Date dimension to a table with multiple date columns calculations - ensuring it's correct set up

Hello, 

I am in need of a confirmation that I have set up the relationships and connections correct for my report. I am connected to a datalake for my first time so I am in slightly unknown waters. 

I am currently setting up a dashboard for a fact table that has 6 date columns.

The dim_date table has a date_key which is a 8 integer maching every single data from 1980-2030, and the date columns in my fact table are all written as 8 integer entries instead of dates so they can be connected to the dim_date table using the date_key.

(please ignore the auto sum on beslutningsdato and lopedag, I have turned that off in my semantic model now).

Now all the entries in the fact table are all individual cases recognised by a unique ID case number, that all has a "received day/INNDAG", "valid day" and  "expiration date". In this case I'm setting up a dashboard based on total received cases per year and facts about these cases, thus I need the dashboard and data to be filtered on "INNDAG". To ensure this, I have set up the relationdship between the tables based on date_key and INNDAG with a one to many relationship. 

To be able to summarize and count the amount of the cases per year I opted for a measurement with "COUNTROWS" based on the unique case_ID column in the fact table. My main issue is that my numbers aren't exactly accurate with what has been reported for prior years so I was wondering if there's something I should do differently or if my countrows measurement is correctly filtered and calculating the amount of rows based on the INNDAG with the current set up.

I was thinking setting up inactive relationships for the other date dimensions and utilising USERELATIONSHIP/role-playing dimensions for other analyses if necessary. 

Thank you so much for your time.