iSixSigma

Averaging an Average Range

Six Sigma – iSixSigma Forums General Forums Tools & Templates Averaging an Average Range

Viewing 6 posts - 1 through 6 (of 6 total)
  • Author
    Posts
  • #55738

    Tony Groves
    Guest

    Hi there,

    We have been monitoring the performance of technical drawings when issued to manufacture. To do this we have looked at the number of drawings processed and compared to number of complaints to understand a Yield.

    This was being done weekly, and therefore understanding the average yield overall was quiet easy. However as time goes by we noticed dips in the trend line from week to week, reviewing the data suggested we should be calculating the yield as an average over two weeks due to time spent on ordering material meant the complaints could have took over a week to come through.

    We have applied this change and can now measure fortnightly, and still report the results weekly, however calculating the average overall yield gives me a different result from the previous weekly method. Would we not need to use the mean Xbar now for the overall average?

    Apologies for the war and peace, but i hope the context can help explain the question, any advice on averaging an average range would be most useful.

    Thanks!

    0
    #201454

    Chris Seider
    Participant

    don’t forget averaging of averages isn’t technically correct unless the sample size is always the same for each average.

    0
    #201455

    MBBinWI
    Participant

    @tgroves-EXTA – so, what you’re dealing with is a system where you want to report weekly, but the system performance is on a longer time period, perhaps 2 wks in periodicity?

    0
    #201458

    Tony Groves
    Guest

    Hi,

    Yes, the weekly yield is worked out by number of drawings processed, compared to number of complaints within that given week.

    To show a more stable trend on our yield performance, we now include the previous weeks drawings processed and complaints and take an average over the last two weeks. We then report that result out every week.

    The overall average yield would have been based on the sample of weekly yields, whereas now those weekly yields are technically an average, and using the same methods for an overall average yield, would mean we are averaging an average which doesn’t seem right?

    I hope that makes sense!

    0
    #201477

    MBBinWI
    Participant

    @tgroves-EXTA – what you seem to be describing is a rolling average reported weekly of data from the past two weeks. This is a reasonable method of damping out noise/variation that occurs at a periodicity larger than the reporting period. It is not an average of an average. Your approach is well within acceptable practice.

    0
    #201480

    Tony Groves
    Guest

    Yes it would be a rolling average as time progresses. Sweet, sometimes working with data can make you go go number blind!

    Thanks for all your help, im new to this forum but have already spotted a few posts i might be able to offer some advice on!

    thanks again!

    0
Viewing 6 posts - 1 through 6 (of 6 total)

You must be logged in to reply to this topic.