Datamodel Minx over group and period

[hansolu]

Hi

I created a measure in my datamodel called [% calc] which is calulation the usage of a raw material vs the stock level of this rawmaterial

Now I need to calculate the min value of this measure for each assembly (for all sub2items used for the specific assembly)
Finally I show this in a chart per assembly and week with the row total being the min for each assembly over all weeks.

The measure I use is this one:

Please Login or Register  to view this content.

Its working as I hope to be but I am afraid its not the correct way as the Dax Editor is showing me some Callback ID which is never good as far as I understand

Is there an expert who can help me to verify this measure?

If needed, I can prepare a sample data sheet but in the next days

thanks a lot

[JeteMc]

If needed, I can prepare a sample data sheet but in the next days

Please be sure to include the expected result(s) so that we will know whether or not our proposed formulas/code work.
Let us know if you have any questions.

[hansolu]

Hi,

attached the file with the desired output below the pivot table

I use a measuer to calculate Stock Case. This stock case is then used in the % calculation.

Issue in Version1 of my % measure is the row total of the Subtotals (Sub1Item level) are wrong. (always 4%)

Please Login or Register  to view this content.

Issue in Version2 is the performance in my model and that the values itself are wrong but the totals (column and row) are correct (the min. values of the wrong initial values)

Please Login or Register  to view this content.

Sub2Item Level should be same as result as the Stock case.

Its a bit different behaviour than in my original file. If I figure out how to solve this example, I am able to solve my original file I guess.


thanks for any help

[Bo_Ry]

Please try

=Var t =ADDCOLUMNS(CROSSJOIN( VALUES(dSubItem2[Sub2Item]),VALUES(dCalendar[Week])),"St",[Stock Case])
return MINX(t,[st])

[hansolu]

Hi, work like a charm and is also faster than before. I cant say thanks enough

Crossjoin in the measure is new for me, I guess there are some other cases where I can use it. I will have a deeper look into\
thanks a lot