Ralph Nader: Boeing has been Mismanaged for Years

Part of the phenominon is technology's accelerating complexity.

(i.e.: there are increasingly more and more details that have to be evaluated as not confounding more and more systems, under more and more scenarios, on a daily basis, that robustness requires an ever-rising bar to be hurdled...and then, triple-checking off-shored software for all of it...)

It seems to me that much of the embedded software failures today come from programmers finding a complicated way to do something simple. When one or two complications are the only things added then the complications dissolve in today's fast processing speed. Once every programmer on a project adds their two complications it then becomes a statistics time bomb waiting to produce a stack overflow somewhere that feeds false data into the system.

Engineering has always had people who design perfect machines and then suffer in silence while the 'value engineers' and accountants strip out what they see as excess/ over-priced materials before passing the design to buyers to get the machine made as cheaply as possible. Having worked as a service manager for many years, (petrol stations) My engineers had to deal daily with the same system faults caused by poor quality circuit boards, etc. It was hopeless trying to explain to our MD and production manager that better quality pumps would increase MTBF which would allow engineers to look after more sites to increase our revenue and would also make our customers much happier with the products.

More recently I had to caution my son who was a buyer for a very large German electronics company because he was so pleased at screwing the best price from small suppliers. "You'll get caught out when it all goes wrong" I warned him and you may even bankrupt that suppler!

Of course, when you are dealing with aircraft systems the rigorous quality controls should eliminate any errors and there has always been a pride in producing such systems. In recent times outside training has not always followed the strict aerospace rules and employees have tended to drift into the critical aerospace environment from other industries and brought with them some of the bad practices we see every day in general industry - especially with software which is still a black art to many hardware engineers. Have you ever sat in a board meeting and tried to argue with a software engineer - ours couldn't even bother to change his T shirt for the meeting!

"Part of the phenominon is technology's accelerating complexity"

As I understand it this was not a technology problem in the first place, it was a mechanical problem that they tried to solve with technology and that's a different thing entirely.

There's nothing wrong with utilising software to make efficiencies, which is the proper use of software but it's not the way to fix a mechanical problem.

I am not an aircraft engineer so the following may be a naive suggestion but it's been stated that the underlying fault is that the engines are physically larger and because of that they were forced to position the engines too far forward in order to achieve ground clearance using the old fuselage, and that repositioning of the engines upset the plane's balance and then they tried to counteract that imbalance with software.

So why not now develop new and taller undercarriage so they can then relocate the engines farther back where they balance the aircraft and consequently get rid of its tendency to stall?

It's not an easy thing to do but then scrapping the whole fleet is what's being proposed and designing a new undercarriage and repositioning the engines so the plane is balanced again seems a whole lot better than software fixes that no one is ever going to trust within the next 40 or 50 years.

I cannot believe that the designers of the 737 Max, used only one computer and only one sensor, to "govern" such important NEW flight characteristics. REALLY Cheapo!

Next error was, the way for a pilot to "handle" the failure was changed from previous versions, and the changes were as good as hidden. Just to allow previous 737 trained pilots to "not need" any training on the new version. Cheapo!

The whole problem of having to move the new larger engines forward and upward, because Boeing were too cheap, years before, to give the original 737 a properly high enough undercarriage/wheel system.

Which is exactly what Airbus did, so that larger diameter motors could be correctly accommodated, without compromising engine safety clearances.

Tests have since shown that no pilot can correct the sensor problems, when simulated on simulators, in a fast enough time to save the aircraft! That is without special training first!

The pilot who ditched an aircraft on a river in New York some years ago, and who saved everyone onboard at the time, tried and could not prevent the simulator crashing! He had been flying 737s for years.....

On earlier versions of the MCAS software, if the pilot "insisted" on pulling the stick back, the software disengaged automatically - a very sensible and intuitive method.

But this was "disabled" in the new Max version!!!

The person who signed off that software change needs to be "hung".

Any other Boeing personnel who were involved, need also to be "hung".

I suspect though that top management were involved as having the final "say", people not fully trained as aircraft engineers. Let us hope that these people will be shortly "hung" out to dry.

If Boeing does not collapse completely, it will be a surprise, as they are STILL building the aircraft at almost the previous rates, in spite of so many people dying in the first two crashes.

They have even filled most of the employees car parking with finished aircraft they are still not allowed to deliver, because the normal storage areas are full to overflowing!

The whole business, from start to finish, smells of an old Germany saying, that I translate loosely to "He saves money, no matter what it costs!"

I had that saying printed out on a poster, on the wall behind my desk, for a great many years when I was still working.

Shame that it was obviously not known at Boeing!

It reminds me a little, of when I was also involved in writing software, I would have a working program within a very short space of time.

But the need to quantify inputs from any user, restricting them only to proper data, informing them how to make only proper inputs, how to understand the outputs and and and, would usually triple the physical size of the program, and slow it down considerable.....But this was the correct way to work - SAFELY!

Even though nobody died if they was a program error!!

I hope there will be a full board of inquiry into the 737 Max disaster which will bring out the sorry tale of technical mismanagement and political/marketing pressure. Behind the scenes the story is probably:-

The aircraft modifications were completed weeks before the software people who were probably struggling to design a modification to the operating system to compensate for the extra power of new engines mounted forward and higher on the wings. In this situation Boeing should have insisted on at least 2 independently operating angle of attack sensors plus a completely fail-safe software system. I understand that it was still possible for the pilot to overcome the MDAS as long as the airspeed was reasonably low, this probably why the 737 Max got though flight testing because the test pilots were ready for it to happen and had time to correct it.

