iSixSigma

welding metrics

Six Sigma – iSixSigma › Forums › Old Forums › General › welding metrics

Viewing 3 posts - 1 through 3 (of 3 total)
  • Author
    Posts
  • February 8, 2008 at 1:15 pm #49304

    Alen
    ★★★★★★★★★★
    Participant

    Hi,
    I have a new Six Sigma Project, reduction of welding defects on a structural engineering product.Each product undergoes about 4 identical welding processes.(4 sub parts are welded on to the main part).I have the following choices for Process metrics:1. Number of product units which do not have any welding defect1. Number of defect free weldings, out of 4 weldings2. Length of defective weldings out of the total welding lengths over a period of one month Now on finalising one metric from the above I will have to estimate the current sigma level and a target sigma value.
    Now my issues for a discussion will be
    1. What is the logical and correct metric for the process2. How to go about estimating Sigma level? (From Firat pass yield, or from defective percentage or from the Distribution?)
    Request your valuable suggestions.
    Alen

    0
    February 8, 2008 at 2:03 pm #168359

    Anand
    ★★★★★★★★★★
    Participant

    Hi Alen
    In my opinion you can look forward to 3 metrics:
    1) Number of defect in the welding process, each part will have 4 opportunities as there are 4 joints.
    2) Number of defective units, where either one or more defects are faulty.
    3) Strength of the welding joint, where you can measure the strength of each weld by destructive testing. In this case you will have to set up a sampling frequency and based on the samples you can derive your process performance. Technically this should be the most suitable metric as it will provide you an opportunity to improve upon the strength of the welded joint which results in defects.
    Anand

    0
    February 12, 2008 at 1:29 pm #168476

    Alen
    ★★★★★★★★★★
    Participant

    Thanks Anand,
    Unfortunately we dont do destructive inspection on the welding, but only sampling inspection by Non destructive testing. I think, from this we arrive at a distribution.
    Alen

    0
  • Author
    Posts
Viewing 3 posts - 1 through 3 (of 3 total)

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