Sample Size and Probability

Six Sigma – iSixSigma Forums General Forums Tools & Templates Sample Size and Probability

Viewing 2 posts - 1 through 2 (of 2 total)
  • Author
  • #53518


    We are looking to use some internal audits as a means to measure potential claims in our business. Currently the management is using 25 audits across the board for each person. If within a 2 month period the auditors find 5 errors, an associate may lose his/her job. There is no consideration given to the external error rate nor to the amount of work produced in a day…. In some cases 50 lines a day vs. 400 lines a day….

    With the current process, we may be losing an employee whose external error rate is way below industry standard, but with the internal audits, they’re toast…



    This violates the caveat that quality data should never be used to evaluate the performance of individuals or teams. Doing so makes it difficult to collect accurate quality data and turns quality assurance/process improvement/six sigma professionals into enforcers — i.e. to be feared rather than welcomed in the spirit of continual improvement.

    It is much better to collect the data, normalize the error rates, and do causal analysis when variation is outside of specifications. The cause may turn out to be human error. If so, the root cause is often due to process deficiency or poor training, both of which can be corrected without punishing employees. Most people really do want to a good job, as Deming and others have noted. It sounds like your management doesn’t get it.

Viewing 2 posts - 1 through 2 (of 2 total)

You must be logged in to reply to this topic.