(The talk used to be available on YouTube, but was removed sometime in the last year by copyright claim. I can't find an alternate source for it. The video was at https://www.youtube.com/watch?v=mZYHMMMeHic)
It's such an inspirational story to read about, even though a lot of people died.
This was one of the strongest proof points that crew resource management really worked. It was the final nail in the coffin for forces that resisted or only reluctantly adopted CRM. It's clear that it took a whole lot of teamwork to get that plane down.
Aviation is much safer than it was in the 80's, and the lessons learned from relative successes like the Sioux City crash can be thanked for it.
Interesting random fact about this crash. On board was a man by the name of John Stille. He had invented a very versatile carbon-carbon bond formation that bears his name “the Stille reaction”.
His work eventually resulted in a Nobel prize for the scientists who built upon it.
A similar incident took place in 2003 on a DHL cargo plane after being hit by a surface-to-air missile. Miraculously they did manage to land the plane intact despite the total loss of hydraulics and needing to rely on engine thrust alone to control the plane.
The Mayday episode might also mention (least from my memory of last watching it) an experimental NASA program developed sometime after this crash that would allow "full airplane control" via computer in the event one more more control surfaces could not be used. It achieved this by translating normal flight inputs into sophisticated throttle manipulation of the engines, just like the pilots on flight 232 did, but with the reflexes and modeling of a computer.
With all the attention given to the 737 MAX lately, it's easy to forget that airline travel is dramatically safer today than it was just 30 years ago when this incident took place.
Fatalities per revenue passenger-mile have decreased by a factor of 15x from 1989 to 2018 (and 54x from 1970 to 2018) [1].
Yes, especially in the US. I think the only crash of a large commercial airliner since 9/11 was AA 587. In the 80s/90s it seemed to happen once a year.
Is that just because personal and private aircraft tend to have fewer passengers? If we measure fatality by "per mile flown" instead of "per passenger-mile" would there be a difference?
Less experienced pilots, less reliable equipment, poorer planning, overconfidence.
Just one of many examples of what can go wrong: overreliance on instruments and information that is lagged vs. reality, around severe conditions, at night. Five souls lost. ATC had repeatedly advised the pilot against his course of action, to no avail.
Remember that the failure we complain about on the 'MAX was an effort to improve margin on the design (to make over-rotation and nose rising on climb a lot less sporty on a heavier aircraft)... that was completely screwed up by both Boeing and regulators.
Mistakes will be made; automation in the name of safety has killed people and will continue to kill people. But overall automation and resource management in the cockpit have conspired to make aviation incredibly safe compared to the past.
So things get better over time but instead of recognizing the advances made, you take the point of view that nothing is good enough because it could have been better?
Well, you can chalk the first 737 MAX crash up to greed, hubris, bad design decisions and a generally rotten culture in a once great, engineering driven company.
Not grounding the plane after that, despite the fact that Boeing must have had the knowledge of what have gone on is corporate mass murder for profit.
you take the point of view that nothing is good enough because it could have been better?
Yes, it would definitely had been better if 346 people wouldn't have been killed by a greedy corporate behemoth, which very obviously cares much more about profits than about the people they killed. This point is pretty obvious when you listen to their mealy mouthed pr bullshit.
That plane crashed because of an unpredictable freak event. The Max planes crashed because of budgets and sidestepping a poorly run regulatory body. I’m sorry, but when capitalism starts crashing planes, I tend to forget about “advances we’ve made.”
> That plane crashed because of an unpredictable freak event.
No, it crashed because of a predictable event that a number of engineers had already warned could happen, based on an obvious design flaw that was not fixed.
I attended a presentation by Captain Haynes back in the early 90's. He showed several photos of the DC-10's hydraulic system. In many places the triple-redundant hydraulic lines were parallel and in close proximity to each other, close enough to be taken out by a single piece of shrapnel from an uncontained engine failure.
Funny you point to "capitalism" where everything in your statement points to cutting corners which has nothing to do with capitalism per se. We would not even have planes that cost hundreds of millions if we did not have a system for investments.
If you replace "capitalism" with "competition for low prices" then I think the comment is a little less inflammatory while still making a similar point. The fact that a capitalistic system is what gives rise to macro economics that facilitate building planes like this doesn't mean it's perfect and can't be improved upon, say with regulation that doesn't outsource the checks to the company making the planes. Capitalism is fantastic, but the extremes it gives rise to can be checked by assiduous oversight. I'm not saying the government is a paragon of efficiency here, but it's better than nothing, and with some reform could be much better.
