iSixSigma

Acceptance Sampling in Quality Audits

Six Sigma – iSixSigma › Forums › General Forums › Tools & Templates › Acceptance Sampling in Quality Audits

This topic contains 3 replies, has 2 voices, and was last updated by  Strayer 10 months, 1 week ago.

Viewing 4 posts - 1 through 4 (of 4 total)
  • Author
    Posts
  • June 14, 2018 at 10:14 am #56021

    Andre

    I’ve been searching for a good sample size for a quality audit, where unfortunately, the team doesn’t know what to exactly look for. The audit consists of looking at unused case files to ensure the staff as a whole is properly filling out information correctly. I assume if there is any error in the file, it is rejected. They honestly don’t know what to look for and I don’t think they are prioritizing any specific question/answer in these files to be more critical than others.

    I am currently looking at Acceptance Sampling and am wondering if this is the best technique for this audit. Normally, sample sizes depend on a confidence interval, but I don’t know what a CI would be given a pass/fail system.

    Would something like this be appropriate? http://garmentstech.com/full-aql-acceptable-quality-level-inspection-level-2-5/

    Concerning the number of cases, we have an mean of 1650, median of 1566 and a population standard deviation of 198.94

    0
    June 15, 2018 at 6:35 am #202674

    Mike Carnell
    ★★★★★★★★★★
    Participant

    @arrr127 If you are interested in understanding acceptance sampling you need to understand OC curves. Get a book by Grant and Levenworth it will help.

    0
    June 18, 2018 at 9:29 pm #202693

    Amanda

    You can look into using the QSIT sampling plan, and choose your specific sample size based on risk.

    https://www.fda.gov/ICECI/Inspections/InspectionGuides/ucm170830.htm

    0
    June 18, 2018 at 10:52 pm #202694

    Strayer
    Participant

    You cannot do a quality audit if the acceptance criteria are unknown. You say they don’t know what to look for. If you know what they should look for, they need to be taught. If you don’t know, it’s the blind leading the blind. You should be concerned about the underlying problem rather than sample size for auditing the symptoms.

    0
  • Author
    Posts
Viewing 4 posts - 1 through 4 (of 4 total)

You must be logged in to reply to this topic.