Start of Content Area

Function documentation Forecast Strategies  Locate the document in its SAP Library structure

Use

The forecast strategy determines how forecast values are calculated.

All forecast strategies are based on statistical forecast procedures and forecast models that represent the time series mathematically.

The exponential smoothing methods are currently the most widely used time series patterns (see Exponential Smoothing).

If you expect historic values to continue to develop as they have in the past, choose a forecast model that fits the time series pattern.

Note

The Automatic Model Selection strategy allows you to let the system select the forecast model that best fits the trend of historic data (see Automatic Model Selection).

Features

The following forecast strategies are available:

Average

The forecast value is calculated from the arithmetic mean of the historic values.

      Optional forecast parameters: outlier correction, logging statistical key figures, ignoring initial zero values.

Moving Average

The forecast value is calculated according to the order.

      Obligatory forecast parameter: Order of MovingAverage

The order of the moving average is a number N that determines the length of the time interval for calculating the average. This is the number of chronologically sequential historic values. The forecast value is calculated as the average of the last N historic values.

Do not enter a negative number for the order.

      Optional forecast parameters: outlier correction, logging statistical key figures, ignoring initial zero values.

Weighted Moving Average

When the system calculates the moving average, each historic value is given a particular weight.

      Obligatory forecast parameter: Order of MovingAverage

The order of the moving average is a number N that determines the length of the time interval for calculating the average. This is the number of chronologically sequential historic values.

Do not enter a negative number for the order.

      Mandatory forecast parameter: Weighting Factors.

The weighting factors specify the relationship between the individual historic values and the average calculation. The sequence is important: Weighting factor 1 refers to the previous periods; weighting factor 2 refers to the periods before that, and so on.

Example

You want to create a forecast based on monthly values and choose a weighted moving average with an order that has the value 6. In this case, you want to place more weight on the most recent monthly values than on the less recent monthly values. The historic data is taken from months 5 to 10. The 6 weighting factors and the relevant months are as follows:

No.

Weighting Factor

Month

1

3,00

10

2

2,00

9

3

2,00

8

4

1,00

7

5

1,00

6

6

1,00

5

      Optional forecast parameters: outlier correction, logging statistical key figures, ignoring initial zero values.

Linear Regression

Simple linear regression (ordinary least squares).

      Optional forecast parameters: trend damping, outlier correction, logging statistical key figures, ignoring initial zero values.

Seasonal Linear Regression

Seasonal linear regression is based on the same statistical procedures as used in demand planning.

Note

For more information, see http://help.sap.com/ SAP Business Suite SAP Supply Chain Management SAP APO 3.1 Application Help Demand Planning Demand Planning Process Definition/Redefinition of Forecast Models Creating a Master Forecast Profile Univariate Forecasting Forecast Strategies Seasonal Linear Regression

      Mandatory forecast parameter: Periods per Season

      Optional forecast parameters: trend damping, outlier correction, logging statistical key figures, ignoring initial zero values.

Simple Exponential Smoothing (Constant Model)

Simple exponential smoothing is suitable if the historic data shows a constant trend.

      Smoothing factor settings: Alpha (base value)

      Optional forecast parameters: outlier correction, logging statistical key figures, ignoring initial zero values.

Simple Exponential Smoothing with Alpha Optimization (Constant Model)

This procedure corresponds to the “simple exponential smoothing“ described above, with one modification; in addition, the system calculates the Alpha smoothing factor. The Alpha value is variegated in the interval using the defined step size and a forecast calculation (for the historic time frame) is performed in each case. The optimum value for Alpha is the value that produces the smallest error in the forecast results.

      Smoothing factor settings: Optimization Variable, Alpha From, Alpha To, Alpha Step Size.

Linear Exponential Smoothing (Trend Model)

The forecast is calculated according to Holt’s method and is suitable if historic values display an upward or downward trend.

      Smoothing factor settings: Alpha (Base Value), Beta (Trend Value).

      Optional forecast parameters: trend damping, outlier correction, logging statistical key figures, ignoring initial zero values.

Seasonal Exponential Smoothing (Seasonal Model)

Choose this strategy if your historic values show seasonal fluctuations (for example, annual fluctuations) from a constant base value.

      Mandatory forecast parameter: Periods per Season

      Smoothing factor settings: Alpha (Base Value), Gamma (seasonal components).

      Optional forecast parameters: outlier correction, logging statistical key figures, ignoring initial zero values.

Seasonal Trend Exponential Smoothing

The forecast is calculated according to Winter/Holt’s multiplicative method and is suitable if historic values display seasonal fluctuations from an upward or downward trend. The fluctuation increases with an upward trend.

Example

Ice cream sales in summer: Assume that ice cream sales rise by a trend of 10% annually. A seasonal increase of 30% each summer then leads to ever greater absolute fluctuations.

      Mandatory forecast parameter: Periods per Season

      Smoothing factor settings: Alpha (Base Value), Beta (Trend Value), Gamma (seasonal components).

      Optional forecast parameters: trend damping, outlier correction, logging statistical key figures, ignoring initial zero values.

Croston Method

The Croston method was developed specifically for sporadic trends. This procedure uses exponential smoothing to calculate a mean time interval between the values in the time series that are not equal to zero.

Note

For more information, see http://help.sap.com/ SAP Business Suite SAP Supply Chain Management SAP APO 3.1 Application Help Demand Planning Demand Planning Process Definition/Redefinition of Forecast Models Creating a Master Forecast Profile Univariate Forecasting Forecast Strategies Croston Method

Check whether you want to aggregate the data in order to remove the gaps in the time series so that you can use procedures that consider trend or seasonal time series patterns. You can aggregate data in this way by choosing a rough time characteristic (month instead of day) or by forecasting values for product groups instead of individual products.

 

End of Content Area