iSixSigma

test

Viewing 8 posts - 1 through 8 (of 8 total)
  • Author
    Posts
  • #51090

    nancy
    Participant

    We’re trying to find out the reasons for high volumes of calls and would that in any way relate to the volumes processed…what test can we perform for the same

    0
    #176547

    RK
    Member

    would require more info…but what I can guess is that the set up is a helpdesk where you receive calls and you want to establish the relationship between the volume processed ( invoices?) by the team is any way related to the call volume. let me know if I read this right?
    If I am right, then your X & Y are continuous. So you do a regression
    RK

    0
    #176548

    Clive
    Participant

    From what you’ve said a regression test would be the best test to run with your volume of calls on one axix and volumes processed on the other.
    The difficulty will be to link the 2 values. Without understanding your business I’m guessing that there’s a delay between processing and getting calls. There a few ways on handling this one the best solution is, if possible, link the call to the actual processed item then you can compare the actual item to if it got a call. If this isn’t possible then   decide on what your average lag is say a week and then compare last weeks processed volumes with this weeks calls.

    0
    #176549

    nancy
    Participant

    Can we consider volumes as continious . I thought it’s a discrete measure .

    0
    #176550

    Clive
    Participant

    Depends on the number of different values. I would.

    0
    #176551

    nancy
    Participant

    Its a collection s helpdesk which collects on money from its customers .

    0
    #176553

    RK
    Member

    actually there is a fair amount of confusion whether to consider volume as discreet or continuous. many people treat it as psuedo-continuous.

    0
    #176557

    Clive
    Participant

    I take it your write busy at the moment then. I have worked in a loans company and have run projects in collections and yes you can use the volmes as continuous in this case.
    The problem with looking at predicting call volumes in this environment is how many external factors will effect it. Major factors which I’m sure your aware of are day of week, time of day, day of month (pay days), how your running your dialler  and different sub groups within your customers, gender, age, credit scores, total debt etc act differently.  This is not to say it’s impossible but you’ll be looking at not just one X and Y. Good luck

    0
Viewing 8 posts - 1 through 8 (of 8 total)

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