Time Series
You can use different types of time series to provide the system with historical demand data (transaction data).
The Demand Data Foundation (DDF) module in SAP Customer Activity Repository can receive the time series from an external SAP or non-SAP application. For more information, see Integration Information and Inbound Processing.
The Unified Demand Forecast (UDF) module in SAP Customer Activity Repository can use the provided time series and master data as input for demand modeling and forecasting. For more information, see Unified Demand Forecast and Outbound Time Series.
Make sure that the prerequisites under Data Management have been fulfilled.
Make sure that your historical data is at daily granularity. Weekly data is currently not supported. For more information, see Customizing (transaction SPRO) under .
Recommendation
Time series length: We recommend that you provide 2 years of historical demand data to ensure the proper interpretation of seasonality, trend, and other yearly demand influencing factors (DIFs).
Sanity check: A data validation report is available that you can run prior to modeling and forecasting. It allows you to identify potential issues in the provided input data and rectify them as needed. When you run the report, the system executes modeling and forecasting in simulation mode on the input data and indicates any issues found. For more information, see SAP Note 2161484
.
The following time series are supported:
Consumption data (Generic)
POS data (Sales)
VDMs providing access to aggregated POS data and sales orders in the SAP HANA content of SAP Customer Activity Repository (Virtual Data Models for Unified Demand Forecast)
You can select the time series in Customizing under . For more information, see the Customizing activity documentation (transaction SPRO).