US pressured Dutch Safety Board to downplay tech faults in 2009 Boeing 737 crash
nltimes.nl
nltimes.nl
[0]: https://nos.nl/artikel/2319562-van-vollenhoven-vaker-druk-ui...
"When the truth offends, we lie and lie until we can no longer remember it is even there, but it is still there. Every lie we tell incurs a debt to the truth. Sooner or later, that debt is paid."
"I tell you, Winston, that reality is not external. Reality exists in the human mind, and nowhere else. Not in the individual mind, which can make mistakes, and in any case soon perishes: only in the mind of the Party, which is collective and immortal. Whatever the Party holds to be the truth, is truth."
"Even a fix now proposed for the Max has similarities with the past: After the crash near Amsterdam, the F.A.A. required airlines to install a software update for the NG that compared data from the plane’s two sensors, rather than relying on just one. The software change Boeing has developed for the Max also compares data from two sensors.
Critically, in the case of the NG, Boeing had already developed the software fix well before the Turkish Airlines crash, including it on new planes starting in 2006 and offering it as an optional update on hundreds of other aircraft. But for some older jets, including the one that crashed near Amsterdam, the update wouldn’t work, and Boeing did not develop a compatible version until after the accident."
I'm obviously thinking about this from a software point of view where redundancy is cheap but I am curious if it's just greedy cost cutting or if there are legitimate reasons for this stuff.
Even if cost wasn't an issue, space and weight are. There are just so much stuff you can tackle on the fuselage or inside the engine cowling before it starts affecting the airplanes ability to fly.
In other words, you can't just order from a catalog and install it. You need to order from a vetted supplier, who maintains testing and documentation, and can prove that it meets spec and also that all the materials used to make it meet spec. Then it has to be installed by a certified A&P mechanic etc. There's a long chain of events for something that seems relatively simple, but that's also why airlines are about the closest thing to a truly six-sigma organization.
Compare this to the ancient 737 where instead of the airplane detecting errors the pilots are supposed to do it by cross checking two flight computers. Thus we see now after the crashes Boeing forced to implement cross checking in a system not designed for it because things reliant on a single sensor has ended up being classified as safety critical.
There are actually very few similarities between this crash and the MAX crashes. The failure was in one part of a highly redundant system with additional warning systems. The pilots failed to correctly react to the failure and failed to do basic airmanship like monitor their speed until it was too late.
Whereas the MAX failures may well have required well-above-average piloting skills and instincts to successfully recover from, this one simply required basic competence and attention.
This is not an automatic system relying on a single sensor, as MCAS was. An airliner in flight is a complex system of which the two pilots are a very significant part. There were multiple failures here, and the technical failure was not particularly unusual. What was quite unusual was a failure of basic airmanship on the part of the pilots. A good investigation would (and did) look at how that situation came about so that training or human-machine interface can be improved to prevent it happening again. It's not about blaming, it's about understanding.
Air travel has made absolutely enormous strides in safety mostly because of rational investigations that don't focus on one party but try to understand all the factors that lead to an incident. The irrational, emotional response to the MAX incidents is eroding that.
Boeing and the FAA definitely have significant issues in culture and oversight. They should be resolved so that Boeing can continue to build safe aircraft and the FAA can effectively oversee the US aviation industry.
At the same time, not every pilot in every flight deck is the same. Some are more capable than others. Some have been trained better than others. Some come from a culture where the "power gradient" between captain and first officer is hard to overcome in order to provide effective CRM. Some come from a culture where they're afraid they'll lose their job if they accurately report incidents. All of these things, and more, are factors in air safety.
Focusing on one factor or one company really doesn't help.
Agree willingness to own up and discuss errors is critical to long term airline safety.
However you only seem to be applying this to pilots and not to Boeing!
Isn't the issue with the MAX, and potentially now earlier episodes, that for perhaps perfectly understandable commercial reasons, Boeing and even the FAA haven't been upfront and honest about engineering issues and how to improve stuff - ie they have been guilty of playing the blame game - on to pilots - who are often conveniently dead.
If you ignore the shrill media and look at the actual investigative work going on, I don't think anybody is seriously suggesting that the pilots were the sole cause of the MAX issues, and certainly nobody is "blaming" them.
At the same time, I don't know anyone working in aviation who thinks it's 100% a coincidence that the first crashes happened with those two airlines. There are always multiple factors in play.
My personal view on it is: the MAX was a flawed design due to the expectations it placed on pilot training, competency and in particular, instinct in a surprising situation. That's it. The aircraft was still capable of flying if the pilots reacted correctly to the failure, and the reaction was one that they were in theory trained for, but it's not reasonable to expect an average pilot to be able to do that instinctively in the time available.
Talking about "cost-cutting" or whether a certain engineering decision was made to save money is irrelevant, because engineers make trade-offs every day - that's an important part of their job - and every single aircraft flying (yes, including Airbus!) has a design that involves trade-offs, even in safety-related things. For reasons that are now being explored, Boeing's engineers made faulty assumptions when evaluating those trade-offs, probably augmented by pressure to avoid mandated additional training of pilots, and the regulatory oversight didn't catch it, or didn't want to due to being in a too-close relationship with Boeing. Those are the problems that need to be solved on the aircraft side.
