AI-Based Matching of D365 F&O Product Attributes with External Data

In Dynamics 365 Finance and Operations, every product has specific attributes (specifications) associated with it.

In our case, each product is a cable item, we have a set of attributes like:

• Conductor: Steel Wired Armour
• Diameter: 10mm
• Colour: Blue, Green, Yellow
• Etc.

The challenge arises when we need to match these product attributes with another set of attributes that are retrieved from external documents. Since the attribute values in the documents and D365 F&O might not exactly match (e.g., slight variations in how they are written or formatted), querying the D365 F&O database directly isn’t a viable solution. The values may be similar but not identical.

Since direct querying of the D365 F&O database isn't feasible for exact matching, we rely on AI to perform more flexible and intelligent matching.

Our Objective:

To find the most relevant products in D365 F&O by comparing the attributes from the documents with the product attributes stored in D365 F&O.

Possible Approach: We’re considering the following approach and would love feedback on its feasibility or suggestions for improvement:

1. Extract the product details and attributes from D365 F&O.
2. Store these records in an Azure database.
3. Use AI (through platforms like AI Foundry) to analyse and match the document-based attributes with the product attributes stored in Azure.

Would this approach—using AI Foundry, exporting the data to an Azure database, and applying AI methods to find the most relevant product—be feasible? Or are there better alternatives to consider?

We’d greatly appreciate any suggestions or alternative methods that may help with this task!