User:LearningwidVJ/sandbox

Workload

In order to manage contact center effectively all the contact center resources  must be accurately matched up with the workload. So, effective contact center management  would  eventually translate  into:  “Having the  right  number of  skilled people and supporting resources in place at the right times to handle accurately forecasted workload, required service level and with quality”. So,  one can  imagine  that if  forecasting  is not  accurate  then the  rest  of the  planning  process would  also  go off  the  mark. Forecast  is  the basis  for determining  staffing needs  and  requirements for  other  resources.

So,  it  provides foundation for:

1.  Calculating Associates required to meet the service level

2.  Minimizing abandoned and blocked calls

3.  Calculating System and trunk requirements

4.  Preparing accurate Schedules

5.  Predicting Future Costs

6.  Meeting Customer Expectations

7.Ensuring Quality of Service

Simple Methodology (Art & Science)

Forecasting Timelines

In some contact center environments,

•Long term forecasts are done for a year  and beyond  and  they are  used  for estimating  annual  budgets, establishing  hiring plans  and  defining system  requirements.

•

•Medium  term forecasts  in  some contact  center  environments are  done  three months  ahead  and are  necessary for hiring and training plans and as well as for accurately adjusting  scheduling requirements,  planning  for holidays,    and  anticipating seasonal  trends (Month on Month) etc.

One way of classifying forecasting is to consider  the timescale involved in the forecast i.e. how far forward into the future we  are trying to forecast.

Short, medium and long-term are the usual categories  but the actual meaning of each will vary according to the situation that is being studied,  e.g. in forecasting  energy  demand in  order  to construct  power  stations 5-10  years  would be  short-term  and 50  years  would be  long-term,  whilst in forecasting consumer demand in many business situations up to 6  months would be short-term and over a couple of years long-term. The table  below shows the timescale associated with calls or transactions.

Long Term  6 – 12 Months  Medium Term 2 – 6 Months  Short Term  0 – 2 Months

The  basic  reason for  the  above classification  is  that different  forecasting  methods apply in each situation, e.g. a forecasting method that is appropriate for forecasting sales next month (a short-term forecast) would probably be an  inappropriate  method for  forecasting  sales in  five  years time  (a  long-term forecast). In  particular accuracy  of  numbers (data)  to  which quantitative  techniques  are applied  typically  varies from  very  high for  short-term  forecasting to  very  low for  long-term  forecasting when  we  are dealing  with  business situations.

Understanding  Data  and Patterns  is  the primary  step  in forecasting  Ladder.

Plotting Data.

Basic data required for forecasting would include the following:

1.  Sufficient historical data on the past transactions

2.  Data on past Average Handle Time (AHT)

3.  Past data on Shrinkage

4.  Once this data is validated it can be plotted to visualize trends or patterns.

There are various types of patterns, which may exist in a single set of data.

Trend

The  long-term  tendency of  a  series to  rise  or fall  (upward  trend or  downward trend). As per Dictionary meaning, trend is the general direction in  which something tends to move or trend is something which extend, incline, or bend in a specified direction.

Cycles

Long  departures from  the  trend due  to  factors others  than  seasonality. Cycles generally occur over a large time interval, and the lengths  of time between successive peaks or troughs of a cycle are not necessarily  the same

Seasonality

The periodic fluctuation in the time series within a certain time  frame. These  fluctuations  form a  pattern  that tends  to  repeat from  one  seasonal period to another. As per Dictionary meaning,  Seasonal  applies to what depends on or is controlled by the season of the year:  a seasonal rise in  employment. Seasonable   applies  to what  is  appropriate to  the  season (seasonable clothing ) or timely ( seasonable intervention in the dispute ). Rains  are  seasonal if they occur at a certain time of the year; they are  seasonable at  any time if they save the crops.

Repeating Patterns across Various Timelines.

All Incoming contact centers experience three dominant patterns in how calls  arrive, and these are mentioned further: Month  of  Year Pattern  or  Seasonality : At  least  3 years  of  data is  recommended  to arrive  at  seasonality. Seasonality  means  that there  is  a pattern in the data on account of the month of the year. For example, for a particular process February may experience high volumes, while July may be  the month with least volume. For some other process this seasonality may be entirely different. So, it depends on the nature of the client’s business. Further  is an illustration of the seasonality.