I'm sure that given enough time you would have seen MAX crashes from other airlines, but it's no surprise to me that the first crashes (and, in an alternate reality where the MAX continued to fly, likely the majority of the crashes) were with airlines that are well-known in the industry to have lower standards of hiring, training and maintenance. Digging into that might well find some problems that need to be solved on the human side.
Sorry, I'm not as up to speed on this fiasco as I should be. Which two airlines did the first crashes occur with? And is the implication that they are subpar airlines?
I understand there are many factors at play here and Boeing is mostly responsible but I'm just curious about this.
Indicative of their company culture is that on the MAX accident flight, there was a maintenance engineer in the jump seat observing in an attempt to diagnose the AoA issue which had also occurred on the previous flight of the same aircraft. On the previous flight, the specific combination of factors to cause the trim runaway didn't occur. This aircraft should have been grounded while the issue was resolved, and instead was taken for a test flight with unwitting passengers on board.
The second was Ethiopian, who are growing rapidly and thus hiring rapidly. The first officer of the accident flight had flown an aircraft (any aircraft) for a total of 350 hours.
FAA wrote to Ethiopian in 2016 decertifying them, with 60 findings identifying a systemic failure of the entire quality management and training management systems. They were recertified in 2017 but there have been whistleblower reports that nothing significant changed, with politically-oriented decisions, nepotism, unsafe practices, task cards being signed off without executing the required maintenance actions, etc. EASA still doesn't consider the issues resolved. Look up Yonas Yohannes Yeshanew, who used to be Director of Aircraft Engineering there.
As you eluded to earlier, there are usually multiple causes. One that the above statement seems to overlook is the poor application and adherence to the system safety analysis. According to their own process, even though they misclassified the failure they should not have relied on a single sensor.[1] It seems to me there's engineering, process, and human factor causes involved.
[1]https://www.seattletimes.com/business/boeing-aerospace/faile...
I don't know much about Lion Air, but Ethiopian is considered a good airline. IT actually does a lot of training and maintenance for several major airlines in Africa. Their crews tend to be young, but that is because they are rapidly growing.
Not now the facts have come out, but originally - how about:
Congressman Sam Graves:
"In May, Graves insisted that “facts in the preliminary report reveal pilot error as a factor”. He went on to claim that “pilots trained in the US would have successfully been able to control this situation”.
https://www.bbc.co.uk/news/extra/sd9LGK2S9m/battle_over_blam...
> At the same time, I don't know anyone working in aviation who thinks it's 100% a coincidence that the first crashes happened with those two airlines.
And here you are effectively repeating that.....
You basic premise is ridiculous. The idea that because some airlines/pilots aren't as good as others, then it absolves Boeing.
Boeing should not be selling a plane into a market that don't cover markets natural variant in pilots and airlines. That's like selling a car only F1 drivers can safely drive to the general public ( and crucially without telling them it's really F1 driver only... ) and claiming it's perfectly safe despite lots of crashes because Lewis Hamilton doesn't have a problem with it.
You could also argue the second crash happened because of the blaming of the airline/crew on the first crash.
Sure there are other factors - it's a question of how many times do you have to roll the dice before you get a crash - clearly far too few in this planes case - and that is entirely Boeing's fault.
There are serious issues to be resolved by Boeing and the FAA. There are issues to be resolved around airline company culture. There are issues to be resolved around pilot training and other human factors.
This "entirely Boeing's fault" stuff just encourages a blinkered view of the situation and is exactly the absolutism I was referring to in my grandparent post.
This absolutism is encouraged by the media (just look at that BBC headline: "Battle over blame") and is thankfully absent from the actual investigative work.
While I have no idea why hot air from a congressman is relevant here, the first quote from Graves is accurate. Pilot error was identified as a factor in the preliminary report, and every subsequent report has also identified it as a factor. Obviously there are far more important factors, but it's still a factor, and every factor should be considered and possibly acted upon. As I mentioned in a grandparent comment, the multi-factor approach to accident investigation is one of the reasons air travel is so safe today.
The second quote is pretty much impossible for him to back up and should probably be treated as the usual political noise -- he is, after all, a congressman.
> Sure there are other factors - it's a question of how many times do you have to roll the dice before you get a crash - clearly far too few in this planes case - and that is entirely Boeing's fault.
You said:
>This "entirely Boeing's fault" stuff just encourages a blinkered view of the situation and is exactly the absolutism
You haven't understood what I wrote.
Let's say there is a dice - that represents the chances of a combination of different factors occurring ( weather, pilot experience, plane maintenance, chance of debris hitting sensor, cosmic ray hitting computer - whatever )
What numbers that dice rolls is entirely out of Boeing's control.
However what is entirely in it's control is what happens when particular combinations, represented by the numbers, come up.
