AA Crash Jamaica

[safetypee]

Similarities with AA accidents and many other overruns – plan continuation bias (press-on-it is).
We need to ensure that pilots have appropriate knowledge about the risks when in landing in poor conditions and how to use that knowledge – 'know what' and 'know how' – risk management, how and when to think about these issues. Risk Perception and Risk Management in Aviation

Other refs:
UK AIC 14/06.
UK AIC 86/07.
http://www.skybrary.aero/bookshelf/books/874.ppt

Safety aspects of aircraft performance on wet and contaminated runways.

Safety aspects of tailwind operations.

For some ‘How To’ aspects – Judgement, etc, see ‘library’ section - presentations on Critical Thinking, Situation Awareness, and Decision Making (free registration)

[sAx_R54]

p51guy

I would have thought there to be a degree of mutual exclusivity between someone selling drugs and/or his sister (to you perchance??), and aircraft handling that appears to have mis-judged the MAP on its FAF to the extent that 50% of the available landing area at its disposal had dis-appeared!! Maybe we can better understand the cause of the accident this way.

sAx

[captjns]

PJ2 queeried...

I know go-arounds are taught at "every" airline - it is a required manouevre just like the rejected t/o, engine fail at V1, etc. But just to make sure we're talking about the same manouevre, the assumption/suggestion I am making here is, this isn't a go-around from minimums, this is a go-around from just above the runway, at touchdown, or on the runway, all after the thrust levers have been closed for landing but reverse has not been selected. That to me is a "balked landing", just so we're on the same page.

I believe this is what 411A and one or two others are talking about and it is what I am focussing on.

That is the manouevre I don't think many airlines teach. As we are aware, the manoeuvre is high-risk and requires a high degree of skillful, coordinated actions.

You are correct... every airline I have been associated with trained go-around from balked landing just above the runway. We have even covered engine inoperative balked landing procedures too.

When conducting base training in the B-727 I would have the student fly just above the runway without touching down so they could see what the sight set up would be. Then a go-around would be executed. Of course all this was briefed before the base training commenced. With that being said, by default, I trained balked landings both in the simulator and in the aircraft.

It's a good idea to provide training in the aircraft too. However, heads of flight training at a number of airlines don't have the stomach for demonstrating such manuevers in the aircraft.

[PEI_3721]

Recent focus on go-arounds overlooks two important points:-

• Although a crew may have been trained how to fly a GA; they still require judgement and motivation to choose to fly the GA, i.e knowing when to change the approach to a GA – when to change the plan.

• The other point is to have a plan; one which ideally considers the risks during the landing to avoid a change, choose the safer option; or have a plan which clearly recognises the need for a GA because of the potential risks of the chosen option, e.g. choice of tailwind approach would require close attention to approach and touchdown accuracy, which if not achieved, a GA is the safer option.

“A Landing is an approach without a Go Around”

[ELAC]

ELAC asks:

Quote:
There's an unresolved question here: How does an aircraft come to touch down 4,000' down a runway at Vref speed?

Flying technique, continuing the landing phase beyond the touchdown zone without initiating a go-around as required in the company's FCOM.

captjns,

To be clear, the question is not how did the aircraft get to the place that it did, but how did it get there with the same speed it should have had 10 seconds earlier at the presumed flare point on the glide path?

ELAC

[Flight Safety]

I remembered something from my flight simulator days many years ago now. I have maybe 300 hours or so in full sims on several types, again from many years ago. For fun we used to practice night scene low ceiling ILS approaches, to break out of cloud at very low altitudes near the threshold. I remember early on my instinct was to pull back on the yoke as soon as I broke out, a reaction to seeing the runway lights and the ground suddenly coming up at me. I had to train myself not to reach and to recognize that if I'd flown the ILS correctly and was stabilized in the approach, I shouldn't change anything immediately after breaking out of the cloud.

I know all professional pilots do the same thing that I learned to do all those years ago, when breaking out under low cloud. However this pilot chose to fly an ILS approach downwind in cloud, and broke out somewhere between 700 and 1000 feet. Even with the tailwind, his approach should have been stablized long before he broke out, as following the glide slope should have adjusted his decent rate to compensate for the tailwind. Is it possible he just had a bad day and suddenly pulled back on the yoke when he broke out, resulting in an altitude change and the extra energy in the landing?

For a professional pilot probably not, but just a thought.

[protectthehornet]

/The Silent Pressures

For too long we have overlooked the silent pressures of being on time...of saving fuel and the like

And don't pretend it doesn't influence you. The on time ratings come out every year...without allowances for go arounds during bad weather.

so he landed 4000' down the runway...bad in my book too. but he probably had gotten away with it before somewhere else and saved a few minutes and a few pounds of fuel.

