Boeing’s answer to the 737 Max problems: more automation?

In all the hype about the problems surrounding Boeing’s 737 Max airliner, particularly the two deadly crashes in Ethiopia and Indonesia, I couldn’t help noticing one thing.  Airlines and pilots in First World countries appear to have had few similar problems with the aircraft.  It’s those in Third World countries that did – and not all of them, either.  The Lion Air 737 Max that crashed had experienced control problems just the day before the accident – but a third pilot on board, who knew what he was doing, told the flight crew what to do (as was pointed out in the aircraft manual), and the problem was resolved quickly and easily.  The cockpit crew on the fatal flight appear to have lacked that level of knowledge and/or training and/or understanding.

This brings up a very important point, one that the Wall Street Journal recently highlighted.

Boeing Co. is increasingly committed to transferring more control of aircraft from pilots to computers after two crashes exposed flaws in an automated system on its 737 MAX that overpowered aviators in the disasters.

. . .

… such changes also seek to address the fact that average pilots may not react to problems—including those tied to automation—as quickly or proficiently as designers traditionally assumed, according to former and current Boeing officials and industry executives. The view took hold after a flight-control system known as MCAS put two MAX jets into fatal nosedives within the past 14 months that together killed 346 people.

“We are going to have to ultimately almost—almost—make these planes fly on their own,” then Boeing Chairman Dave Calhoun said in a CNBC interview in November, roughly six weeks before relinquishing that job to become the Chicago plane maker’s CEO.

. . .

Executives at Boeing and Airbus have said they are also designing flight-control systems tailored for younger pilots, who generally have less flying time in their logbooks—and a more innate familiarity with technology—than aviators of years past.

. . .

Boeing also plans to tailor its design and training to better serve the more globally diverse group of pilots now flying its planes, said former and current company officials familiar with the plans.

. . .

Some of the new systems Boeing and other companies are working on are designed to maintain stable flight while pilots troubleshoot in moments—like those during both MAX crashes—when crews face cascades of emergency alerts and warnings that can be confusing or contradictory.

There’s more at the link.

There are three aspects to this issue that the WSJ article (and Boeing’s comments) side-stepped, probably for fear that politically correct critics would pillory it for being “insensitive” or “elitist”.  However, I don’t give a damn about politically correct critics – so here goes.

First is the language issue.  English is, by regulation, the international standard language in air traffic control, and is also the language in which most aircraft instruments (particularly those made in the Western world) are labeled and operated.  Candidates for commercial pilot certification in the USA have to have a working knowledge of the English language in order to take their examinations, and this is supposed to be verified during the check ride.  However, that is not always the case overseas, particularly in the Third World.  If a non-English-proficient crew is overwhelmed by warning messages in a language they don’t properly understand, then problems may develop too quickly for them to respond appropriately.  I think this hasn’t been mentioned at all in the mainstream media – but it should be.  Having flown in many countries in the Third World, I can assure you from personal experience that the problem is very real.  Discussions with those who sign up pilots from other countries for training in the USA will reveal just how widespread the problem really is.

Second, there’s the cultural mindset.  Technology is downright challenging at times.  Even First World pilots sometimes battle to get through all the prompts, warnings, alerts and messages being thrown at them in the cockpits of airliners.  That’s why simulator training is so important.  It can reproduce problems that are almost impossible to overcome, and repeat them again and again, so that pilots learn the hard way how to deal with them.  (Simulators also show how impossible some problems really are.  Take, for example, the crash of United Airlines Flight 232 at Sioux City, Iowa in 1989.  Astonishingly, the cockpit crew was able to land the plane, although it wrecked on touchdown, and saved the lives of many people on board.  Post-crash re-enactments in simulators were uniformly catastrophic, resulting in a crash with the loss of all lives aboard.  They not only helped to explain what happened, but also helped in the design of modifications to similar aircraft in an attempt to prevent any recurrence.)

If you take people who do not grow up with an intimate familiarity with technology, they often find it hard to master it.  Many take refuge in learning-by-rote, checklists, and “heads-down” flying – looking at their instruments, rather than out of the window at what’s going on around them.  The crash of Asiana Airlines Flight 214 is a good example.  A simple glance out of the cockpit windows would have shown the aircraft to be far too low – but none of the flight crew appear to have done so.  Instead, concentrating on their instruments, heads down in the cockpit, they essentially flew into the ground.

