4 ms·

uh no. I'm a food engineer for one of the largest manufacturers in America and there isn't any pressure from corporate to just pass lines that might be dangerou

uh no. I'm a food engineer for one of the largest manufacturers in America and there isn't any pressure from corporate to just pass lines that might be dangerous. the problem is in testing and just how statistically unfeasible it is. case in point, we had a line that passed every QA test but we would still get back 1 product every now and then from a customer with spoilage. we'd test thousands of samples each year and they're all fine but still something was the problem. on top of that, we had "nearly identical" lines that never had any problem. luckily we found the problem in the line and how to push the failures alway the way down to zero for little cost instead of corporate wanting to pay $10 million to fix it. so no, corporate isn't some boogy man.

yourapostasy 4y ago
SRE folks really should be speaking with engineers like you. I'd love to spend a few months crawling in and around the QA problems your industry faces in production, and how your industry tackles it. There are many echoes of only-happens-at-scale and statistical quality control that software folks think are the first time anyone has ever solved, but are really process engineering problems old hands in industries like yours have long since codified into practices. This is part of the reason I like going on factory tours and see "How It's Made". The US lost a hell of a lot more than the ability to make goods when we shipped off our manufacturing infrastructure. We lost the capacity to cross-fertilize between industries and create compounding innovation effects. If you can share as many interesting stories of production line QA as hits your whimsy in as much detail as you care to share, it would be greatly appreciated.