So if there is a plane that crashes only when you roll a six, and a plane that crashes when you roll a five or a six, clearly the second plane is 100% worse and yes the individual crash depends on what the dice rolled - but the relative safety of the plane doesn't!!!
Do you understand now?
Because you wrote, "certainly nobody is "blaming" them"
Clearly that was wrong - here's another article: https://www.theguardian.com/business/2019/may/23/boeing-737-...
Where the Amercian airlines pilots union is saying Boeing blaming pilots is inexcusable - whereas you say 'nobody is blaming them'.....
> I encourage you to read what I'm actually writing rather than imagine what I'm thinking.
Hmm - see above.
This reminds me of that fatal crash with "Fast and Furious" star Paul Walker: the Porsche he was driving really wasn't fit for the road, and was only safely driveable by a race car driver. It was too powerful and the center of mass was too far to the rear (thanks to Porsche's idiotic rear engine placement), giving it a strong tendency to fishtail and lose control.
Even with the higher levels of automation, the pilots are literally there to monitor the automation and intervene when it does something wrong. The 737 tells you when it will stall in on the speed tape on the PFD, auditory and through the feel of the stick shaker.
The issue here is not anything technical, but Boeing's alleged pressure on the inquiry. I say alleged, because Boeing should be allowed to, and in fact should be expected to provide a fact-based analysis of the accident. It is ultimately for the investigators and regulators to decide if it is being tendentious and/or manipulative.
The "whether most pilots would have done the same" is probably where I see the biggest differences between this and the MCAS incidents. Every student pilot knows to monitor their speed, especially in a low-energy situation like approach and landing. Very few student pilots could successfully have recovered from the trim upsets caused by MCAS.
Every single Boeing plane has two buttons called "autopilot disconnect" and "autothrottle disconnect", and when you press them, then you can fly the plane and it will do exactly what you tell it to do. I mean fly the plane, i.e. left hand on yoke, right hand on throttle, feet on pedals. If those systems fail, then yes, blame the manufacturer. But those systems basically never fail.
* THY A333 at Port Harcourt on Dec 31st 2019, runway excursion, tyre damage
* THY B738 at Odessa on Nov 21st 2019, runway excursion and nose gear collapse on landing
* THY B738 at Lviv on Jan 6th 2019, runway excursion during backtrack
* THY A321 at Bucharest on Jun 22nd 2017, overran runway on landing
* THY B738 at Pristina on May 2nd 2016, runway excursion on landing
* THY A319 at Sochi on Sep 28th 2016, ATC prevented controlled flight into terrain
* THY B738 at Nice on Aug 27th 2016, went through extended centerline and descended below safe height
* THY A321 at Tbilisi on May 5th 2016, hard landing
* THY A343 at Antananarivo on Jan 9th 2016, touched down short of runway
* THY A333 at Kathmandu on Mar 4th 2015, runway excursion and nose gear collapse
* THY A320 at Istanbul on Apr 25th 2015, hard landing, go-around, engine problem, gear problem, gear collapse, runway excursion
I could go on...
I’m not qualified to judge how much of this was down to pilot error, but the risk is ascribing it purely to this and moving on takes away emphasis from redesigning the system to give the flawed human beings piloting the plane the best chance of making the right decision next time.
The original NYT article, with only a few HN comments:
https://news.ycombinator.com/item?id=22098270
"Boeing’s Responsibility in a Deadly Crash ‘Got Buried’"
From my understanding, every 737 has two flight computers, but only one is in command at a time (I.e. generating commands to drive things like A/P, A/T). If other hardware in the plane is connection-wise split between the two computers, you will end up in situations where the two computers fundamentally disagree on what reality is.
The only backup in that case is good airmanship to know that you are in an anomalous condition, diagnose whether you have a set of gauges that you can rely on, and finally, what to do about it. ATC could have been utilized to cross-check instruments.
This is honestly non-trivial machine operation 101. Your sensors can be wrong.
A resolution I might suggest moving forward is an MFD function that allows pilots to manually remap avionics data sources in flight. Structure it as a Source->Sink system. Then again, I'm reluctant to drop that type of burden on them as it sounds quite similar to the structure of certain other software of I'll-repute; however, it would provide a remediatory mechanism a Pilot could utilize in flight.
Though it doesn't help at all if everything is hosed. That's what A/T A/P disconnect and stick and rudder skills are for though.
Overall, it does strike me as possibly being same in type in terms of manufacturer philosophy playing a part; but in this case there was enough done wrong by the pilotsthat it doesn't necessarily seem like a worthwhile hill to die on to go after Boeing for it.
Interesting there was explicit pressure against doing so though. That raises questions, but sufficient time has gone by that likely most documentation to be investigated is long gone.
[0]: https://www.onderzoeksraad.nl/en/page/16227/turkish-airlines...
this is completely different from the mcas where the default configuration lacked redundancy, the pilot lacked training and the bypass procedure ineffective unless executed at low speed and high altitude before the dive would lock the trims