How many of the Boeing 737 Max’s problems might be blamed on a similar technological overload of its flight crew?  Again I point out that First World crews appear to have either had no similar problems to Third World aircrew flying that plane, or they were able to deal with them as a matter of routine.  I can’t help but think that a basic comfort level with technology had a lot to do with that.  (One example:  why on earth did the Ethiopian 737 Max remain at full takeoff power for the entire time it was in the air?  That should have been throttled back soon after lift-off, but it was not.  Was the crew simply too overwhelmed by other issues to think about so basic a requirement?  It should have been drummed into them during their basic training, never mind advanced commercial pilot status!  As it happened, it simply made the situation worse, bringing on disaster even faster.)

Finally, there’s an issue of culture.  Status, prestige and rank are vastly more important in Third World countries than they are in, say, the USA.  A senior or command pilot expects – and usually gets – instant obedience from his co-pilot, rather than informed input and (if necessary) disagreement.  If the command pilot makes a mistake, or isn’t coping with an issue, a US pilot will likely at least say something, if not intervene to help correct the problem.  Not so much in the Third World.  There, the boss is the boss is the boss – and if he’s doing something wrong, he’s still the boss!  “Talking back” will likely get his assistant reprimanded or fired, rather than commended.  Again, I can’t help wonder how much this problem was a contributing factor to the two fatal 737 Max crashes.  I have no way of knowing . . . but having been a passenger on many Third World aircraft (including Ethiopian Airlines – never again!), I have my suspicions.

The problem with Boeing’s proposed solution is that automation can be as hazardous as anything else.  Just look at the problems reported with the Tesla “autopilot” (not to mention the deaths allegedly attributed to it).  I know that DARPA is currently engaged in developing a “robotic co-pilot” for military aircraft.  I presume that Boeing envisages similar technology for its passenger airliners.  I just hope it works better than Tesla’s iteration!

Peter

19 comments

  1. While this is obvious to anyone who has watched third-world oopie-doopies try to imitate proficient Western professionals, it will be buried by political correctness and these crashes blamed on failed technology.

  2. You also have the third world problem of status being everything. How does one get to be a pilot? Not by competence, but by who's relative you are. Ties are resolved by bribery. Ability to do the job figures nowhere in the process.

  3. I have a good friend that's a commercial pilot. I asked him what he thought about the crashes after they happened. He shrugged his shoulders and said "Not enough SIM time. There's nothing wrong with the plane. Many third world airlines pilots don't spend enough time in the simulator"

    The problem isn't technical (although there were design flaws, mostly with redundancy. The real issue was marketing. They wanted to sell a plane under the guise of not having to train your pilots – "It's the same as any 737!". MCAS is a software solution to make the plane feel like an older model.

  4. The MCAS automation is what killed all those people. The automation was fatally flawed. The answer to this is not "we'll give you more automation".

  5. Automation is only as good as the people who wrote it and those who tested it, hopefully NOT the same people. Boeing replaced their experienced IT staff with $12.50/hour staff from India (NCL, I believe) in exchange for a deal with an Indian airline.

    In my 40 years in IT, I have worked with Indian IT staff extensively. When they are good, they are very, very good but, and it's a big but, many from todays colleges are incapable of thinking problems through. For example, if you ask them how many values a light switch can have, the common answer is 'Two. On and off'. The correct answer in IT is 'Three. On, Off and something else'. Failing to account for that 'something else' is what leads to these kind of problems.

  6. The blancoliro youtube channel has a long series of videos that covers the facts that are available.
    Author is a 777 pilot who was a 737 pilot for many years, along with lots of other experience including Air Force instructor
    and Airframe and Powerplant Mechanic certification and experience. You can't find a better, more comprehensive
    source on the subject.

    My opinion after watching all those is that the crashes were caused by the poor system design but were preventable by
    pilots with adequate training (as someone else mentioned re: sim time).

    The move to more automation by Boeing would leave them far short of the level that Airbus has already reached, where pilots
    don't so much fly the plane, rather they issue instructions to the computer which then flies the plane. This has contributed to
    the practice adopted by airlines in less wealthy parts of the world to place passenger jets in the hands of pilots that have very
    little experience. When you are flying an Airbus that often does not matter because the pilot cannot override the computer.
    When flying a Boeing it often does matter because the pilot needs to remain in charge and disable the automation when it is
    suboptimal.

    However, the MCAS system had a built in crash mode, and it is not easy for a pilot to recognize when it has happened and know
    what to do to disable it. So yes, better training matters, but the planes crashed themselves, it wasn't the pilots that caused it.

  7. Boeing certified the airplane as just another 737 with minimum training needed to save development time and cost, but the new bigger engines changed the characteristics.

    Outsourcing the real time software development.

