4 ms·

Worse, that type of thing is long been known as an dangerous issue and the industry has processes and protocols to deal with it. That's why it's "worse", Boeing

Worse, that type of thing is long been known as an dangerous issue and the industry has processes and protocols to deal with it. That's why it's "worse", Boeing knows about all these things and conscientiously decides to don't give a f#$%k. Airbus ( and I'm sure pretty much all the other manufacturers ) has all the processes for training the workers about this problem, how to avoid it, how to deal with it when happens, check-in and checkout all the tools at the start and the end of the task, inspections, etc. this is so basic it hurts.

kayfox 7y ago
I think its probably a lack of supervision and not a lack of protocols, if it were a lack of protocols the 787s assembled in Everett would have the same issues. Also, I have personally observed the use of protocols to prevent this in tours and (a long time ago) working in the Everett factory.
- atoav 7y ago
  In the daily podcast episode that QC employee explicitly tells a story of how managers tried to hush him, change the protocols, ignore issues etc. That same guy also said it had to do with the location of that new factory and the lack of company culture paired with managers who just had the numbers in mind.
- LeifCarrotson 7y ago
  The tricky thing about protocols is that you have to be willing to follow them and deal with the consequences when you step off the normal path through the flowchart. If the protocol says "(1) Check out tools before using them. (2) Check in tools when returning them. (3) Verify all tools checked out have been checked in before marking the job complete." that's one thing. Sending employees to training on how to do this and seeing daily clipboards by the tool room with initials might make you think the protocol is in use. And if there are supervisors making sure that the clipboard is full of initials at the end of the day, that can give you a nice warm fuzzy feeling. The problem comes when supervisors want to 'look good' by having no deviations from the protocol. If people are post-dating their check-in or pre-dating their check-out, or checking stuff in that's still in use by a coworker, the protocol is less than useless. When humans are tasked with following those protocols, they'll make human mistakes sometimes. If there are no mistakes that's not good, that's bad - it indicates the metrics are being fudged. I think any supervisor that fails to find a deviation from the protocol or who can't list an event when the protocol required action outside the norm does not look good. Similar stuff applies to software development: If you're doing code reviews, or testing before check-in, or signing off on releases, and can't point to a time when the protocol said "nope, go back to the drawing board" you're not actually making your development safer you're making it a hassle.
  - kayfox 7y ago
    The tool crib checkout is done on computer at least. The toolbox inspections might be fudged. It also might be downstream or upstream of the Boeing factories, like at one of the companies that manufactures major components (example: Spirit AeroSystems manufactures the 737 main fuselage and I'm told Section 41/cockpit of many aircraft) or at one of the maintenance facilities.
- abakker 7y ago
  I wonder if there needs to be some personal liability here, too. Many individual bankers have been banned from working in the industry for willfully ignoring the laws/rules. I wonder if that is necessary in this industry, too?
  - kayfox 7y ago
    I have heard from union members about union pushback on such things, but that may be anti-union sentiment or rumors.
atoav 7y ago
Which is why this raises many red flags: it is indicative of a company culture that sooner or later will murder innocent people.
- frandroid 7y ago
  Sooner? Two 737-MAX8s have gone down already!
heisenbit 7y ago
Boeing knows how to do this. The problem is manager shifting responsibility from one mature manufacturing site to a less mature one without properly compensating for it. A problem fortunately unknown to SW engineers.
onceUponADime 7y ago
There actually is a process of beeing interrupted while working on a plane, which consists of retreat ( you carefully undo the last step you remember doing before beeing interrupted, checking you have all parts in the location they started out, protected against dust etc.). Then you make a note on the step you expected to do. <Interruption> <Return to work> You start out three step before your last step. You repeat/check those steps. Its time intensive as hell, which is why the usual interruption to egostroke for meddle management does rarely happen in aircraft maintenance and production.

kayfox 7y ago

I think its probably a lack of supervision and not a lack of protocols, if it were a lack of protocols the 787s assembled in Everett would have the same issues. Also, I have personally observed the use of protocols to prevent this in tours and (a long time ago) working in the Everett factory.

atoav 7y ago
In the daily podcast episode that QC employee explicitly tells a story of how managers tried to hush him, change the protocols, ignore issues etc. That same guy also said it had to do with the location of that new factory and the lack of company culture paired with managers who just had the numbers in mind.
LeifCarrotson 7y ago
The tricky thing about protocols is that you have to be willing to follow them and deal with the consequences when you step off the normal path through the flowchart. If the protocol says "(1) Check out tools before using them. (2) Check in tools when returning them. (3) Verify all tools checked out have been checked in before marking the job complete." that's one thing. Sending employees to training on how to do this and seeing daily clipboards by the tool room with initials might make you think the protocol is in use. And if there are supervisors making sure that the clipboard is full of initials at the end of the day, that can give you a nice warm fuzzy feeling. The problem comes when supervisors want to 'look good' by having no deviations from the protocol. If people are post-dating their check-in or pre-dating their check-out, or checking stuff in that's still in use by a coworker, the protocol is less than useless. When humans are tasked with following those protocols, they'll make human mistakes sometimes. If there are no mistakes that's not good, that's bad - it indicates the metrics are being fudged. I think any supervisor that fails to find a deviation from the protocol or who can't list an event when the protocol required action outside the norm does not look good. Similar stuff applies to software development: If you're doing code reviews, or testing before check-in, or signing off on releases, and can't point to a time when the protocol said "nope, go back to the drawing board" you're not actually making your development safer you're making it a hassle.
- kayfox 7y ago
  The tool crib checkout is done on computer at least. The toolbox inspections might be fudged. It also might be downstream or upstream of the Boeing factories, like at one of the companies that manufactures major components (example: Spirit AeroSystems manufactures the 737 main fuselage and I'm told Section 41/cockpit of many aircraft) or at one of the maintenance facilities.
abakker 7y ago
I wonder if there needs to be some personal liability here, too. Many individual bankers have been banned from working in the industry for willfully ignoring the laws/rules. I wonder if that is necessary in this industry, too?
- kayfox 7y ago
  I have heard from union members about union pushback on such things, but that may be anti-union sentiment or rumors.

atoav 7y ago

Which is why this raises many red flags: it is indicative of a company culture that sooner or later will murder innocent people.

frandroid 7y ago
Sooner? Two 737-MAX8s have gone down already!

heisenbit 7y ago

Boeing knows how to do this. The problem is manager shifting responsibility from one mature manufacturing site to a less mature one without properly compensating for it. A problem fortunately unknown to SW engineers.

onceUponADime 7y ago

There actually is a process of beeing interrupted while working on a plane, which consists of retreat ( you carefully undo the last step you remember doing before beeing interrupted, checking you have all parts in the location they started out, protected against dust etc.). Then you make a note on the step you expected to do. <Interruption> <Return to work> You start out three step before your last step. You repeat/check those steps. Its time intensive as hell, which is why the usual interruption to egostroke for meddle management does rarely happen in aircraft maintenance and production.