iSixSigma

Relationship Between Factors Analyze Phase and DOE

Six Sigma – iSixSigma Forums General Forums Methodology Relationship Between Factors Analyze Phase and DOE

Viewing 5 posts - 1 through 5 (of 5 total)
  • Author
    Posts
  • #54210

    Marc van der Peet
    Participant

    Dear all,

    We’re about to move to the improve phase in my lean six sigma project but I had some questions about how i should use a potential DOE in the improve phase.

    My idea would be to use the DOE for measure of the levels of the analyse factor which could be used best. So for example:

    in the analyze phase i determined that seniority of employees is relevant for reducing the cycle time. Then in the DOE i’m going to investigate which levels of experience will generate the best results (shortest cycle time). Good approach?

    Dear Regards,

    Marc

    0
    #194237

    MBBinWI
    Participant

    @marcvanderpeet12 – Marc: Only if you can somehow control “seniority.” I think that you need to peel the onion back a bit further to find the underlying factor that distinguishes some employees as being more capable than others, and time itself doesn’t seem the most likely candidate.

    0
    #194245

    Henk
    Guest

    @MBBinWI Thanks for your reply. So if we manage to to define the link between seniority and reduced cycle time a bit further we should do a DOE with the different levels. Say for example we find out that the amount of cases handled(linked to seniority) we can do a DOE with:

    a 1000 cases dealt with
    b 2000 cases dealt with
    c 3000 cases dealt with?

    0
    #194246

    Hari VS
    Participant

    Hi Marc,

    From my experience, I think while considering seniority, it will be a good option to also consider what makes the senior employee to do that exceptional work. Because sometimes providing an extra training for junior employee may bring them upto speed. Also if you are able to identify that extra factor/factors, you can consider the same for your experiment.

    0
    #194247

    MBBinWI
    Participant

    @henkverstraten – Henk: I don’t know why you’d want to do a DOE using cases dealt with as a factor.

    @marcvanderpeet12 – Marc: Why do you think that this effort requires a DOE? While DOE is the most efficient means of developing data, it is really only needed if you suspect that there are interactions, or multiple Y outputs where the inputs may have different contributions to the outputs (especially if in different directions). I agree that you want to find how to get your people to come to more consistent evaluations, but I’m not so sure that a DOE is called for in this situation.

    0
Viewing 5 posts - 1 through 5 (of 5 total)

You must be logged in to reply to this topic.