iSixSigma

Call Center Related Question

Six Sigma – iSixSigma Forums Old Forums General Call Center Related Question

Viewing 15 posts - 1 through 15 (of 15 total)
  • Author
    Posts
  • #47586

    Rolando
    Member

    Our actual capacity is higher than our call volume.  Yet we still can’t handle all calls and abandoning so many calls.  AHT was met though.  What’s our problem?

    0
    #158735

    Chris Seider
    Participant

    Sounds like your capacity is your theoretical capacity which is calculated too high or not have the correct input data to model the capacity.

    0
    #158736

    Rolando
    Member

    Thanks for the help.  I was actually guessing someone would point that out.  Based on your experience, is this a training issue?  quality issue?
     
    What can you suggest?

    0
    #158737

    Chris Seider
    Participant

    I’ve learned to never suggest a solution without data.  But you should be able to find out easily within one day of data collection where gross explanations come from….though the solution may not be identified.
    Good luck.

    0
    #158738

    Rolando
    Member

    Thanks.  It’s a good challenge.

    0
    #158744

    Quality Novice
    Participant

    I agree with C Seider and if I may suggest, also look into your trending data. Your historical data may have an anomaly. Your historical data should only have regular scenarios. If it has, might as well take it off from the trending table and replace it with another….just my 5cents.
    Good Luck.

    0
    #159096

    Pasi
    Participant

    Hi,
    Have you considered figuring out where the variation lies? Is it time -related, location or between ?
    Regards
    RB

    0
    #159098

    Pasi
    Participant

    Hi,
    You could do a minima-maxima and this graph should merge with your processing capacity.
    Rgds
    RB  

    0
    #159102

    Shelby
    Member

    I agree with many of the other replies.  However, I have a suggestion that worked for me while diagnosing a similar process recently.
     
    Go to the work area and observe.  i.e. Do a Lean Manufacturing project starting with a waste walk.  You will likely find potential improvements that can be tested quickly.  While doing this, make sure you learn every value stream that the call center is involved.  In my recent project, other team members performed the pre-work.  They had missed 2 very time consuming responsibilities. 
     
    By going to the actual workspace, you will be able to best understand the situation.
     
    If this process fails to identify opportunities for improvement, you should definitely study this as a SS problem.
     Good Luck.

    0
    #159106

    Dan M
    Participant

    Hi all.  This is an interesting problem, very similar to manufacturing capacity.  Even though we state a weekly capacity, that requires making units at a defined frequency throughout the week.- lots time can not be made up.  For example, if we can make one unit per minute, we can state our capacity as 480 per day or 2400 per week.  But if you loose a minute on Monday, you are a unit behind on Friday afternoon.
    Perhaps the call volume capacity is sufficient at the gross level, but if we receive a disproportionate number of calls in a period- say 12-1 when people are on lunch and call to resolve in issue- then the “instantaneous” capacity is too small.
    I like the idea from others about observing the process, and charting the calls over time- you may see that the call loading is uneven, and if there is any cycles or trends, you might be able to adjust your staffing to match, as one example.

    0
    #159109

    Mr Kid
    Participant

    Your main problem is that you are looking at averages and not taking variability into account.  Undoubtedly, you have variability in call volume and call length.  I would guess that you have times where you abandon calls and other times where operators are sitting idle.  Is there a way to decrease the variability in either of these areas?  Do you have standard work implemented?
    If you can’t do that, or you’ve reached the limits of what you can do, to avoid abandoning calls, your capacity needs to be able to handle your peak volume, not just your average volume.  So either you can increase your capacity across the board (hire more operators), or figure out a way to have peak capacity at the times that you have peak call volume.  This would start with trend analysis of your data, and I would think lead to potential solutions like staggered shifts, flexible work hours, part time employees, etc.
    Also, if you need to add overall capacity to handle the peak times, is there something else that you could do with this capacity during the off-peak times?

    0
    #159111

    Mr. White
    Participant

    So basically, your problem is impossible to solve, based on your understanding of it.  
    Reliable metrics?   Prove it!
    Gather data.
    Determine distribution of data.
    R&R gauge study?
    Test hypothesis.
    Rinse.
    Repeat.

    Good luck, this is not a tough problem, you failed to understand it, so start over and do it again. You have this basket of tools, let your intuition guide you.

    0
    #159112

    GrayR
    Participant

    Fully agree with Mr. Kid.  You have two sources of variation that your prcess is dealing with — variation in the number of calls, and variation in how your operators handle the calls.  Ideally you would like to reduce variation from both sources so that you don’t need the excess (your actual capacity).  Since it may not be easy to reduce demand variation, you will prbably still have to maintain excess capacity to handle the peaks.  On the supply side however, using standard work is the first step.  Second, monitor the call lengths and understand why there is variation in handling a call.

    0
    #159113

    IVR/VRU
    Participant

    Please let’s not forget that you also have the IVR/VRU. How is the VRU structured and how do calls flow to the rep. Also, are you dealing with a VRU that has loopbacks, dead ends or is based on voice technology? In essence there is an intervening variable between incoming call volume and handling of calls. The analysis needs to review all three critical elements. 

    0
    #159177

    Pritam Parekh
    Participant

    You should install auto answering machine and record the matter and compliance to be done accordingly studying the facts.

    0
Viewing 15 posts - 1 through 15 (of 15 total)

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