80-style "fail passive" autoland system.

This CAT-3 evolution since the late 1970's, regarding very low visibility landings, has now is become foggy. The MD80 was certificated during 1980. By 1980, the L1011 had almost a decade of low-vis autoland history. At Boeing, in 1980, one customer had order ten B747SPs, but they wanted full Cat-3 autoland on that new "model" SP [new because of the Autoland cert'].

That test program (B747SP w/Autoland) proved to be a show-stopper: That test-version red&white B747SP stayed-on at the Test Center for about 10 months extra, with the customer getting late-fee payments, from the manufacturer, for every day-late.

Meanwhile, down south at Long Beach, the MD80 Cert' seemed to get some sort of odd certification for a Cat-3 ILS landing system, employing a human's Decision Height (rather than the usual Cat-3 Alert Height), with only a Fail-Passive Autoland System: Something goofy was happening!!! This stretching of the Cat-3 philosophy, just to get that "Fail Passive" product certified, was leading us to the sort of COMPLEXITY that neither pilots nor investigators would comprehend. And this "fail-passive" Cat-3 Certification WEAKNESS (complexity), is NOT mentioned anywhere in the NTSB's P.C.

Graybeard
4th Jul 2011, 16:36
Lack of a Fail Active rudder kept the MD-80 and 737 from even being considered for Cat IIIb with its rollout steering requirement. They are only Cat IIIa qualified. The 737NG got a Fail Active rudder in 2000 or later.

GB

beardy
5th Jul 2011, 06:50
non-Cat 3 runway

Sorry my sloppy writing. According to the NTSB report the company were not allowed to execute a Cat 3 approach to 36c by their own company rules. It was that that I was using.