Import historical data for demand forecasts

To help guarantee the accuracy of demand forecasts, you must have as much historical demand data as you can get per item or item allocation key. If the historical demand data isn't already imported, use the Historical external demand (ReqDemPlanHistoricalExternalDemandEntity) data entity in Microsoft Dynamics 365 for Finance and Operations to import it.

In the Data management workspace, you can see an overview of all the fields in the entity.

  1. Open the Data management workspace.
  2. Click the Data entities tile.
  3. Search the entity list for Historical external demand.
  4. Click Target fields. The following entity fields are mandatory: site (DeliveringSiteId), date (DemandDate), quantity (DemandQuantity), and either item number (ItemNumber) or item allocation key (ProductAllocationKeyId).

To use the data entity, you must have a Microsoft Excel file or comma-separated values (CSV) file that contains the historical demand data. The following example shows how to import the data from a CSV file.

Example

You can use the following file as an example. Download the HistoricalDemandData. This file contains the historical demand data for item D0001. It contains only the following mandatory fields: site, quantity, and the demand date.

  1. Select the company to import the historical demand data into.
  2. Open the Data management workspace.
  3. Click the Import tile.
  4. Enter a name for the import project, such as Import historical demand for item D0001.
  5. In the Source data format field, select the file format of the file that you're importing. To import the HistoricalDemandData file for this example, select CSV.
  6. In the Entity name field, select Historical external demand.
  7. Save the file to your computer, and then upload it.
  8. Click Import.
  9. The Execution summary page is opened automatically. Verify the imported data on the page.

After you've imported the historical demand data, you can generate a demand forecast.

See also

Generate a statistical baseline forecast