Alternative Historical Data 

Use

You can improve the accuracy of forecasting by using alternative historical data (AHD) as the data basis for forecasting for the process of consumption-based planning.

In this context, alternative historical data can be seen as freely definable data sources containing historical data, which can be used for forecasting in stores or distribution centers.

Note that the alternative historical data function should only be used in certain cases. For more information on the use of alternative historical data, see Requirements Planning: Forecast using Alternative Historical Data

Integration

In order to use alternative historical data for forecasting, you need to load data from freely definable data sources in table WAHD of SAP R/3. 

·         In the standard system, a scenario is delivered for which the alternative historical data is loaded from the component SAP BW to SAP R/3.

·         In principle, you can take alternative historical data from any other data source. You can link such data sources using the Business Add-Ins provided.

Prerequisites

If you want to load alternative historical data from SAP BW, note the following requirements for this scenario:

...

       1.      The scenario requires suitable data to exist in SAP BW.  Define which data from SAP BW is to be used as alternative historical data as the data basis for a forecast. Note that a suitable query for the AHD scenario must be available in SAP BW. 

       2.      Using Customizing for Data Retraction in SAP R/3 (transaction MCW_AA) you can determine how and which data you want to load from SAP BW to SAP R/3.

       3.      You can control alternative historical data from the Standard SAP R/3 Customizing. The alternative historical data itself is controlled using RP types.

Features

Using the alternative historical data functions, you can maintain the consumption values as follows:

·        Using transaction WAHD1, you load alternative historical data into table WAHD in SAP R/3. You load the alternative historical data from the data source you designated (for example, SAP BW). Once you have loaded the data, the alternative historical data is immediately available to use in forecasting.

With these functions, it is possible to look at the individual processes involved in loading the data using a flow trace. Use this function to help you recognize and deal with any problems that may occur when loading alternative historical data. Logs that are generated are not saved.

These functions also enable you to work in a test mode. In the test mode, you can simulate the loading of alternative historical data, without the data being posted to the database.

In order to achieve the best possible performance, you can process these functions in parallel. Use parallel processing when you want to process large amounts of data. 

·        Using transaction WAHD2, you can display alternative historical data for a combination of site and article and you can change this data. Use this function to check and, if necessary, alter alternative historical data. You can also display other time series data, such as consumption values, forecast values or the promotion smoothing of consumption values.

·        Using transaction WAHD3, you can display alternative historical data for a combination of site and article. You can also display other time series data, such as consumption values, forecast values or the promotion smoothing of consumption values.

·        Using transaction WAHD4, you can delete alternative historical data for a combination of site and article. Use this function regularly to delete redundant alternative historical data from the database table.

With these functions, it is possible to look at the individual processes involved in deleting the data using a flow trace. Use this function to help you recognize and deal with any problems that may occur when deleting alternative historical data. Logs that are generated are not saved.

These functions also enable you to work in a test mode. In the test mode, you can simulate the deletion of alternative historical data, without the data being posted to the database.

In order to achieve the best possible performance, you can process these functions in parallel. Use parallel processing when you want to process large amounts of data.

You can carry out the functions for loading and deleting alternative historical data online as well as in the background. SAP recommends that you carry them out in the background. Online processing is only recommended when processing very small amounts of data or for testing and simulation purposes.

Constraints

Loading alternative historical data is a very runtime intensive process.

In order to avoid runtime bottlenecks, you should make sure whether, and above all to what extent, it is worth loading alternative historical data.

You should only load the complete alternative historical data if really necessary, for example, if you need to make significant changes to the existing delivery relationship. Otherwise you should just load missing periods.