Need help determining process sigma level

Six Sigma – iSixSigma Forums Old Forums General Need help determining process sigma level

Viewing 4 posts - 1 through 4 (of 4 total)
  • Author
  • #37557


    Hey folks,
    I am currently seeking my greenbelt and am wrapping up the Measure stage and am stuck.  I am trying to determine the process sigma level for my process and am not sure what to do.  The histogram is positively skewed which is what I think may be throwing me off as most of findings are outside my upper spec limit. Here is what I have for info:
    Potential Within Capability
    Z. Bench = -0.79
    Z.LSL= 1.21
    ZUSL= -0.45
    Cpk= -0.15
    CCpk= 0.13
    Overall Capability
    Z. Bench = -0.85
    Z.LSL= 1.01
    ZUSL= -0.37
    Ppk= 0.13
    Process Data
    LSL = 0
    USL= 2.5
    Sample Mean = 3.96701
    Sample N = 288
    St Dev (within) = 3.28972
    St Dev (overall)= 3.93974   
    I am calculating number of unproductive hours per technician per week. (36 techs, 8 weeks).  Can anyone help me out?  Thanks.



    How were the data collected? Were these 288 back-to-back parts? If so, how long was the time it took for you to collect data? In that time frame, do you think all sources of variations would have acted upon the process? (Example: Operator, raw material, etc) Or, did you have subgroups? If so, how many subgroups? How many per sub-group? What is the average time to gather the sub-group? These will help you to characterize the standard deviations that you have calculated. Please post your response. Thanks. Karthik Subbiah



    Thanks a lot for the quick response.  First of all, my project is basically to reduce the amount of negative unproductive time (time that is not billable to the customer, and excludes vacation, sick time, holidays and training).  Examples of unproductive time includes sitting idle, doing paperwork, cleaning out their service vans, etc.
    Because all types of unproductive time are counted in our JD Edwards system as one code, I had to take a sample of the 36 technicians time sheets.  I decided to do 8 weeks worth of time sheets for each technician thus giving me 288 points in my sample.  the values range from 0 hours per week to 16 hours per week.  Right now the process is not monitored so we are shooting for a LSL of 0 hours per week and 2.5 for the USL. Btw, I tried calculating this by day per tech instead of week but most of the datapoints were concentrated at the 0 or 1 mark thus not giving a very good graphical picture in Minitab. 
    Anyway, most of my data is outside of scope and the histogram is poitively skewed with the ZBench being a negative value.  I have read all my books and searched the net for an answer on how to calculate this but have not been able to do this yet so I am getting quite frustrated.  My blackbelt mentor was away so I was unable to seek his advice. Does this pait a clearer picture?  Thanks id advance. 



    Some ideas pertaining to change in approach here. Please define the target  time (set a goal) using customer demands that need to be met, Takt Time, within a scheduled time period (use number employees x nunmber of hours per day or week of available time and divide this product by the customer demand that needs to be completed within the scheduled time period). Determine the cycle time per employee, actual time to complete an order, and compare this to the Takt Time. Then, do a value-stream mapping for the order-to-ship cycle that will enable you to capture, with the employees’ participation, 1) Value-added activities (is it in line with the customer expectations), 2) waste (eliminate these and these could be the unnecessary activities that you have compiled), and 3) incidental activities (reduce these activities).
    I am suggesting these because you are dealing with people (transactional process) and therefore, an exercise such as this will help you design meaningful work for people, hence satisfaction and loyalty.
    Your analysis so far has been on capturing unproductive time without defining what productive time is. Use the data that you have by creating a matrix, employee along the columns and weeks along rows. Put in your data in this matrix and determine the range of time (unproductive) per employee and identify the one with lowest range and this may be helpful in benchmarking during the value-stream mapping process. Please go to and buy a book on value-stream mapping and it will walk you through the steps. Good Luck. Please post your questions.

Viewing 4 posts - 1 through 4 (of 4 total)

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