A go around should get the crew a bonus, not anything else.


to the FEDS...drop the on time reports...unless you compensate for wx, and other safety of flight issues.

To the airlines...train for go arounds and what is a good reason for a go around. In ppl (that's private pilot training) I use to train...if no landing in the first third...GO AROUND.

That should be ingrained into everyone's flying skills.

Every winter at our airline, we talk about the criticality of the approach and landing especially on contaminated/wet runways. And not wasting runway!

[PJ2]

captjns;

Yes, and to discuss it a bit for others...
If nothing else, doing it in the airplane certainly gives one confidence in the machine and familiarity with one's own capabilities and enhanced skills.

That said, history has lessons of its own here in terms of training accidents and I don't suppose too many would blame the heads of flight training for not having the stomach for them - to save costs, time is intentionally compressed because each landing is a touch-and-go, essentially a go-around by any other name. After the t/d, the instructor takes over momentarily, to ensure the flaps are re-set, the engines spooled and the student is "cleared to go" again. Such work is worth its weight in gold however, even with the occasional engine throttled back at rotation to simulate failure. After one has the experience, one realizes just how far the simulator is from the airplane, even today. They're good, but inertia, mass, g, and response to same simply cannot be duplicated.

PEI_3721;

Even the go-around is not automatically consdered the safest manoeuvre even though it may be the best/safest option "in the moment". There are a number of accidents which have occurred on the go-around due to various reasons including disorientation, mode-confusion and lack of familiarity and/or skill/training as well as an almost universal time-compression during the manoeuvre. Mental preparation and mental state are, as has been observed here, key elements which avoid surprise, distress and a possible loss of control.

At Kingston, whether a go-around from the runway was available, doable, wise or whether the accident began to occur after touchdown we do not know yet even though many will have an idea or two.

[ELAC]

That is why I included the possibility that, similar to AA1420, the aircraft got high in the very last stages of the approach during the "inside to outside" perceptual transition, in darkness and heavy rain.

Nor do we know the power settings across the threshold, either but I wonder - will the N1s be low to idle indicating a higher than normal threshold crossing height, or will we see N1s reducing and then, if the autothrust was still engaged, increasing again in response to autothrust commands to maintain the approach target airspeed; - we do not have that information and cannot discount either notion.

PJ2,

These are both very possible explanations. Most of us have seen both occur one time or another, though fortunately not with the same outcome. I've certainly witnessed approaches, especially in strong crosswinds, where the PF got so absorbed in handling one aspect of control that another, such as thrust managment, was totally forgotten in the moment even with, in the case of the 'bus, repetitive autocalls of "Retard".

ELAC

[punkalouver]

All this talk about the AA 1420 accident in Little Rock. Why don't we just admit that the decision to continue that approach was one of the stupidest ever seen.

At one point on the approach with the flightcrew knowing that a thunderstorm awas at the airport, the controller gave a second windshear alert and winds of 32 knots gusting to 45 knots followed by an RVR of 1600 feet in heavy rain. What is the likelyhood of a successful landing, especially when the wind is 50° off the runway?

[PJ2]

punkalouver;

All this talk about the AA 1420 accident in Little Rock. Why don't we just admit that the decision to continue that approach was one of the stupidest ever seen.

I think that categorizing a decision merely as "stupid" has a material outcome beyond criticizing a pilot or a crew. It does not produce a resolvable outcome - we are all capable of stupid decisions and the idea isn't elimination, it is awareness. The notion of "stupid" will always have a context and therefore a stupid decision, but for small turns of circumstance usually beyond our personal control, may have been brilliant. Where something goes wrong and ends in an accident or something goes wrong and we get away with it, neither category, (stupid, brilliant), nor the outcome, provides a basis for analysis and, where indicated, for change.

The Report on AA1420 states that the change for AA was an arming of the spoilers by the captain, with confirmation of arming of the spoilers by the F/O among other changes. The report is severely critical of the decision to continue, indicating that each factor in and of itself may not have warranted a go-around but collectively they did. Yes, it seems a simple lesson with hindsight bias but it made sense for many years. I know another airline which, after a major fatal, changed its spoiler arming procedures to ensure arming by one crew member and confirmation by the other. That did not prevent another accident in which loss of spoiler deployment was implicated in another fatal over-run accident on takeoff. The TAM accident comes also to mind although there are WOW and logic factors (#2 TLA) involved.

