Monday, April 20, 2009

Choose Your Own Misadventure

You are the commander of a Boeing 737 passing through 10,000' and setting up for a visual approach. You realize that you have misjudged the wind and your descent profile. You are likely to be too high and fast. You:
a) extend speed brakes
b) fly further out than usual before turning final
c) recheck your calulations
d) nose down and go faster, hoping it will work out

You are still high and fast. You call for flap 15 but the co-pilot does not respond to your request. You:
a) ask him why he did not do as you asked
b) yell at him
c) extend them yourself
d) continue the approach without flaps

You are on final approach and the flaps are set to five degrees extension. The manual calls for at least thirty degrees of flaps for landing, but the manual also calls for a much lower airspeed before extending that much flap. You:
a) try to extend the flaps anyway
b) leave them up and land anyway
c) conduct a missed approach

You believe the flaps are malfunctioning, as you have selected them down but the airplane is not slowing as it usually does. You:
a) continue the approach at high speed
b) check all your power settings and other parameters
c) conduct a missed approach, extend the flaps at altitude and have a crew member go aft to the passenger cabin to confirm the position visually

You encounter windshear on approach to a runway. You:
a) conduct a missed approach in accordance to your company procedures
b) land with minimal flaps at above normal speed
c) continue at whatever speed and configuration you can attain, just to get the airplane on the ground.

You are at the position on the approach where your company manual calls for you to be stabilized. You are so high and so fast that the airplane automated systems think you are crashing. You:
a) attempt to land
b) conduct a missed approach

You are almost a hundred feet above the runway threshhold, well above landing speed and not in landing configuration. The copilot is telling you to conduct a missed approach, and does not confirm landing checks complete. You:
a) attempt to land
b) conduct a missed approach

You touch down over a third of the way down the runway, at well over maximum tire speed. The airplane bounces twice, pulling almost three Gs, and the nose gear collapses while the copilot yells at you to go around. One of the engines comes off its pylon. You:
a) kill 22 people and injure almost everyone else.

Sadly, there is only one choice in the end, and that is the adventure a real life commander chose.

Indonesian pilot Marwoto Komar has been sentenced to two years in jail after being found guilty of criminal negligence for attempting to land a Boeing 737 in the wrong configurations and at almost twice the normal speed. The jet overran the end of the runway into a rice field. Twenty-two people were killed and fifty were seriously injured in the crash and ensuing fire. The Daily Telegraph story here includes a dramatic photo of the post crash fire.

According to the investigation report, the aircraft was already fast through 10,000' on the approach, as they exceeded the crew accepted a straight in visual approach to Yogyakarta but continued on the ILS approach without informing ATC. As long as someone is looking out the window, there's not a lot of difference between a well-flown long straight in visual and an ILS approach, but unfortunately this was neither. The aircraft arrived overhead the initial approach fix high and fast. Not just a little bit high, but at 3927' instead of 2500'. It wouldn't be the first time a crew had cut a few corners and come in too high, but the captain, who was the pilot flying, elected to rectify the situation by lowering the nose into a steep descent. There's an expression in aviation: you can go down or slow down. They had been trying to do both for the whole descent, at one point reaching 293 knots below 10,000', knowing they had a tailwind.

The captain called for more flaps, but the first officer did not extend them because the airspeed was 33 knots over flap extension speed. The captain continued to ask for flap fifteen right through the approach, but the FO never extended them nor gave a reason why he was not doing so. The ground proximity warning system reacted to the configuration, speed and rate of descent by issuing alerts and warnings, which the crew ignored, fifteen times.

They arrived over the beginning of the runway still 89 feet in the air and travelling at 232 kts with flap 5 (as opposed to the manufacturer's recommended 134 kts with flap 40). The aircraft touched down 860 m from the threshhold (more than a third of the way along the runway), bounced twice, then landed again still well over the recommended tire speed. The nosewheel tire burst, sending up sparks as what was left of the gear scraped along the runway. Both thrust reversers were deployed, and presumably the brakes were applied, but with too much speed and too little runway remaining, the overrun was inevitable.

