I would like to know how many of you use some sampling plan for re-verifying corrected nonconforming product VS. how many of you perform 100% re-verification of corrected nonconforming product. I would also like to know your rationale too.

I have always been used to performing 100% re-verification of corrected nonconforming product. I am not use to a sampling plan for accepting this re-verification.

Thanks for your feedback.

Doug

It depends. Sometimes I find the re-worked parts will be done in batch processes (i.e. 20 pieces machined at once in 5 "loads") and one load after another. A reasonably constructed sampling plan will often do the job.

I would like to know how many of you use some sampling plan for re-verifying corrected nonconforming product VS. how many of you perform 100% re-verification of corrected nonconforming product. I would also like to know your rationale too.

I have always been used to performing 100% re-verification of corrected nonconforming product. I am not use to a sampling plan for accepting this re-verification.

Thanks for your feedback.

DougAlways did 100% verification. Then again, it was required by "regulation".

It depends. Sometimes I find the re-worked parts will be done in batch processes (i.e. 20 pieces machined at once in 5 "loads") and one load after another. A reasonably constructed sampling plan will often do the job.

In the current situation, the sampling plan used for accepting jobs is used for re-verifying Nonconforming Product. This plan accepts certain lot sizes with 1 reject and not-accepts with 2 rejects. Does this seem appropriate to you?

Doug

Hi Doug,

I for one use sample tests/inspections for reworked products. If the product fails this test/inspection, I opt for 100% test/inspection.

surendro sahoo

In the current situation, the sampling plan used for accepting jobs is used for re-verifying Nonconforming Product. This plan accepts certain lot sizes with 1 reject and not-accepts with 2 rejects. Does this seem appropriate to you?

Doug

Doug,
I would suggest using a tightend sampling plan with Zero acceptance of any rejects. If fails then go to 100% f or that lot.

I would like to know how many of you use some sampling plan for re-verifying corrected nonconforming product VS. how many of you perform 100% re-verification of corrected nonconforming product. I would also like to know your rationale too.

I have always been used to performing 100% re-verification of corrected nonconforming product. I am not use to a sampling plan for accepting this re-verification.

Thanks for your feedback.

Doug
There's a lot about conventional wisdom in implementation of sampling plans that doesn't make sense to me. Inspection is all about probability, and using probability calculations to make predictions. Decisions should be based on the level of risk inherent in any given plan, and that decision must always be based on one's personal knowledge of all of the different factors that contribute to risk. One thing I don't understand is why it's considered prudent (aside from regulatory requirements) to do 100% inspection of a lot that's undergone rework, and ostensibly twice the scrutiny given to the original lot. The other thing I don't understand is why, when sampling inspection results in discovery of defectives, sampling is arbitrarily tightened for a certain number of subsequent lots. The operation was a success--the "normal" sampling plan was proven able to detect defectives at the accepted level of risk--so why change it?
The bottom line here is that the level of scrutiny given to reworked product (again, assuming that regulatory or customer requirements are not a factor) should be driven by the level of risk, just as the selection of the original sampling strategy was (or should have been).