NIST Handbook of Engineering Statistics (2012)

I talked with someone in food manufacturing who mentioned the NIST guidelines for "did you short fill to your stated quantity" are more lax if you have variability in the size of the pieces. With well calibrated measurement devices, you could slightly short fill and but be totally legal. I'm looking for the specific formula they use to determine this, but my Google-fu is a bit lacking on this one.

I talked with another person who was at a meat packing plant. At the end of the manufacturing line for ground beef, there was a little machine which would extrude a tiny string of ground beef. The weight was precisely measured, a prescribed amount was added to get it to its required package weight, then it was packaged. He said the plant ran on ~3% profit margin, so they went quite far down the praeto of cost saving ideas.

I also know of an intern at Proctor & Gamble who noticed an overfilling of mouthwash. With the volumes they ran, they had annual savings of hundreds of thousands of dollars by better calibrating things.

PDF link: https://www.itl.nist.gov/div898/handbook/toolaids/pff/E-Hand...

One of the OG knowledge websites for me. What a great resource.

Does anyone know of any applications of statistical techniques, like the ones in the handbook, to software engineering in general or testing in particular?