It does not appear that there were any mitigating issues such as approaching weather, critical fuel, or pre-existing emergency.

There are so many issues here. I'll touch responsibility, training, co-pilot assertiveness and the role of the criminal justice sytem in air safety.

The captain never took responsibility for the crash. "Lack of remorse" was even cited as a factor in his sentencing. Compare to Captains Sullenberger and Haynes who saved lives landing airplanes crippled by circumstances beyond their control, but later agonized over what they could have done better. It's almost bizarre how Komar blamed everyone and everything but himself. He attended the trial in uniform, and fully expected to return to work.

He claimed windshear at touchdown. It's a known problem in the area and something a captain with over 13,000 hours of experience should be able to recognize, but there were no other reports of windshear in the area, no indication that it might be present from the detailed winds aloft, and a professional approach marred by windshear is inconsistent with the speed profile recorded throughout the descent.

He claimed that the flaps were faulty and did not extend when selected, but investigation found "no evidence of any defect or malfunction with the aircraft or its systems that could have contributed to the accident." The chief crash investigator, Mardjono Siswosuwarno, said the aircraft's wing flaps failed to extend for landing and that might have been caused by the high speed, but that's not something the captain can blame on the flaps.

The captain also levelled blame at the co-pilot, for not lowering the flaps, and at the poor emergency services. Admittedly the copilot did not clearly refuse the flap extension command by stating "Unable, we're thirty knots high" or whatever his SOP callout was, and the emergency services at the field were poor and contributed to the death toll, but none of that excuses the captain's actions.

The PIC said that he was unaware of the actual airspeed and expected that the copilot would inform him of any speed concerns, but that doesn't align with the comments he made during the descent, or the knowledge that someone with 3700 hours on the airplane would have to possess. The copilot called for a go around before and after touchdown, and the captain did not respond. It has been suggested, (but the captain has denied) that a company incentive to save fuel led to the decision to continue the unstable approach.

Poor training probably contributes to the decisions the crew made. For example, training records showed that the pilots had attended an "introductory" GPWS course with no evidence of related sim training. Non-pilot staff seem to be lacking some basic training in public relations, seeing as they got took their pictures taken, smiling, with the burned out wreckage. But how much training does it take to choose a missed approach at some point in that accident sequence? The copilot doesn't seem to have had sim training in the actions he needed to take in order to take control of the airplane from a captain who has not stabilized the approach. He did try. He said, he had shouted at the captain to go around because "that was the proper procedure ... to ask people to go around with yelling".

And that's tough. You need training, confidence and a company culture that will support taking an airplane away from a captain. This co-pilot stuck up for the captain after the crash. I don't believe that he literally blacked out from the g-forces. I think he needs a story to tell himself to explain why he did not take control and perform the missed approach. It reminds me of Michael Origel, the co-pilot of American Airlines flight 1420 that overran the runway in Little Rock Arkansas during a landing in a thunderstorm. Origel insists that he called for a go around, despite the fact that experts can find no trace of it on the CVR. I believe him. The voice in his head screaming that this was not right was so loud that he can't believe the captain and the recording microphone didn't pick it up. Has there ever been an accident in which a copilot's loud demands that a captain go-around were cited as a cause or contributing factor? Ever hear of a copilot fired for asking a captain to go around? I can name names, hell I have a pocket in my formal coat containing funeral programmes that name names of people who might be alive today had they demanded a go around.

But considering all the things that people did or didn't do, should or shouldn't have known to do, was a criminal act committed? They broke company policy in the lack of a stabilized approach, but I don't know that they broke air regulations. Could the possibility of the CVR being introduced as testimony in court cause pilots not to admit errors in flight?

It is very uncommon for a pilot to be charged with a crime in connection with a crash. It's a somewhat disturbing precedent when you're a pilot. We're usually assumed to be doing what we believe is best for the flight, in defense our own necks, and air law gives us a lot of latitude to do things that would otherwise be against the rules to defend our safety. Any misjudgements we make already force us to face possible injury or death and then we have to defend them to our own conscience, our boss, and the aviation regulatory authority. Do we need to add the civil justice system?

