Help with relationships.

[Syndicate_Admin]

Good morning friends, I wanted to know if someone can help me know what should be the best way in this case.
Currently in my model I have 3 tables of facts, Sales, Purchases and inventory, I have 3 dimensional tables. Calendar-headquarters (In this table are the headquarters of the 3 headquarters of the company) and the table of products where I have all the details of the products, the problem is that the codes are not unique, I can get the same product in two or all 3 stores. So in some reports I get the inventory of another.
It should be noted that the headquarters table is the result of Headquarters = DISTINCT ('products'[headquarters]).
Try to do this by placing all products in my product table, but each product may have different behaviors in each location, e.g. product A may be active for sales at headquarters 1, but at headquarters 2 and 3 it is not active for sales. then the report is placing me Active in all the headquarters the product A that is inactive in the headquarters A and C, but is active in the headquarters b.

I hope I have explained. Thank you.

[Syndicate_Admin]

Thank you very much for your response 😄
The problem I have is that for example in the 3 locations I have the ca10 product, but in each location it has a different behavior:
ca10 at headquarters 1: Asset for sale. Classification: High turnover.
ca10 at headquarters 2: Inactive for sale Classification : No inventory
Ca10 in headquarters 3: Asset for sale Classification: low turnover.
So when creating the reports and creating a classification segmenter places me erroneously in some cases, I can place:
ca10 in headquarters 1: Inactive for sale Classification: Without Inventory, if I handle a unique code of products I am placing more classifications.
I thought about creating another dimensional table with the rankings to see if this improves the behavior, but I also thought that a good solution could be to create a table of products for each location, but I think this is not quite right or can worsen the performance of the reports. That's why I wanted to ask the experts for an opinion. Thanks a lot. Best regards

[Syndicate_Admin]

Thank you 😄 very much for your reply.
The problem I have is that for example in the 3 locations I have the ca10 product, but in each location it has a different behavior.
ca10 at headquarters 1: Asset for sale.
Classification: High turnover.
CA10 at Headquarters 2: Inactive for sale
Classification : No inventory
Ca10 at headquarters 3: Asset for sale
Classification: low turnover.
So when creating the reports and creating a classification segmenter it places me erroneously in some cases, I can place ca10 in headquarters 1: Inactive for sale Classification: No Inventory,
If I handle a unique code of products I am placing more classifications.
I thought about creating another dimensional table with the rankings to see if this improves behavior, but I also thought that a good solution could be to create a table of products for each site, but I think this is not quite right or can worsen the performance of the reports. That's why I wanted to ask the experts for an opinion. Thanks a lot. Best regards

[lbendlin]

Your data model looks great - kudos!

the problem is that the codes are not unique, I can get the same product in two or all 3 stores. So in some reports I get the inventory of another.

Not sure what you mean. Please elaborate.