I could point to today's commercial airplanes which exist in a culture & regulatory framework, not pure free-market capitalism, that has consistently reduced the risks of commercial flying for decades.
I could point to a capitalist society where customers demand and expect safety and are part of the process that brings forth regulations. I'm sorry I didn't imply we live in a pure capitalist society.
A fundamental principle of capitalism -- the pursuit of financial profits, that is, revenues in excess of recognised costs -- leads directly to multiple practices which prove hazardous at scale and over time.
Financial and economic theory don't recognise ecological inputs including the time and replenishment rates associated with nonrenewable resource formation. (By definition, a "nonrenewable resource" is on whose extraction or extinction rate exceeds that of its formation.) It invokes utterly nonscientific principles such as Hotellings Rule (precisely zero geological references cited) and the Rule of Capture to justify pricing and business behaviours.
Capitalism relies on and exercises externalities of pollution, risk-shifting, and cost-burden shifting to increase firm profits at a cost of larger social burden.
The legal structure of incorporation is literally a risk-externalising engine, recognised by name in some cases ("limited liability corporation"), in which risk-based costs are externalised or stoplossed. This can be useful in cases, but can also be grossly overused.
"Wealth, as Mr Hobbes says, is power" is among the shortest sentences in Adam Smith's Wealth of Nations, a work not given to short sentences. Financial wealth leads to political and market power, including the power to distort markets. Much of Smith's book was a criticism of such practices. That lesson is largely lost by modern boosters of the work, who instead prefer stringing together synthesised passages across hundreds of pages of multiple sections of the book to advance their own, power-centralising ideologies.
Capitalism, finance, and business have a schizoid and faulty relationship with risk. "Cutting corners" is precisely a capitalistic measure, one which arbitrages a potential cost with a present realised savings. The bet is that the potential cost will never be realised. Sometimes that's not the case, but sometimes it is, and the firms which win that bet, based entirely on luck, are rewarded by the market. The market makes no distinction between profit gained by full realisation of probabalistic costs, and profit gained by ignoring those costs, in the individual case.
(In the aggregate profile case, the risks balance out, but specific performance matters and is independent of the profile returns.)
In market returns to various goods and services, pricing behaviours of commodities (undifferentiated bulk goods), services, labour, rents, assets, and highly differentiated specialty goods vary. In particular, the iron law of wages (tending to subsistence, or lower, levels) and the law of rents (capturing all consumer surplus value) mean that over time wealth accrues to rentiers and owners of capital, at a cost of labour.
Addressing these issues would require massive changes to accounting, risk premium, and legal regimes. That same "wealth is power" dynamic means that such reforms are exceedingly unlikely, because those who benefit by false models also benefit by ensuring that those models are protected and preserved.
"A fireball and smoke rose from the middle portion of the plane as banks of seats began vaulting and somersaulting high above the flames. Some of the banks of seats were thrown far above the fuselage in great parabolas, shot as if from a cannon by the centrifugal force of the aft end of the fuselage swinging in its majestic, flaming arc. What must it have been like to take that ride, alive, aloft, alone, aware, unhurt as yet, and looking down on the green earth?"
It was a grim situation, but I don't think the end of the article does justice to the fact that nearly two-thirds of the passengers survived what by all rights should have been completely unsurvivable, thanks to the sterling performance of the flight crew and the flight instructor, as well as the flight attendants for keeping everyone calm and prepared, and the control tower personnel for getting a prodigious amount of local firefighters and EMTs on the scene before the plane arrived.
It was, as Wikipedia dryly puts it, "a prime example of successful crew resource management," and had a big influence on training practices as airlines sought to replicate that professionalism.
United Flight 232 was an engineering disaster, but a piloting miracle.
Between the four pilots they had nearly 90,000 hours of flight time. Haynes had 30,000, of which ~7,200 where on type. Records had 20,000, Dvorak 15,000 and Fitch had 23,000. Fitch had amazingly practiced flying in a simulator with throttles-only after hearing about JAL 123.
If any crew was going to land this plane, it was this one.
Amazing how that turns out. Like when Sulley had to ditch that flight into the Hudson, and just so happened to be an instructor of water landings. What are the odds?
