If you are working with inventory planning then likely you do some form of forecasting on your items. It might be simple analysis in spreadsheets and then enter the forecast against your items. If you want to do more statistical based forecasting by looking at history and trying to predict forward then you will need to use some sort of algorithm. In AX2012 the approach that was released is to export to SSAS and use the time series analysis there. Now with D365 we have the ability to use Azure Machine Learning.
You can research more about Azure ML on this link https://studio.azureml.net. You can try it for free to start to learn. Azure ML uses experiments which are the logic or algorithm of what it’s going to do with the data. Dynamics provide one that you can start with which you can find here http://aka.ms/dynamicsax7-demandforecasting
In later posts we’ll investigate the functional configuration in more detail but to start we’ll walk through getting the technology configured so you can run the base line forecast generation and have the data going back and forward to Azure ML.
Here is a quick overview.
D365 July 2017 Update 8 (7.0.4565.16211)