Transitions from status counted multiple times when using calculated member aggregate function

I have a problem calculating median days in transition status. I use this custom measure:

Descendants([Issue].CurrentHierarchyMember, [Issue].[Issue]),
([Measures].[Transitions from status],[Time].CurrentHierarchy.DefaultMember) > 0
[Measures].[Days in transition status])

Then in columns I also add “Transition status” so I can selecte a specific Transition status.

Some issues are calculated with a value which is a multiple of the actual “Days in transition status”. the factor is the same as the value of “Transitions from status” measure.

It looks like this behaviour is cause by the use of such a “calculated member”


What I want to accomplish with the calculated member is to get all Issues in Project P1 which have component API or Web,
But now in my measure “Days in transition status” is summed up not per issue, but by issue/component.

Can I build the “calculated member” in another way so the measures are not summed up?

After changing the calculated measure as shown below the result looks OK now.


  1. take median of all “Days in transition status” (issue can be more than once in transition status)
    -> ( [Measures].[Days in transition status], [Time].CurrentHierarchy.DefaultMember )
  2. use measure only for period when issue was the last time in transition status
    -> see Filter
  3. do not sum up if report filter selects multiple components and issue is linked to this multiple components (this is the problem why I created the post initially)
    -> solved by divide with ( [Measures].[Transitions from status], [Time].CurrentHierarchy.CurrentMember)
          [Measures].[Transition from status last date],
          [Transition Status].CurrentMember,
    ( [Measures].[Days in transition status], [Time].CurrentHierarchy.DefaultMember ) /
    ( [Measures].[Transitions from status], [Time].CurrentHierarchy.CurrentMember)

The solution works, but the performance is very bad. Is there some better (more performant) way to get the same result?