How to make an attribute relationships and hierarchies when you have two or more attributes with the same key value but different name?

Question

When making a date dimension in a SSAS multidimensional cube, it is simple to make a hierarchy and attribute relationships for day, month, quarter, and year, which all roll up into one another. But what if I have a couple of attributes that combine month and year, such as "Jan 2019" and "01/19"? I made these because I know the charts will use them. If I take one of these attributes called [Year Month] and place it as seen below where Date -> Year Month -> Quarter -> Year, then it gave me a warning that I create another hierarchy, which I did, but that seems not user friendly to have a few nearly identical hierarchies when they are browsing fields. Or should I have kept [Year Month] in an attribute relationship with Month Name, like I did with the other Month Year fields in that list?

If I don't add additional hierarchies, then I get a warning that says "Design hierarchies for each incoming relationship path when attributes have multiple incoming relationships or, if any of the incoming relationships are unnecessary, you should delete them."

In summary, I have a few attributes whose key value is the same (month/year)- how to approach this?

The warning (blue squiggly) is just saying to avoid visible attributes used in hierarchies. I will get around to fixing that once I solidify things.

Answer 1

After googling the warning message that I added into the question above, I discovered some posts/articles suggesting you really should add a hierarchy for each path or otherwise the "attribute decoding" won't work. I also think it does make sense to put Year Month on its own path (so the diagram is correct). And then I discovered you can make a hierarchy invisible to the end user. If you go to your cube and expand a dimension and then click on an individual hierarchy, there is a Visible property you can set to false.

However, I also decided to reduce these additional attributes of type string/varchar down to one after reading how it can hurt performance to have a lot of string attributes.