Show TOC

 Forecast with Dynamic Moving Average Model

Purpose

With this process the system executes a forecast with the dynamic moving average model. This forecast model can be used if there is a large deviation of average demand in the demand history, for example, for slow-moving items. The basis for this model is the modified moving average based on weekly periods.

The system carries out the forecast according to this model in four steps. It carries out each of the four steps for both the forecast of order items and for the forecast of average demand size per order item (demand/item) separately. The system firsts defines the optimal amount of historical data that is relevant for the forecast, and then carries out outlier correction and calculates the forecast. At the end the system then uses stability rules .

The basis for this model is the data of the 52 previous weekly periods.

After calculating the forecast, the system calculates the standard deviation and the MAD separately.

For more information about how the system chooses this model, see Automatic Model Selection .

Note Note

This forecast model is not available for leading-indicator-based forecasting because the system only creates the leading-indicator-based forecast for demand and not for the number of order items or the average demand per order item.

End of the note.

Process

Forecast for Order Items

The system adds all order items that there were during the optimal number of past weeks and divides the result by this number of weeks.

Example Example

The system has calculated the optimal number of past weeks as 26. There were 104 order items during the last 26 weeks. This means that the system calculates 104/26 =4 as the weekly forecast for order items.

End of the example.

The system converts this forecast result on a weekly basis to the periodicity that you defined in Customizing under Define Forecast Periodicity . For more information, see the Implementation Guide (IMG) for Advanced Planning and Optimization under Start of the navigation path Supply Chain Planning Next navigation step Service Parts Planning (SPP) Next navigation step Basic Settings End of the navigation path .

To do so, the system divides the forecast result on a weekly basis by the scaling factor for weeks, and multiplies this result by the scaling factor for the periodicity that you defined. You have specified the scaling factors in Customizing under Make General Settings for Historical Data . For more information, see the Implementation Guide (IMG) for Advanced Planning and Optimization under Start of the navigation path Supply Chain Planning Next navigation step Service Parts Planning (SPP) Next navigation step Forecasting End of the navigation path .

Forecast for Average Demand Size per Order Item (Demand/Item)

  • If there was no demand in the last 52 weeks, the system sets the demand/item to the pack size.

  • If demand occurred exactly once in the last 52 weeks, the system sets the demand/item to the current demand.

  • If demand occurred two or three times in the last 52 weeks, the system sets the demand/item to the average of the number of times that demand occurred.

  • If demand occurred four or more times during the last 52 weeks, the system tests the following cases:

    • The optimal number of past periods for the historical periods for the forecast of the order item was 6 or 13 weeks and the following tests are successful:

    • The demand forecast per month is greater than the Param1: Relevant Periods for Demand /Item in DMA Model parameter in the forecast profile on the Model Parameter tab page. You get to the forecast profile on the SAP Easy Access screen under Start of the navigation path Advanced Planning and Optimization Next navigation step Service Parts Planning (SPP) Next navigation step Planning Next navigation step Forecasting End of the navigation path .

    • The increase of the forecast order items of last month’s forecast is a least as large as the Param2: Relevant Periods for Demand /Item in DMA Model parameter in the forecast profile on the Model Parameter tab page.

    • The number of order items during the relevant historical periods for the order item forecast is at least a large as the Param3: Relevant Periods for Demand /Item in DMA Model parameter in the forecast profile on the Model Parameter tab page.

      If these cases occur, the system uses the same length of the demand history for calculating the demand/item forecast as for the order item forecast. If not, the system calculates the forecast average number of order/items for the optimal number of past periods as follows:

      The system calculates the demand/item for each week of the relevant time period, adds the results and divides by the number of relevant weekly periods. The system converts this result based on weeks to the periodicity defined by you as described above.

Demand Forecast

For the demand forecast the system multiplies the forecast demand/item by the number of forecast order items.