Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors

Expressing Schedule Inflexibility (math)

  • This topic has 8 replies, 1 voice, and was last updated 11 years ago by Jim Mathew.
Viewing 9 posts - 1 through 9 (of 9 total)
  • Author
    Posts
  • #18555
    Matt
    Guest

    Specifically, I’m considering the disconnect between calculating required staff at a monthly or yearly level and the translation to the daily, 30-minute interval level.

    For example – based on a set of shrinkages, handle time, and other factors, there might be an FTE goal of 150 that a call center needs to be at for given month.

    However, if you are creating schedules for 150 FTEs over a week, you\’ll end up very short due to schedule inflexibility factors such as:

    Agents working 8 or 10 hour contiguous shifts (versus coming in for 2 hours, taking off for 5, and coming back in the other 6)

    Agents having at least 1 weekend day off (or having 2 consecutive days off)

    Having a minumum number of agents on the phone for inteligent call routing (you might only need 3 people on the phone to meet the demand at 6:00am, but need to schedule 10 to allow for vacations, sick time, etc)

    So in reality, that 150 required FTE ends up being something like 200 schedules to address all of your requirements across 7 days a week and 24 hours a day.

    Is anyone aware of a way to mathematically express this increase to required staff?

    #18556
    Catalin
    Guest

    I don’t know but if you found out pls help me. I need a formula too.

    Thanks

    #18557
    Matt
    Guest

    We eventually brute-forced it by making a matrix of volumes and schedule types, so we could very quickly respond with what the load calculation of staff was vs the number schedules that would be required to actually cover based on minimum staffing levels.

    So if the volume was ~100000 calls per month and the coverage was 5 days a week, we’d be able to return our numbers very quickly.

    I moved on to other things, but that remained in on the backburner for me. I eventually made some contacts with actuaries and was planning to work with them to figure it out since they specialize in equations with massive numbers of variables, but my role changed within the company and I am no longer responsible for having to justify to financial planners why their calculated number of FTEs is so much lower than what we need to actually meet the service levels 🙂

    #18558
    Rishabh SIngh
    Guest

    Did you get a chance to crack this piece, i have been looking for the same since last 6 months and am constantly surf this page, hoping for something good.

    As per normal WFM tools,
    Inflex = (Sched open – Req)/Sched Open.

    #18559
    Rishabh Singh
    Guest

    Guys – Is there no one from contact centre, who have some thoughts around inflex quantification

    #18560
    Catalin
    Guest

    I resigned to search for a formula…
    Fortunately my WFM soft calculates schedule inflexibility by it’s self 🙂

    #18561
    Chris Linden
    Guest

    I’ll have a crack at answering this…

    To determine a figure to use in the planning process you will need to analyse historical data from your WFM system or ACD. What you need to measure is the Actual FTE Delivered vs Actual Required FTE and compare that to the Service Level achieved.

    This can only be measured after the day has passed and an accumulation of historical data will give you the figure that should be used for planning purposes. You can either select days where your actual SL was on or close to target or plot all ‘normal’ results on a graph and use that to determine what the schedule inefficiency figure will be to achieve SL target.

    You should find that to achieve target your Actual Required FTE figure will be higher than the Actual FTE delivered and the % difference between the two is what you should use in the planning process.

    #18562
    Rishabh Singh
    Guest

    I think I have a crack on this piece (mathematically), and its based on MAPE (mean average percentage deviation/error)- if we considering the variance b/w intraday required heads and delivered heads, the average of absolute % variance would let us know, what is the schedule inflexibility of a day.
    By this methodology, we would know, how aligned are we with the workload arrival curve on an intraday basis…

    #18563
    Jim Mathew
    Guest

    Inflex if planned to be applied to capacity planning, should be calculated as follows:

    Sum of Intraday Excess scheduled against required / sum of intraday required

    Efficiency of schedule can be calculated as:

    sum of absolute difference between intraday required and scheduled / sum of intraday required

Viewing 9 posts - 1 through 9 (of 9 total)
  • The forum ‘Call center’ is closed to new topics and replies.