Not the case with the poor guys who lost their lives - was airspeed included in the systems analysis for the MDAS?

You posted:-

I understand that it was still possible for the pilot to overcome the MDAS as long as the airspeed was reasonably low, this probably why the 737 Max got though flight testing because the test pilots were ready for it to happen and had time to correct it.

I am sure that it was done on aircraft with a working sensor and at the correct speeds!

A low airspeed (would be nearer to stall speed) would have made the problems even worse!

Remember, the system was made (I find the word "designed" to be inappropriate!) to prevent stalls/nose too far up, a sensed "low speed" would have resulted even earlier the nose "push down" that killed so many people!!

I have not yet seen a design of the sensor in question online, but if anyone has seen one, I would like a link please.

But totally guessing only, I suspect that it was something with possibly a Pitot tube as well....which have, when blocked for some reason, killed more than their fair share of air travelers, over the years....

And how many of the decision makers at GM have been prosecuted?

Did GM suffer total collapse?

GM at least did it the right way, (if there is a right way for corporations to kill people) Binding non disclosure agreement of sufficient payout to guarantee silence.

It should be very disturbing to somebody else besides me that the immediate patch did not include reverting to the previous proven practice of having an automatic disconnect of the MCAS control software upon the insistance of the pilot pulling back on the throttle?...

(...and, just how many simulations do the bean-counters fly before making their budget cut feedback?...)

I agree entirely.

If MCAS pushes the nose down, but the pilot (for any reason)wishes the nose to be further up, pulls back on stick, MCAS automatically disengages. Maybe posts a warning that MCAS is disengaged.

Remember, the stall alarm is normally a separate sensor and system, so the pilot will receive an audible warning if he pulls the nose up too high anyway....Unless Boeing screwed around with that as well!

Only a "bean counter" would authorize a change in that, which was basically the old system!

I hope that the CEO (where the buck stops) goes down for this....

If Boeing had redesigned the undercarriage to be "higher/taller", allowing the engines to stay where they were placed before, that would have been far cheaper and safer than all this shit with software....

Gees, even Tesla got that right.

If I twitch the wheel or prevent it from turning the autopilot disengages.

If Boeing made cruise control the brake pedal would not disengage the throttle.

in boeing logic the more you push the brake the harder the car should accelerate.

If it was not so very serious, your post would have amused me, but you have got Boeing banged dead to rights!

We must never lose sight of a little humor, many thanks.

The pilot that ditched the plane in the Hudson River, Chesley "Sully" Sullenberger, disagrees with the embedded processing of today's aviation control systems. He wants every system to be capable of a completely manual control override, regardless of how difficult it might be for a pilot to control. (I failed to find his exact quote.) He strongly advocates that pilot training simulators must include these manual control conditions so pilots both learn what it takes to manually control without software assistance and the actual aerodynamics of the craft.

Good post.

I agree with Sully entirely, though I did not know of that wish....I only knew that he fought the simulator for the MCAS problem, and could not prevent a crash, before he got the special training needed.....

He maintained that no human could react fast enough without the special training....on a simulator of course.

Old school but good school?

My belief is someone in project management saw there was a (mostly) functioning system and because they were over budget and late on schedule ordered engineering to release it as is and stop working on it.

For want of a (manual-overide)-nail, a (survival-option)-shoe was lost ... (yet again!)...

Well put!

...and thank you. I greatly appreciate objective recognition each time I receive it, especially since my obscure reference should not have been so un-memorable by others...

Various versions have been around for centuries, but I was specifically refering to Benjamin Franklin's usage in his ''Poor Richard's Almanac'' published in 1758...

...no, I was not there when it was first published, but sometimes, it feels like I was ...(haha...)

GA from me

As usual,it all boils down to $$$.

The Titanic sank quickly because the original plans called for the water tight walls to go all the way to the ceiling on each compartment.

Budget managers finally convinced the engineer to stop the walls short of the ceiling,saying the water will never get that high anyway.

The managers had no experience,except looking at the live well on their one man boats,and noticing that they never overflowed.

The engineer never really like the idea,and was under severe pressure and duress to make the changes.

He was aboard the Titanic when it sank,refusing to take a space on a life boat,and decided to go down with the ship because he could not live with the guilt.

There were not enough life boats aboard the ship because "It is unsinkable anyway, why do we need all those lifeboats $$$$?"

Too bad the Boeing engineers are not required to ride on their planes to prove their ideas work.

"My message going forward is that I want to remind everyone in the aviation industry - especially those who manage aviation companies and those who regulate aviation - that we owe it to our passengers to keep learning how to do it better."

Chesley Sullenberger - Miracle On The Hudson Pilot

I think the board of directors,who had the final responsibility for the plae should be held personally responsible for the failure.

The company can be fined millions,but they are untouched,and go merrily on their way.They have no personal stake in their decisions.

The directors should be tried for negligence and total disregard for public safety.

Real time should be served in a real prison,not a country club pseudo-prison where you can't walk outside of the lines.

That will never happen.Too many personal favors will be called in.Too many greasy palms to make anything stick.

If it ever went to court,it would be dragged out till all of the families' survivors died out or got tired of showing up in court to find that it had been rescheduled.

So sad.So true.

A Roman philosopher once said:"They hang the petty thieves,and elect the great ones to office."

Same goes for other criminals.