Which released-bracket measurement belongs in our fixture estimate?

Only for three tagged brackets. The inspector kept those with separate first readings because they were queried. The rest have final readings only.

Those three were selected for trouble, so they won't give you a correction rate.

They can still show what the correction involves, Sam. I've seen a disputed sample get dismissed as unrepresentative and then the next estimate quietly omit the operation again. Keep it for the work description, not the frequency.

Agreed, Jo. Useful examples, rotten basis for a percentage.

The operation is now on the route. Its frequency is still unknown. We are planning a current batch with separate released and post-correction checks tied to each bracket.

Which drawing revision will that batch use? The old sheet is no help if the requirement changed.

And keep the fit-up observations with it, since the fixture designer will need the condition before welding.

Who writes down the correction work itself? An operator should not have to reconstruct it from the inspector's final number afterwards. Even a clear indication that correction happened would beat the current sheet.

Ella, drawing revision F. Sofia, fit-up stays linked to the bracket identity. Jo, the operator records correction on the route and inspection adds the subsequent measurement. We have agreed that split for the trial.

Will the corrected and uncorrected brackets stay distinguishable on the bench too?

Good point, Ada. A tidy sheet won't rescue mixed samples.

Yes, individual tags and separate marked positions while inspection is outstanding. No trial results yet. The estimator has stopped asking me to turn the three queried brackets into a production percentage.

Please post the eventual split. Especially whether the ordinary brackets need the same correction as those three.