"One way of looking at this might be that, for 42 years, I've been making small regular deposits in this bank of experience: education and training," said US Airways Capt. Chesley "Sully" Sullenberger. "And on January 15, the balance was sufficient so that I could make a very large withdrawal."
Could be some survivor bias here. One 747 Max happened to have a jumpseat pilot who knew how to fix the MCAS issue and the plane flew on without issue and thus without much media coverage. [1]
Presumably, 2 other 747 Max's didn't have enough SME's on board to save them and crashed.
I heard that expression on live television in 2008 which turned out to be a nonevent: https://www.reuters.com/article/us-usa-american-miami/faulty... I happened to be in the USA and watching this on TV and they said this about the pilot, who had some ridiculous amounts of flight experience, if I remember correctly decades in the Air Force and then even more as a commercial pilot and they said if this happens, he is the best to happen to. He landed that plane, no problems.
I notice that the flight instructor disobeys the captain, who says to says to remove power from the engines, and does the opposite right as they're touching down.
Was there ever any thought as to whether that contributed to the loss of life?
I read the wiki article[0] about this after reading the article and found it a lot more informative. It says just before touchdown "Fitch noticed the high sink rate and that the plane started to yaw right again, and pushed the throttles to full power in an attempt to mitigate the high sink rate and level the plane". It seems a more complex situation than him just deciding to apply power to the engines as the right wing would likely have made contact with the ground anyway.
I wasn't able to find any speculation if Fitch could have acted differently and had a better outcome in the official report[1]. The report states that the "captain's acceptance of the check airman to assist in the cockpit was positive and appropriate".
Regarding disobeying the captain we can see from transcripts from the CVR that the captain (probably) knew what Fitch was doing:
Records: "Close 'em off."
Haynes: "Left turn."
Records: "Pull 'em off"
Fitch: "Nah, I can't pull 'em off or we'll lose it, that's what's turning ya."
Unidentified: "Okay."
Fitch: "Back ya’ll!"
Haynes: "Left, Left throttle, left, left, left, left, left, left, left, left, left, left, left!"
Ground Proximity Warning System (GPWS): "Whoop whoop pull up. Whoop whoop pull up. Whoop whoop pull up."
Haynes: "Everybody stay in brace!"
GPWS: "Whoop whoop pull up."
Haynes: "God!"
[Sound of impact, end of recording]
My impression is that they might have had a better result if he'd obeyed orders, or they might have pancaked into the ground at an even greater velocity and killed everyone.
Maybe a simulation could figure out, after the fact, which one it would have been. But it's a moot point, because there was no way either the captain or the flight instructor could have known for sure. Each man did what he guessed was best in the heat of a impossible, unprepared-for situation, and they managed to save nearly two-thirds of the passengers. IIRC, The investigation praised them all and exonerated them completely, which the FAA doesn't do lightly.
Presumably using only one engine would have prevented the plane from flipping like it did, but it might have smashed into the ground harder or turned into a fireball.
Ultimately it is impossible to say what kind of effect it would have had, though I’m inclined to say it would likely have been better if the thrust was applied equally.
Thrust was applied equally in the cockpit, but jet engines don't respond immediately and don't spool up in sync. Normally that's not an issue because you're not making large adjustments in short spans of time, you can use various control surfaces to compensate for the brief asynchrony, and you're far enough away from the ground to have room to maneuver.
If you remember the Korean airlines flight that crashed at SFO a few years back, the spool up time for the engines came into play as well. The pilots noticed the issue with too little time for the engines to deliver more power before the plane crashed.
Yah. The thing is, long-period oscillation modes (phugoid, etc) exist in aircraft but are so readily controlled by pilots with conventional controls that they mostly don't matter from a certification standpoint (and would be difficult to avoid, anyways).
But jet engines are slow enough that any significant control of 10-30s period oscillations with engine thrust alone would be a real, real challenge.
This is one of those classic aviation tales I'm never not going to re-read when they pop up, right there with the Gimli Glider [1] and the various SR-71 stories.
“Ladies and gentlemen, this is your captain speaking. We have a small problem. All four engines have stopped. We are doing our damnedest to get them going again. I trust you are not in too much distress”
And for absurdity the one where the pilot was sucked out of the window and the cabin crew had to hold on to him
I wonder if the same McDonnell Douglas corporate culture behind the flawed DC 10 is what led to the MAX? It's often said Boeing's acquisition of Douglas was more of a reverse takeover.
On a side note, I once met an aircraft investigator who called it the Death Cruiser 10, and that it put his kids through college!
