MDX issue

This is a Microsoft BI environment, where we’re using SQL Server 2008 Analysis Services for cube building. The below scenario uses MDX in the SSAS Cube.

Please help me

Your scope expression will not work as you are referencing CurrentMember values for [dDataType].[Category Data Type]. Unless these are included in the query there will be no current member for them. You could try nested Scopes (rather than explicitly testing the current member values) but I think this might have the same problem. I will try to find an alternative for you later.

Tim

p.s. as an aside, rather than using:

[dDataType].[Category Data Type].CurrentMember.Name = "Budget"

use

[dDataType].[Category Data Type].CurrentMember is [dDataType].[Category Data Type].&["Budget"]

as it's more efficient

No this doesnt work either. Please help

Did you try nested scopes?

i.e.

SCOPE ([dOrganisation].[Organisation Hierarchy].[Reporting Unit].&[BELUX HYPERION]);  

SCOPE ({[dDataType].[Category Data Type].&[Budget], [dDataType].[Category Data Type].&[Forecast]});

THIS = ([dOrganisation].[Organisation Hierarchy].[Reporting Unit].&[BELUX HYPERION_]);

END SCOPE;

END SCOPE;

I have tried this in an equivalent calculation and it works with the value only in the slicer.

Tim

Still no luck... Same result...

Here is how I think the scenario got closer. I created a MEMBER like below:
 
CREATE MEMBER CURRENTCUBE. [Measures].[000TD_FC_TEST] AS
(([dOrganisation].[Organisation Hierarchy].[Reporting Unit].&[BELUX HYPERION_]
,[dDataType].[Category Data Type].&[Forecast]),[Measures].[Balance])
 
And then create a Scope Statement:
 
SCOPE ([dOrganisation].[Organisation Hierarchy].[Reporting Unit].&[BELUX HYPERION]);
THIS = [Measures].[000TD_FC_TEST];
END SCOPE;
 
And this solves some of the problem as it gives me the desire result when slicing using “Forecast”. However it messes up the solution if I put “Budget” or any other Category in the slicer, it still  gives the same results like “Forecast”.
 
Then I tried something like below:
SCOPE ([dOrganisation].[Organisation Hierarchy].[Reporting Unit].&[BELUX HYPERION]);
SCOPE ([dDataType].[Category Data Type].&[Forecast]);
THIS = ([Measures].[000TD_FC_TEST]);
END SCOPE;
END SCOPE;
 
But this one won’t work either. Any suggestions on this. Thank you!

anybody expert there to help

Hi,

Your calculated member isn't really closer to sorting your problem out - you have simply created a tuple that explicitly returns the value for Belux hyperion_ sliced by Forecast. But it won't be sliced by anything else - it's not an approach that will resolve your problem.

What I don't understand is why the nested scope solution I provided didn't work - in my recreation of the problem it worked fine.

I have replicated the problem using Adventure Works: the idea being that for June and July, the measures for 'Bikes' is set to be the same as the values for 'Clothing'. Calculation is:

SCOPE ([Product].[Category].&[1]);

SCOPE ({[Date].[Month of Year].&[6],[Date].[Month of Year].&[7]});
THIS = ([Product].[Category].&[3]);
END SCOPE;
END SCOPE;

MDX queries to check this:

For June Bikes = Clothing:

SELECT ({[Measures].[Internet Sales Amount],[Measures].[Internet Order Count]}) on rows,
NONEMPTY ([Product].[Category].[Category].MEMBERS) on columns

FROM [Adventure Works]

WHERE ({[Date].[Month of Year].&[6]})

                                       Accessories      Bikes      Clothing      Components
Internet Sales Amount      $65,200.93      $35,206.48      $35,206.48      (null)
Internet Order Count      1,723      778      778      (null)

(note that month is only included in the where condition)

For August, Bikes has it's own value:


SELECT ({[Measures].[Internet Sales Amount],[Measures].[Internet Order Count]}) on rows,
NONEMPTY ([Product].[Category].[Category].MEMBERS) on columns

FROM [Adventure Works]

WHERE ({[Date].[Month of Year].&[8]})

                                      Accessories      Bikes      Clothing      Components
Internet Sales Amount      $52,056.61      $1,824,214.67      $23,335.39      (null)
Internet Order Count      1,323      932      519      (null)

Isn't this what you are trying to achieve?

Tim

Please see attachment and reply me ASAP
mydoc.doc

Hi,

As I understand it, if you have custom rollups then they will take precedence - I don't think you'll be able to work around them by attempting to provide other calculations (as you've been attempting). You would need to modify the custom rollups themselves to 'build in' the mapping between the Belux hyperion members.

Tim

can you show how to do it please

Maybe - but I haven't used custom rollups myself. You would need to post the calculations to begin with...

Please can somebody give me the solution as i am now tired of this.

I don't think you can expect a solution if you don't post the requested information. Can you post the custom rollup code?

Please find attached code. Please change extension of attached file  to .mdx
Copy-of-Formula.txt

Looks like you will have to replicate the logic in the scope statements into these calculations.
You may be able to place these calcs inside the scope statement to simplify the checks on Forecast  and Budget but modifying these calculations will be necessary.

but cant you give me logic for it?

I'm not going to rewrite your query for you - it will take too much time. You have the logic, it's just plugging it into the existing calculations.

Thanks for your help again! I did try modifying the calculations just like I did in the Scope statement, however it looks like we cannot achieve this simply within our cube. I've got to do this at the Report Level.
 
In the scope statement we cannot specify the SubCube query (the slicer pane in the SSAS Browser). We opened up a case with MS and got the above conclusion.

ok close it