iSixSigma

data quality average sigma

Six Sigma – iSixSigma Forums Old Forums General data quality average sigma

Viewing 7 posts - 1 through 7 (of 7 total)
  • Author
    Posts
  • #28442

    Payne
    Participant

    From what I’ve read the average sigma for most companies is around 3sigma. However, I was wondering if the average was different when measuring quality of manually entered data? The team I am working with is trying to project a realistic goal sigma for this process. Does anyone know where I might find a case study or article related to data collection processes? Thanks!

    0
    #70885

    Jim Johnson
    Participant

    Kelly: One of the key parts of the Measure phase should be a Measurement System Analysis which I think would answer any question you may have concerning variation caused by the Data Collection process.

    0
    #70888

    Payne
    Participant

    I understand how there might be variation in the data collection process in the Measure phase of Six Sigma. The project we have chosen is related to improving the quality of data that is manually entered (accurate and up to defined standards). My question is – is there an industry average sigma for a data quality process and is it different than the reported 3sigma average. Reason I ask is because we’d like to set an attainable goal sigma for our project. Also, if anyone has heard of Six Sima projects that are related to manually entered data collection processes, I would appreciate the help! Thanks again, Jim.

    0
    #70889

    Jim Johnson
    Participant

    I don’t know of such an average nor of such a Sigma value. Perhaps there are others in the forum who are? You may consider trying to set your own standard based on what your needs are for the data. Obviously, this would be driven by the Critical to Quality for the Measurement System. As such it would almost become a CTM (Critical to Measureable) characteristic.The way that I would go about defining it would be to start with the end in mind. For example, if you want to be 95% confident in your measurement what do you have to do to ensure that the data going in is correct? This should lead you to the critical x’s (inputs) that go into the data entry process. Study these and determine what the optimal settings are (perhaps perform a DOE) for the critical components and viola! You may have the standard that you seek.Does this make sense or am I missing something?

    0
    #71017

    Peter Blanchard
    Participant

    This white paper contains some useful information:-
    http://www.vikingsoft.com/vdewp.htm
    I’m nothing to do with them, but they’ve obviously done some benchmarking. What I found confusing was that the real life error rates I’ve come across are much higher than the white paper suggests – but then what I was looking at was defectives rather than defects. The white paper is talking about defects.
    I don’t know what measure of error you’re using, but it is relatively simple to calculate a Sigma score even from number of defectives data provided that you can estimate an average number of opportunities for an individual data item. Clue is that the probability of a non-defective is given by 1 – (error rate)^number of opportunities.
     

    0
    #71030

    Mikel
    Member

    Kelly,I don’t know where you are learning Six Sigma from, but the goal of your project is simple. Establish your present capability as part of the Measure phase, and improve that capability by a minimum of 50%. How did you decide this was the right thing to improve?

    0
    #71046

    Groll
    Participant

    ASQ’s Measurement division has spent fruitless hours on the concept of data and entry of the data.  I have found that it is best to automate as much data entry as possible and into a very reliable place.
    Best of luck!!!

    0
Viewing 7 posts - 1 through 7 (of 7 total)

The forum ‘General’ is closed to new topics and replies.