iSixSigma

reflecting a backlog to senior management

Six Sigma – iSixSigma Forums Old Forums General reflecting a backlog to senior management

Viewing 10 posts - 1 through 10 (of 10 total)
  • Author
    Posts
  • #33528

    Cormac
    Participant

    I need to reflect a constantly changing backlog in a service environment to the executive team.  They like things simple!I do not want to just show the mean as very old cases distort the story and the majority of items in the backlog are resolved quite quicklyAny ideas?

    0
    #90832

    DaveG
    Participant

    Do a power curve or a histogram.  A power curve is 2 lines showing cumulative values over time – 1 for starting events and 1 for finishing events.  [email protected] and I’ll email you an example.

    0
    #90834

    Clark
    Participant

    Cormac, When you say service environment, do you mean the core product being sold is a service or are you in a service department supporting a team that produces a product?
    The only reason I ask is that backlog can be seen from several perspectives depending on the audience and what the actual backlog is.  It would be helpful to know if you provide service to the team or if the entire organization is a service organization.

    0
    #90835

    Mahesh
    Participant

    Hostograms on ‘age’ of request will probably be sufficient.
    Also, if it is a very central piece, you could look at the backlog through the 7 wastes prism of Lean, and analyse how backlog causes waste.

    0
    #90857

    Cormac
    Participant

    The product is a service and the backlog represents the time from initial referral until the close of the sale.  There are multiple teams each dealing with product variations.  Senior Management want to be able to see anomolies in the the backlog growing or shrinking both across time and between teams.
    Thanks Cormac
     

    0
    #90936

    Srinivas
    Member

    I would suggest that you do a span measurement (difference between the first percentile and ninety ninth percentile) for the time taken to resolve cases. This measurement is not influenced unduly by the outliers. Also the customer will notice variation more strongly and this measurement will help you drive home the point that your variation in the extremes is high, and hurting your customers.

    0
    #90943

    Cormac
    Participant

    Srinivas,
    I really like this idea, Why pick the 99th percentile rather than the 98th or 99.9th though; Is there an objective way to select this or is it just ‘gut-feel’.

    0
    #90945

    Srinivas
    Member

    There is no scientific basis for this. Its just a convenient number since we are effectively saying not more than 1% defective on either end of the spectrum. It used to be 95% initially, but then moved to 99% to be more robust. If you feel 99% is too stringent, you could look at 5% and 95% too.

    0
    #90955

    Cormac
    Participant

    I’ll try it at 99% and see how that works — I’ll keep this thread posted on my results; thanks for your help

    0
    #90956

    Tim
    Member

    Try searching this site for “span” as I did this a while ago and found good info on how to use the metric.
     

    0
Viewing 10 posts - 1 through 10 (of 10 total)

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