Dynamic Communities Magazine

Dynamic Communities creates technology-centric communities to exchange ideas on how to best maximize industry knowledge through user-produced education, enriched networking, and conference attendance.

Using Azure Machine Learning with D365 Finance and Operations

05-15-2020 12:00 Rahul Mohta Dynamics 365 FO | AX

DC-Magazine-Header-Azure Machine Learning

Azure Machine Learning can bring in leap bound features in Dynamics 365 Finance and Operations in various areas. Starting with inventory forecast, it could also benefit financial and other areas. A time consuming mundane activity of forecasting thus gets propelled by AML and Dynamics 365 together.

With the ever-growing need for relying on technology to produce/forecast upcoming trends, sales projections, planning for cash, and inventory, cloud-based machine learning plays a vital role. Not only does it augment human decision making, but it also adds and computes millions of combinations to get the best possible judgment.

Prerequisites to make available

  • Open Azure Machine Learning experiment from the link: http://aka.ms/dynamicsax7-demandforecasting

    • sign in to Azure Machine Learning Studio.

https://gallery.azure.ai/Experiment/Dynamics-Ax7-demand-forecasting-29

Azure - Demand Forecasting

  • Create your project by copying the experiment from public library to your workspace.
    Copy experiment from Gallery
  • Save and Run and Deploy opened experiment as a web service after the service status shows finished running
    Save and Run and Deploy
  • Save the experiment and deploy as web service
  • Copy 'API key' from web service properties page and paste it to 'Web service API key' control in D365FO - Demand forecasting parameters form

API Key from Web Service

  • Open request response help page from web service properties page and copy endpoint address to Paste it to 'Web service endpoint address' control in D365FO Demand forecasting parameters form
    Request Response API Document

    Note that you should only copy the base URL without any parameters. The expected format of the endpoint address is https://{region}.services.azureml.net/workspaces/{guid}/services/{guid}

    *Remove extra parameters

     

     

    Next we assign the URI and web service end point address in D365FO in AML parameters:

    demand forecasting parameters

    With this you are now ready to configure few other steps in demand forecast parameters and ready to run your statistical baseline forecast generation.

    Azure Machine Learning can bring in leap bound features in Dynamics 365 Finance and Operations in various areas. Starting with inventory forecast, it could also benefit financial and other areas. A time consuming mundane activity of forecasting thus gets propelled by AML and Dynamics 365 together.

 
Rahul Mohta

Written by Rahul Mohta

Terms of Use: Dynamic Communities does not take responsibility for any incorrect or outdated information and looks to the author as the expert to provide accurate content.