One thing I never understood as a kid, was why wasn't there better footage of a crash. Many people at the airport knew the plane was coming in, and they had plenty of time. Yes this wasn't 2019 where everyone has a camera in their pocket, but nonetheless I always thought the could have done better than the crazy shot through the chain-link fence.
It seems to me the reason is simply that it wasn't anyone's job to make sure there was really great footage of the crash. Everyone was too busy preparing to respond to an imminent disaster in which nearly 300 lives were, in all probability, about to be lost.
"Hey Bob, you have that VHS recorder stowed away somewhere in the locker room, don't you? This guy on something called a website 30 years in the future is going to really wish that we got great footage of a horrible crash in which dozens of people die. Why don't you run out a little closer to the airstrip where a nearly completely out-of-control airplane flying at an insane speed is about to slam into the ground? Future spectators will really appreciate it."
> We only gave them two minutes to line up for runway 22. They were all set up for runway 31. And when I told Kevin we saw this runway ahead of us, and that's where we were going to land, he had two minutes to get the equipment off the runway. They were actually positioned on that runway, and we could see them moving off that runway. That's the reason the video is no better than it was. With all the notice we had, everyone was set up for runway 31. The two hangars that are built right off to the left of runway 22 were built after the airport was closed, and you can't see the runway from the tower. And that's where the operator was. So he went clearing down the steps, and if you've ever tried to get down tower steps in a hurry, it's a long way down, and that's as far as he could get, before we got there.
I also remember something about Sioux City itself - the hospital was at shift change, so they were able to have just about 2x the normal amount of staff available.
They had enough advance warning and were able to start getting more fire/EMS resources heading that way as well.
Would ditching (assuming there was a large body of water near) or landing in a huge cornfield (like Ural Airlines 178 out of ZIA (Moscow-Zhukovsky)) have made it better or worse?
Disclaimer: this is total speculation from someone completely unqualified to speculate about aviation
Given the violence of the crash due to the total lack of fine control, likely worse.
A fire in the cornfield could have killed many more people as would the lack of emergency firefighting and medical attention.
A plane that broke apart in the water would likely just lead to most passengers quickly drowning. It might work if the plane remained intact, but that requires landing level and slow, something they had no ability to control.
Did the kids without seatbelts all die? I can't imagine what I would do as a parent of one of those babies if I were told to put them on the ground. Like... Hell no!
I was a senior in high school in a suburb outside Minneapolis, and I remember it very well. What was so heartbreaking is that it looked like they were going to be fine, right up until the very end when the right wingtip went down, dug in and they started cartwheeling. After seeing that, I was amazed that anyone survived.
"When the plane pancaked onto the runway, more than 10,000 pounds of kerosene came out all at once from the ruptured right wing and turned to mist."
I wonder why they didn't "dump the fuel" just before the crash to have avoided the fire. I thought this was standard operating procedure to dump fuel before crash landings.
I had read about the crash awhile ago and as I recall many people that died died of smoke inhalation from the fire and not the impact itself.
Possibly the pilots were reluctant to dump too much fuel, since the two remaining engines were their only method of steering the plane. A lack of fuel was the last thing they needed.
Interesting. Or perhaps it would create more balance issues. Regardless, it sounds like it was considered by someone with more experience flying airliners than myself.
Dumping fuel requires good control of the aircraft, which they didn't have. Plus it would have added more workload to a flight crew already overloaded. On balance trying to do it might have led to missing the landing altogether and everybody aboard being killed.
Good question. Maybe they were too busy flying the airplane and could divert attention to dumping the fuel? Or maybe they need hydraulics to dump the fuel?
https://www.npr.org/2019/08/26/754458583/al-haynes-pilot-fro...
This talk by him is well worth reading:
http://clear-prop.org/aviation/haynes.html
(The talk used to be available on YouTube, but was removed sometime in the last year by copyright claim. I can't find an alternate source for it. The video was at https://www.youtube.com/watch?v=mZYHMMMeHic)
Denny Fitch passed away in 2012:
https://www.npr.org/sections/thetwo-way/2012/05/10/152402632...
Fitch was interviewed by Errol Morris in an episode of "First Person" called "Leaving the Earth." It's still on YouTube:
https://www.youtube.com/watch?v=o8vdkTz0zqI
Wikipedia page on the crash:
https://en.wikipedia.org/wiki/United_Airlines_Flight_232