ISO 2859-2 LQL sampling plan linked to %confidence/%reliability

LarsZ

Registered
Although this is my first post to this forum I have been reading this treasure trove of high quality information for a while now and want to express my gratitude to this community. Thank you for all your guidance and advice!

Then, onto my situation:
- we produce small LOT's (50 to 100 a batch) of disposable, sterile class IIa medical devices
- we have implemented a destructive sample check to verify the quality of these products after manufacturing & sterilisation
- we had implemented a ISO 2859-1 AQL based sample plan (so a inspection by attributes and AQL levels based on Risk Management) BUT we got feedback from our notified body that for design verification and validation a ISO 2859-2 LQL sampling plan linked to %Confidence/%reliability levels coming from Risk Management.

My question being:
From what I have understood the required sample size for a attributes inspection with c=0 (all samples should pass the test) can be calculated as:
n=ln(1-confidence)/ln(reliability)
(e.g. 95% confidence /90% reliability would result in n=29)
However: how as n is not linked to a LOT size how can one link this to LQL levels?
 

v9991

Trusted Information Resource
i was further reviewing certain articles...

ref. slide 20 :- Acceptance Quality Limits and Visual Inspection
• Understand the sample size is unrelated to population size (in most situations).


The Effect of Lot Size
The "discrimination" or protection provided by a sampling plan depends primarily on the number of units inspected and the acceptance number. Lot size has only a minor effect limited to the case when 10% or more of the lot is inspected.

Since Type-B OC curves represent worse case, sampling plans selected based on Type-B OC curves can be used to inspect any lot regardless of size. When selecting statistically valid sampling plans, it is not necessary to use different sampling plans for different lot sizes. A better strategy is to select one sampling plan based on the protection it provides, i.e., its OC curve.


so the emphasis is more on thinking through on selection of sampling plan, vs the lot size.:agree:
 
Top Bottom