[411A]

Landing anywhere at night in heavy rain and the wipers going flat out is not nice though, the runway lights can blur and mushroom and depth perception becomes more difficult.

The subject aircraft was equipped with a HUD (as I recall), and if properly used, should greatly aid in the visibility department.
Comments from those that have used a HUD in the 737 might be interesting, in this regard.

[stilton]

I wonder how much reverse was used.


Idle reverse to 'save fuel' is a common practice now. So common that many Pilots will not use max reverse even when needed.

[captjns]

I wonder how much reverse was used.


Idle reverse to 'save fuel' is a common practice now. So common that many Pilots will not use max reverse even when needed.

The thrust reversers would not have saved the day.

Thrust reverses on the 737 serve two functions... makes a lot of noise and alerts the captain to wake up so he/she can man the tiller to steer the jet on the ground, and two puts money into the coffers of the oil company through increased fuel consumption.

[411A]

Thrust reverses on the 737 serve two functions... makes a lot of noise and alerts the captain to wake up so he/she can man the tiller to steer the jet on the ground, and two puts money into the coffers of the oil company through increased fuel consumption.

Hmmm, sounds like the thrust reverse on the old straight-pipe 707's...
Who wouda thought...

[Kirks gusset]

FROM: THE BOEING COMPANY

SUBJECT: 737-800 Landing Overrun Accident at Kingston, Jamaica – 22 December 2009


Several points from the press release are paraphrased below:

•There were 154 persons on board the aircraft with no fatalities but numerous injuries reported.

•An ILS approach to Runway 12 at KIN was made at 10:23pm local time (night lighting conditions) in heavy rain and with a tail wind of about 14 knots.

•According to the FDR, the aircraft was traveling at the Vref (landing) airspeed of 148 knots, with a groundspeed of 162 knots, i.e. with a tailwind component of 14 knots, when the wheels made initial contact at about 4,000 feet down the 8,900-foot runway.

•Brakes were applied initially by the Autobrake system to a level of Autobrakes 3, then the crew subsequently applied maximum manual braking.

• The FDR indicates that the aircraft departed the end of the runway at a groundspeed of 63 knots.

• The runway at KIN is ungrooved asphalt. An evaluation of the runway surface conditions at the time of landing is in progress in order to assess how runway friction effected the airplane’s braking forces. To this point in the investigation, no mechanical problems have been found with any aspect of the aircraft.

• The JCAA continues its investigation of this event, and will provide additional updates as progress is made.


Boeing has no recommended operator action at this time. If the investigation shows any specific actions are recommended or required, operators will be notified.


Chief Engineer - Air Safety Investigation
The Boeing Company

KG says: Deep landing, tailwind, floating.. was never going to be good!

[Doors to Automatic]

Does anyone know if it was a Flap 30 or Flap 40 landing? The former does cause the aircraft to float as I understand it.

[safetypee]

Re the use of HUD:

Using ILS guidance at low altitude requires good beam quality. Some Cat 1 installations suffer low altitude beam bend and the GS may suffer more than normal ‘parabolic’ distortion, the latter taking the aircraft high.
If there is a flare mode, this may depend on the quality of the undershoot ground plane. In this instance an approach over water followed by a ‘sea wall’, etc, might not be conducive to high accuracy. Thus even where the Rad Alt reduces beam bend problems, the flare and touchdown point might not be as accurate as with a Cat 3 ground installation - and Cat 3 doesn't assure a short touchdown anyway.

In the visual mode the FPV indicates where the aircraft is going. However, this depends on the pilot being able to recognise the immediate surroundings and the aiming point, which in poor visibility can be subject to considerable error. Ultimately this is no different to a manual landing; you should always know where you are and where you are going.
Head-up speed and altitude indications should provide awareness of the flare progress, but in high stress landings pilots don’t always scan the ‘displays’ – they might ‘see’ the information, but don’t comprehend it.
Possibly time to consider HUD fixation, or the if the routine use of HUD generates technology dependency – loss of basic awareness skills.

[Kirks gusset]

Visual perspective may change with flap 30 vs 40 but the aiming point is still in the same place as is the ILS. This is simply pilot error, landing with questionable tail wind on contaminated runway at night. It could have been much worse.

[waves-dubai]

Safetypee...

You have made valid points, but I can't seem to understand:
How the pf or pnf could cross the threshold of a 8,900 ft runway and not realise they were far from the touchdown zone.......
Am pretty sure the localizer gave the accurate distance to the runway...

Also, with an initial touchdown speed of 162kts, tells me it was possible for him to initiate a G.A.