iSixSigma

Time Study Analysis

Six Sigma – iSixSigma Forums Old Forums General Time Study Analysis

Viewing 3 posts - 1 through 3 (of 3 total)
  • Author
    Posts
  • #53011

    Tommy
    Member

    I am working on a contract basis to help a company develop labor standards.  The company refurbishes computer equipment.  The facility is set up like a manufacturing environement, where equipment comes down conveyors to a technician that works on the computer, and then pushes it down the conveyor to the next operation. 
    I am conducting a time study analysis to help creat labor standards for each operation.  The equipment and the conditions can vary greatly, and thus the time a technician spends on the equipment can vary as well.  I have been trying to look at my data as I go and pay attention to the standard deviation to help me figure out if I need to add more data to my analysis.  I am finding that most of my data points fall outside of a 95% confidence interval because of special cause scenarios, which are a part of the work for the technician.  When I look at the data it seems explanable, but I’m wondering if I need to do anything else in my analysis since my data falls outside of the confidence interval.
    Any thoughts?

    0
    #187372

    Dharma Bum
    Participant

    Honestly,
    If the number of special causes are as numerous as you say, then I’d rack and stack the specials in order of impact to the Company in terms of cst/quality/delivery and then start knocking them off one at a time. After you pick off all the easy ones, then go back and delve into further time studies.Low-hanging fruit…snatch it up.

    0
    #187387

    Tierradentro
    Participant

    It sounds like a pretty good application for a VSM/Standardized Work application…essentially focusing on identifying and minimizing the non-work elements (ie 7 wastes) first, then standardize and distribute the remaining value-adding elements to determine a stable rate of performance.  
    For a highly variable process, I think focusing on processing times using mean/std dev/CI is unnecessary and time consuming…juice isnt worth the squeeze….don’t need that much firepower when your process is highly volatile/unstable – the descritors arent reliable estimates of the population anyway. 
    But that is just my opinion – there are many ways to approach it.

    0
Viewing 3 posts - 1 through 3 (of 3 total)

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