Thursday Throughput: Boeing Edition
[ThTh1] One of the big issues to spring up this month has been the problems at Boeing. A door blew out of an aircraft and the landing gear fell off another one. Various grifters have rushed to blame this on “DEI” policies at Boeing. But, as the Real Engineering video below shows, the problems are more conventional: corporate greed, regulatory capture and a culture of cutting corners.
It’s awful to see one of America’s industrial titans suddenly turn into a pumpkin like this. A few years ago, they paid their CEO $60 million to quit. I think they need to fire the entire board.
[ThTh2] One useful thing coming from the AI chatbots. It may give us some insight into how humans construct speech.
[ThTh3] The malaria vaccine is now showing stunning effectiveness over several years. And progress is being made on a universal coronavirus vaccine. And the HPV vaccines is basically eradicating cervical cancer. And if that weren’t enough, a vaccine is showing incredible results for treating colorectal and pancreatic cancers.
[ThTh4] Why do we like science fiction? It turns out, it may make our brains happy to imagine the future.
[ThTh5] A new paper claims that the use of hydroxychloroquine in the pandemic cost 17,000 lives. Orac is dubious and so am I. Their methodology does not seem particularly robust. And even if it were true, the implications people are drawing from it are spurious. The paper examines the first few months of the pandemic when HCQ was being used as a standard protocol in hospitals because it had shown promise dealing with SARS Classic. It was only after massive double-blind studies showed it didn’t work that we abandoned HCQ and it became the refuge of COVID denialists.
[ThTh6] Speaking of deniers, the climate deniers are claiming a new report, not a refereed paper, shows that global warming is a hoax. It doesn’t. What it shows is that some climate models have overpredicted the amount of warming by a small amount. While this is true, the overprediction is small and you can contrast this against the predictions of the climate denialists for flat or falling temperatures that were completely 100% wrong. One plot in particular is galling, showing that the warming is below all the model predictions. But their comparison is to twelve Midwest states. Global warming is — and I don’t mean to be pedantic or anything — global. If you compare global temperature trends to the global models, instead of cherry-picking specific regions, you see that the trend is toward the upper range of the models, while still below the worst ones.
Meanwhile, back in reality, Greenland is melting faster than we thought.
[ThTh6] I can’t think of a joke that would be appropriate for a family website about studies showing that more frequent ejaculation corresponding to better prostate and testicular health. So think of a good one in your head.
Oh, wait.
[ThTh7] During last year’s brief room-temperature superconductor hope, I said that while the result may turn out to be false, it could lead to new science. Well, some new promising science.
[ThTh8] I think the main use of AI for scientific research will be to enhance human investigations. A new study claims AI-enhanced research is finding new antibiotics that could prevent the coming crisis of antibiotic-resistant bacteria.
[ThTh9] So what has JWST been up to? Everything from spotting aurorae on brown dwarfs to probing the depths of star clusters.
What I find amazing about the new and improved superconductor is that, in theory, it works at temperatures that humans are pretty good at creating. Like in your freezer in your fridge in your kitchen.
Sure, humans couldn’t work in that environment for a long time… but they can work in that environment *AT ALL*.
This is so very exciting and I hope that *SOMETHING* comes of it.
Even if it is an interesting third thing instead of a mega-interesting “room temperature” superconductor.Report
Global Warming? ech. An awful lot of Being Soothingly Mean And Comfortingly Angry was excused by the Truly Dire Nature Of The Situation, and if it turns out to Not Be Quite As Dire As Previously Suggested, well…
Oh, don’t I take it seriously? Sure I do! I think we should be building thousands of nuclear power plants about it! We aren’t! And the people saying we shouldn’t are the same people who tell me they don’t think I’m taking the matter seriously. So I guess that I see “seriously” differently from everyone else, where I think “seriously” means “we should do the thing that solves this problem most quickly” and everyone else thinks “seriously” means “we have time for me to argue down everybody that doesn’t like my preferred solution”.Report
The Boeing Door: Based on my experience with this sort of thing, I figure that it’s probably “Step 94: install lock bar and torque bolts to hand-tight. Step 21741: Remove lock bar bolts, apply threadlock adhesive, torque to 5 ft-lb” and they were so busy on the line that they forgot to do Step 21741, and when it came time to close out the documentation they noticed that Step 21741 hadn’t been signed off by the Quality inspector, and said ‘well, there’s no way we’d forget to torque critical bolts like that, go ahead and ship it”.
So, y’know. Not not a cost-saving move, because they’d have to pay four guys to rip apart large portions of the interior fittings to get at those bolts and check to see if they were torqued and glued down. But neither something you’d consider the result of A Poor Engineering Culture or Cost-Cutting Measures or Corrupt Incompetent Management. Like, yeah, it’s easy to say “oh well they should have done it anyway“, and, yeah they should have! But it’s also not like there’s no reasonable incentive to not go back and check.Report
Agreed. I had a convo with one of my Program Managers about this. He informed me that he used to be a materials engineer working for Boeing. 9/10 times when something like this happened, it was an assembly guy using the wrong bolt, lock washer, etc. DEI didn’t even come up.Report
The way he talks about the Boeing door sounds like a classic cascade. One wrong decision along the way cascades down through multiple bad decisions until you end up with a door plug flying out. The manufacturer didn’t seal the door while shipping, Boeing didn’t seal it down while testing, Alaska ignored warning signs. And pow.
The engine failures are a bit different since that seems like a preventable error made because Boeing didn’t want to redo the landing gear.Report
Humans aren’t very good at working through long lists of things in order and doing exactly what they say. Foxconn is using robots for assembly in China now, not because they’re cheaper than the humans but because the robots are really good at that kind of repetitive thing.Report