In today’s post I am going to demonstrate how to use Demand Forecasting using D365 and Azure Machine Learning.
Go to the Demand Forecasting Parameters using the following navigation:
Master planning–>Setup–>Demand forecasting–>Demand forecasting parameters
Set the forecast generation strategy to Azure Machine Learning and save the form. After saving, you will be presented with the following warnings that you will need to resolve.
Visit the url: http://aka.ms/dynamicsax7-demandforecasting and click on open in studio, you will be asked to login for which you can use your microsoft account
This is a standard Demand Forecasting experiment provided by Microsoft that we will use.
In the next screen you will be asked to select a region. Click the “Tick” icon to proceed.
An experiment will be opened in the designer as shown below
There are 3 types of inputs this experiment accepts:
- Web service input – input received (data and parameters) from D365FO
- Sample Data – this is sample data that will be used when we run the experiment stand-alone.
- Parameters – these are the sample parameters used in conjunction with the sample data above to run the experiment
For this demo we will use the “Web Service Input”.
In order to use this experiment as a service you will need to run it and then deploy it. Click on the “Play” button at the bottom of the screen
After the experiment is run you will need to click on “Deploy Web Service”
After the experiment has been deployed you will need the API of the web service:
You will also need the web service URL, to do that click on the “API help page” for the REQUEST/RESPONSE to get the web service URL
There will be a long Web service URL, we don’t need all of that; just include everything up to the /execute?api
We now have deployed the web service, and have 2 values. An API key and a Web Service URL, now its time to go back and update the Demand Forecasting Parameter.
We are done with the setup, now we need to run the demand forecast. To do that follow the navigation Master planning–>Forecasting–>Demand forecasting–>Generate statistical baseline forecast. Use the following parameters
Click OK and wait a few minutes. After the operation has been completed, you will be notified. In order to view the results use the navigation Master planning–>Forecasting–>Demand forecasting–>Adjusted demand forecast
And you will be presented with the results
You have successfully implemented the Demand Forecasting in D365 For Finance and Operations using Machine Learning.
That’s it for this blog peeps. I will be back with more.