The decisions the captain made during this flight seem reckless. I think his conviction is closer to the correct outcome than having the manufacturer sued for the flaps failing to deploy at excessibe speeds, but I'm not convinced that the justice system should have a role here.

21 comments:

ZD said...

So, why doesn't some government licencing body pull this captain's ticket so he never flies again?

Anonymous said...

Train conductors, bus drivers, and doctors have been sentenced to jail time for deaths caused directly or indirectly by their actions while on the job. Why would/should it be different for pilots? Even ordinary citizens driving down the street can get jail time for making a couple of unfortunate (or unintentional) decisions which end up in tragedy.

I think that the responsibility comes with the four stripes on the uniform, just like it does in other occupations where you are put in complete control of the lives of others.

The very dangerous follow-up question that will probably highly offend many readers (but that I will ask because I'm not a pilot) is whether or not you think the aftermath would be any different had this pilot been in the US and belonged to a union? Did he belong to a pilot's union in Indonesia?

IMHO keeping all the accountability within aviation seems like a very slippery slope. It would be great if it was never a problem, but this case seems to show that sometimes external review is an unpleasant but necessary reality.

As an example, the comittees that review and approve animal research procedures in the US are required to have one person whose primary concerns are not scientific, and another member who is not affiliated with the institution that they are reviewing for. This may lead to frustration on the part of researchers, but sometimes the outside viewpoints are necessary to help keep perspective on things.

Norman

Unknown said...

Re the statement: "...but I don't know that they broke air regulations..."

I don't have the Air Regs memorized but I'm pretty sure there must be a blanket statement in there somewhere about the necessity of not endangering passengers or fellow crewmembers...? And I'm not a lawyer, but I'm guessing that if such a statement is not in the Air Regs, then it exists in other codes of law that could be invoked - reckless endangerment, negligence causing harm, etc...?

This kind of gross mismanagement of an aircraft crosses the line to the point where civil law should be invoked. Otherwise it's akin to Doctors and Police Forces and Lawyers and Priests and Financial Investors having no direct accountability to the public. And we've seen enough abuses in those fields to know better by now.

jinksto said...

I think "G" and Anon/Norman have it nailed. If there is gross misconduct involved then there has to be a civil penalty.

You've listed out many missed decision points. In guessing the answers I'm pretty sure that I got the answers right.

I fly SEL and that scenario applies to those as well as a 737 which means that this pilot KNEW he was working near or over the line for safety. He'd have to. He would have learned those things from the first week of training and every day since then. He knew he was wrong, he did it anyway. The accident to most of us is incomprehensible. He violated both company policy and common sense flying rules. I'm not one for criminal penalties that are unwarranted but at some point you have to accept that the level of negligence goes beyond merely foolhardy into criminally stupid.

As my instructor was fond of saying, "The FAA will almost never ask you why you did a go-around but they will Always ask you why you didn't do one."

Aviatrix said...

I don't have the Air Regs memorized but I'm pretty sure there must be a blanket statement in there somewhere about the necessity of not endangering passengers or fellow crewmembers...?I've never seen the Indonesian air regulations, don't even know what language they might be written in, hence my statement that I don't know. They are probably based on the Dutch ones, seeing as they're a former colony. If a blanket statement is all there is to go on then it's very subjective. The captain can maintain, as he did, that he took the safest course he knew and that it wasn't his fault and there was no decision to endanger. Is being bad at your job a crime?

A Squared said...

Is being bad at your job a crime?At the risk of stating the obvious; apparently it is in Indonesia.


Recently in the US a pilot was giving someone a ride and hit a power line over a river, and his passenger died. There was a great deal of hand wringing about jailing pilots for accidents. And as much as I am against jailing pilots or anyone for true accidents, I have to concede that the pilot did make a conscious decision to violate the aviation regulations, and his passenger died as a direct and immediate result of his violation of the regulations.

Generally speaking, if you kill someone because you're breaking the law, it's a crime. I've tried to come up with a compelling reason why this pilot shouldn't be charged with a crime like non pilots in similar circumstances, but I've been unable.