    Changing the culture from one focused on safety to one of cost.

    Moving the headquarters to Chicago.

    Depending on 1 sensor with no redundancy.

    Making safety features an upgrade.

    And Boeing is paying the price for their short sightedness. 737 fiasco is only one of several recent fiascos.

  8. Language is also a major issue. 'Technical competence' in English is NOT the same as actually understanding English. Having flown all over the world, there are many places where if you deviate even slightly from the 'script', you end up with dead air, or the controller or pilot questioning what was said. The same is true for warnings and alerts. If you cannot read English, warning and alert cues that scroll across the screen are so much garbage… sigh…

  9. As a fictional example based on real events, I found the movie Sully to be fascinating. Based on the Hudson river set down of a passenger jet by Captain Sullenberg after bird strikes took out both engines.
    In particular, the movie went into detail over the investigation and the attempt to second guess the captain's decision to set the plane down in the river rather than attempt alternate landing strips.
    Spoiler alert, after proper setup and consideration for the time required for problem analysis, simulations proved that the river landing was the only option that did not result in a land crash and probable loss of life.

  10. Nothing will ever be foolproof because fools are too ingenious.

    That should be tattooed on on the back of every engineer and their bosses hands at Boeing.

  11. Today's Boeing is not the Boeing we knew of yore; when the McDonnell-Douglas "merger" went through, it was the McDonnell-Douglas "suits" who wound up running the company. Being from the Northwest, I knew a bunch of old-school Boeing types who were appalled at the way the new McDonnell-Douglas corporate culture took over and started weaseling its way into every nook and cranny of the old Boeing. That process has been at the root of all of Boeing's problems, from the bribery scandal on down to this one.

    Boeing is dead. The old company exists only in the memories of the retirees, and popular imagination. What exists now is a reanimated zombie of McDonnell-Douglas, with all of its old vices perpetuated forward under a new name.

    McDonnell-Douglas pulled off a corporate Cordyceps takeover, and we're just now seeing where that is taking the company–Which isn't anywhere good. Boeing is dead, and all that's remaining is the husk filled with Cordyceps spores.

  12. @Ray, the most horrifying quotes from the Boeing internal docs (I read on the BBC) were one Boeing guy asking another if he'd let his family fly on a MAX. The replay: "No"

  13. I love how "the more globally diverse group of pilots now flying its planes" has become the plausibly deniable, politically correct way of saying "the more unintelligent group of third-worlders now flying its planes".

  14. A problem that is alluded to but not quite directly stated is that of proficiency in hand flying, which is a problem even among 1st world pilots. I completely agree with everything Peter has said, and what McChuck has said, especially having watched 3rd world pilots in their training. I was a missionary pilot in Africa for 10 years, a flight instructor, and am currently assistant chief pilot at a large aerial work company. All that to say I have direct experience in this subject, though not direct experience with the 737 other than as a passenger.

    The reliance on automation robs a pilot of hand flying skills, necessary to overcome situations in which the automation proves either sub-optimal or actively hazardous. Pilots first of all want to defer to the automation, partly because it's normal to use it. Also because it's easier. (Sorry other pilots but we tend to be a lazy bunch). Flying is, like shooting, a perishable skill. It comes back quickly with training, but not quickly enough in an emergency. And if it was never developed in the first place, such as is often the case in the 3rd world,… well.

    You might checkout code7700.com for further reading on the subject. Mr Albright is a very good author and has lots, LOTS of information on the subject. And lots of experience to back up what he says, and has done lots of research to back up the experience. His article on the Asiana 214 crash is quite good and, I think, has bearing on the 737max situation. You can find it at http://code7700.com/case_study_asiana_airlines_214.htm

  15. Boeing is toast. If the MAX situation doesn't kill it, their deal with China will certainly finish them off. Those McD-Douglas office manager types in Chicago, being only paper shufflers without any technical background, have Boeing assisting China in learning how to build airliners and set up their own production facilities, supposedly for internal China use.

    Those idiots have no idea that that knowledge is a large part of what makes Boeing have any business value. Once China gets up to speed, they can kiss goodbye any future world sales of aircraft. China will be able to undersell Boeing EVERYWHERE. Fools to the max.

    Those guys are so stupid, they didn't do any studies of previous Chinese ventures that other companies have attempted. The fact is, that if you have them build a product, you have then GIVEN them your future business prospects.

  16. This is why the Navy and AF have simulators to teach missile crew members what to do, and why they must do it, and how easy for things to go bad on them.

Leave a comment

Your email address will not be published. Required fields are marked *