System and Method for Long Term Forecasting
Systems and methods of strategic forecasting of demand are disclosed. One such method comprises receiving demand data and computing a variant factor associated with a variant. The variant includes a variant date and a measure of demand on the variant date. The method further includes generating a strategic forecast. The strategic forecast includes a forecast for the variant date, based on the variant factor and on the demand data for the variant date.
1 . A computer-implemented method of strategic forecasting of demand for a customer center, comprising:
receiving demand data;
computing a variant factor associated with a variant, the variant comprising a variant date and a measure of demand on the variant date;
generating a forecast for the variant date, based on the variant factor and on the demand data for the variant date.
2 . The method of claim 1 , wherein the variant corresponds to a special event.
3 . The method of claim 1 , wherein the variant corresponds to an outlier.
4 . The method of claim 1 , wherein the variant corresponds to a queue reconfiguration event.
5 . The method of claim 1 , wherein the variant date occurs in the past.
6 . The method of claim 1 , wherein the demand data comprises past demand data.
7 . The method of claim 1 , wherein the demand data comprises current demand data.
8 . A system for strategic forecasting of demand for a customer center, comprising:
logic configured to receive data describing past demand;
logic configured to receive a description of a variant period, the description including a variant date and a measure of demand on the variant date;
logic configured to compute a variant factor, the variant factor describing an effect of the variant period on the customer center; and
logic configured to generate a strategic forecast based, at least in part, on the variant factor and on at least a portion of the past demand data corresponding to the variant date.
9 . The system of claim 8 , wherein the description describes a special event.
10 . The system of claim 8 , wherein the description describes a queue reconfiguration event.
11 . The system of claim 8 , wherein the description describes an outlier.
12 . The system of claim 8 , wherein the variant date occurs in the past.
13 . The system of claim 8 , wherein the variant date occurs in the future.
14 . A computer-implemented method for strategic forecasting of demand for a customer center, comprising:
receiving an identification of a variant date, the variant date associated with a measure of demand on the variant date;
receiving a variant date range used in determining a variant factor;
finding a non-variant date in the variant date range that corresponds to the variant date but that does not correspond to the identified variant date;
computing the variant factor describing an effect of the variant period on the past demand; and
generating a strategic forecast including a forecast for the variant date based past demand data for the variant date and the variant factor.
15 . The method of claim 14 , further comprising:
computing the variant factor as a ratio of the demand data on the variant date to the demand data at the corresponding non-variant date.
16 . The method of claim 14 , wherein the identification corresponds to a special event.
17 . The method of claim 14 , wherein the identification corresponds to an outlier.
18 . The method of claim 14 , wherein the identification corresponds to a queue reconfiguration event.
19 . The method of claim 14 , further comprising:
identifying a possible variant; and
receiving an indication that the possible variant is to be treated by the method as a variant.
20 . The method of claim 14 , further comprising:
receiving a tactical forecast; and
combining the received tactical forecast with the generated strategic forecast in accordance with a weighting factor that describes relative importance of the tactical forecast and the strategic forecast.