The case of the Indonesian pilot is less clear, but I would imagine that he probably had a legal obligation to fly by company procedures, and I'm almost positive that his company procedures forbid continuing an approach that is that badly off profile.

Anonymous said...

jinksto: Near the line of safety is normal ops, whether we acknowledge it or not. That is to say, aviation is unforgiving.

I think cases like this would be best settled by peer-review. Other pilots know better than anyone else how far over how many lines this guy was, and I don't think they would be too lenient, contrary to what some may believe.

Anonymous said...

I would think that the medical profession would be close to the professional pilot when it comes down to post-event evaluation of real-time decisions which did not have a good outcome. The predictability in medicine is some what fuzzy, to the point where the phrase "the art of medicine" is often used. A doctor can do everything right (or make a minor mistake) and someone dies, or they can perform a massive cluster f**kup and then save the patient and there is no issue.

And let's keep this above the anomaly of litigation in the US, where doctors gets sued if the outcome is not perfect.

Not being part of the medical profession, information is limited to what one reads in the press (an admitted exposure to filtered information). However, the College of Physicians (here in Ontario) does the peer-review of doctors, and their "conviction rate" is miniscule. Major screw-ups get a note in your file, massive screw-ups require someone to review your work periodically, and you'd pretty much have to butcher a small town to get a suspension.

It's rather like judges reviewing judges - there appears to be a inate resistance to causing permanent unemployment to someone in the same profession as yourself. By extension, I would not hold much hope for pilots reviewing pilots.

Which is why the general public should have the ability to sue, and why the regulators have the ability to pursue criminal or civil remedies. Sometimes the miscreant is in the wrong job (pull the ticket), and sometimes while they did the job they did not demonstrate a mistake or an error in judgement, they clearly demonstrate that their actions rise to the level of a criminal disregard for the predictable outcomes of their actions.

Sarah said...

Ok. I choose "abccabb" and avoid the last choice. The most valuable tip I received in training was "when in doubt, imagine how the NTSB report would read". So maybe I'm over cautious - but so far, not sorry.

By the way, (cough) "you go, girl!".
We're waiting...

dpierce said...

I'm projecting US law on Indonesia, but I've sat on two juries that dealt with negligence claims, and in each, the arguments ultimately boiled down to "did the person exercise a 'reasonable' standard of care?" or put differently, "would a 'reasonable' person have done what the accused did?" (or something along those lines).

The law being the law, you can twist and argue this ad nauseum in either direction, but I think he clearly sailed into negligence by that standard.

Of course, in one trial, members of my jury thought it is 'reasonable' for a doctor, who is on vacation and has transferred all of his patients to another doctor, to continue to call in to the hospital hourly for updates and make recommendations about the care of those patients.

Chris said...

I tend to agree with you that jailing pilots for making errors is a slippery slope, in this case I find it hard to defend the pilot involved.

Should a bus driver be immune from prosecution if he is driving dangerously, speeding, etc, and has an accident which could have been avoided and which kills his passengers?

There were many contributing factors here, all of which seem to have been glossed over by the prosecutors for reasons we can only guess at. The airline is at fault for insufficiently training its pilots. The regulator is at fault for inadequately overseeing the airlines operations. The airport operator is at fault for having a substandard emergency plan, and so on.

If a pilot makes a single error, or even a sequence of errors in good faith which result in a fatal accident, leeway should be given. But on balance, in this case, the sheer number of negligent and dangerous actions made the court decide to punish the Captain.

The real failure, in my mind, is to pursue remedial action against all those other incidental factors. It's as if now they have found their scapegoat, everyone else can go back to the way things were and pretend nothing ever happened.

gmc said...

Chris says: "The real failure, in my mind, is to pursue remedial action against all those other incidental factors. It's as if now they have found their scapegoat, everyone else can go back to the way things were and pretend nothing ever happened."

Excellent point! Perhaps that Reason Swiss cheese model of accidents can be applied somehow and the appropriate amount of blame, correction and penalty applied where necessary... but I guess that's asking a lot.

