Search the Elsmar Cove!
**Search ALL of** with DuckDuckGo including content not in the forum - Search results with No ads.

What to do when 100 % verification is not 100 %

First, I am sorry if this comes across as confused - I am trying to figure this out.

We have an automatic measurement assembled machine, in which a robot arm moves an item along a series of steps. After turning, the item is moved to a steam cleaner, then it goes to one measuring device, then another measuring device. Then the computer decides if everything is okay: accept or discard.

So, this production of an item has 100 % measurement, with no sampling needed. My question is: what happens if any measurements is lost due to some incident. Let's say, as an example, that there is only data present to support that 90 % of the items has been measured and accepted. Is there any way to infer that the rest of the items are accepted?

The obvious option is to measure the whole lot again - but is there an alternative? I guess I can choose to do sampling, but then it would not be 100 % verification any more. But it would be a large sample, and it would demand that the measurement loss was not in a large chunk during the production cycle.

I hope I have been able to make myself understandable.

Thanks for reading!

Paid Advertisement - Forum Supporter

Johnny Quality

Starting to get Involved
Nicholas, is this 100% measurement a customer requirement? A requirement for a 3rd party?

Does your data suggest the 100% measurement is redundant? What is the consequence if a part that does not meet the tolerance falls through the chain? Will it be detected later?

Ed Panek

VP QA RA Small Med Dev Company FDA and ISO13485:16
If I were reviewing post production DHR and I was missing 10% of my QC check data I would need to know why that happened, What the risk is, and then go through a MRB and disposition it.

For ISO 13485
If a deviation occurs in production the requirements state:
8.3 Control of nonconforming product
8.3.1 General
The organization shall ensure that product which does not conform to product requirements is
identified and controlled to prevent its unintended use or delivery. The organization shall document
a procedure to define the controls and related responsibilities and authorities for the identification,
documentation, segregation, evaluation and disposition of nonconforming product.
The evaluation of nonconformity shall include a determination of the need for an investigation and
notification of any external party responsible for the nonconformity.
Records of the nature of the nonconformities and any subsequent action taken, including the evaluation,
any investigation and the rationale for decisions shall be maintained (see 4.2.5)


Super Moderator
Either change the requirement or change the process. 100% is 100% (each, every, all). 100% isn't 100% with the exception of unless that exception is well defined, understood and accepted.

Ronen E

Problem Solver
Staff member
Super Moderator
Flag as non-conforming and run through MRB per your established SOP. Any concession should be appropriately rationalised, documented etc.

John Predmore

Quite Involved in Discussions
You say the measurement is lost, but does the sequence in the automatic inspection station change based on the result of the inspection?

If the automatic station takes an action based on the measurement - such as marking a defective part "reject" or placing the defective part in a reject bin, then you have indirectly the pass/fail results of the inspection on individual parts. By the way, a best practice is to also mark (in an inconspicuous way, such as a prick punch) an accepted part "pass" at the end of the inspection cycle. That is your assurance that no part somehow bypassed the inspection procedure. In this way, if any parts in your batch are suspect, you could hand sort for the inspection mark and that might be faster than remeasuring the entire lot.
Top Bottom