Use a custom Azure Machine Learning algorithm

  • 10 minutes

Your organization can continue to use its own Microsoft Azure Machine Learning algorithms for demand forecasting in Supply Chain Management.

Set up a new Microsoft Entra application

The Microsoft Entra application is a business application that allows Demand Planning to connect to your organization's Azure Machine Learning algorithms. To create a new Microsoft Entra application in your Azure Machine Learning workspace, follow these steps:

  1. Sign in to the Azure portal.

    Note

    You must have Cloud Application Administrator privileges to proceed with this setup.

  2. Register a new Microsoft Entra application. For more information on how to do so, see Register a Microsoft Entra app and create a service principal.

  3. Use the default settings and follow the on-screen instructions to complete the wizard.

  4. Create a client secret for the application in the Certificates & secrets section of the new Microsoft Entra application. For more information, see Add a client secret.

  5. Note the application ID and secrets.

Assign access for the Microsoft Entra app to Azure Machine Learning

To assign access for the new Microsoft Entra application to the Azure Machine Learning workspace, follow these steps:

  1. Go to the resource group that contains your Azure Machine Learning workspace.

  2. Select Access control. Under the Role assignments tab, select Add.

  3. Select Contributor on the Privileged administrator roles tab.

  4. Select the User, groups, or service principal option and then select Members.

  5. Locate and select the Microsoft Entra application that you previously created.

  6. On the Review + assign tab, select Next.

Now the application is listed in the All section of the Role assignments tab on the Azure Machine Learning workspaces and storage accounts.

Connect to the Azure Machine Learning service from Demand Planning

To set up the Azure Machine Learning service connection in Demand Planning, select Custom Azure ML on the navigation pane in Demand Planning. Select New to create a new connection.

Screenshot of Demand Planning with the New button highlighted.

Next, enter information in the following fields: Name, Subscription ID, Resource group name, Workspace name, Storage Account Name, Application ID, and Application secret.

Set up a forecast that uses your Azure Machine Learning algorithms

To set up a forecast that uses your organization's Azure Machine Learning algorithm, you must start by creating a new forecast profile. On the Select a forecasting model preset page, select None (Define at later stage).

Screenshot of the Create new forecast profile on the Select a forecasting model preset page.

After you save your configuration, select the Forecast model tab. Set up your model to include a Finance and operations apps -- Azure Machine Learning tile. How the algorithm runs depends on where this tile is set. When you're ready to complete the model, add a Save block.