The main thing however, is that all the contributing factors should be analyzed. Even this pilot did not get up that morning and say to himself, "I think I'll purposely screw up an approach today and kill some passengers." So why then did he do just that?

Mysterious...

Grant (Falcon124) said...

I'm all for good investigations and no recriminations so we can get the facts and make improvements - aviation safety is built on this. Usually, it is proven that all involved were doing the best they could under the circumstances but that changes are required in many areas.

In this particular crash though, the pilot was not only ignoring the sterile cockpit rule below 10,000 feet, he was apparently SINGING! He ignored cockpit warnings. He ignored his copilot begging him to abort. He definitely landed way too fast (many of the surviving passengers who'd flown frequently reported on the high speed on landing) and broke every rule in the book.

The pilot had over 13,000 total hours and 3,703 on type yet he still made this most disasterous of errors and then tried to say it wasn't his fault.

I wouldn't have been going for criminal negligence, I'd have been going for insanity or imparement due to booze or something. It's interesting to note in the accident report that the pilot did not supply information about what he'd been doing in the previous 72 hours.

For those commenting about more than just the pilot at fault, the investigation revealed a number of faults in the airport and the airline as well as the pilot, with Garuda's training and checking getting a major overhaul after this (the pilot had a history of landing faster than required during sim checks). As a result of this accident, a number of major changes have been made at the airport and within Garuda.

chris said...

was a criminal act committed?I don't see how there can be the least doubt.

Aviatrix said...

A criminal act is when you break the laws of the jurisdiction. The only one the pilot appears to have broken is the one restricting a pilot to 250 kts below 10,000' unless otherwise authorized. He almost certainly broke numerous company regulations, and someone speculated that there might be a law requiring him to follow those.

He did a bad thing, unquestionably, but bad things are not always criminal under the local laws.

What law are you certain he broke, Chris?

nec Timide said...

criminal negligence - (law) recklessly acting without reasonable caution and putting another person at risk of injury or death (or failing to do something with the same consequences)

Most jurisdictions have some similar statute. Doing bad things may not not always be criminal, but doing bad things that can be reasonably expected to result in harm to others is, or should be. If this pilot had, at the very last minute, relented and gone around to make a successful landing then I would agree. But pressing on into what his professional training, and years of experience had to tell him was a dangerous situation, with no reason to do so, in my mind makes it clearly criminal.

Skipping stones across a pond, generally not a criminal act. Skipping stones across a pond full of bathing toddlers...

Chris said...

In case there is some confusion, there are two commenters named 'chris' here. I'm not a lawyer, so I'm not qualified to speak on legal matters.

But as for the choose your own misadventure part, I went with AAC, which renders the subsequent parts irrelevant. If things don't look or feel right, it's always best to go around until they do, I reckon.

A Squared said...

What law are you certain he broke, Chris?

Article 359 of the Penal Code of Indonesia which provides: Any person by whose negligence the death of another person is caused shall be punished by a maximum imprisonment of five years or a maximum light imprisonment of one year.

kbq said...

How to properly park your aircraft after an unauthorized landing... :)

http://www.liveleak.com/view?i=945_1240357993


Kevin

Anonymous said...

In principle, the law is supposed to apply to everyone equally without regard to profession, wealth, or status.

The onus should be on those who oppose the criminalization of crashes to justify why pilots should be exempt from existing manslaughter and criminal negligence statutes, not on everybody else to justify why pilots should, in the appropriate circumstances, be prosecuted for criminal negligence.

Is there a case to be made that prosecuting pilots is contrary to the public interest and/or harmful to public safety? Perhaps. But that case has to be made explicitly and not assumed a priori to be true.

Anonymous said...

Aviatrix, the Indonesian DGCA CASR's are based on the FAA FAR's.

The sentencing of Komar in this case is the subject of much debate amongst my workmates.

I whole heartedly do not believe in the criminalisation of breaches of aviation safety but a line does have to be drawn.

This accident despite the sentencing was at least the catalyst for changes in Indonesian Aviation. It is however still atrocious and in just the last 3 weeks we have seen another 3 fatal accidents but things are changing for the better, abliet glacially.