Hi,

I am a newbee to OLAP and MS Analysis Services. I have a cube with Time and
Client dimensions. Measures include shipment amount and payment term.
Payment term is an integer number meaning 'should be paid within N days from
shipment date'.

I need to create an MDX query to sum up shipments for which a) payments are
_not_ overdue and b) payments _are_ overdue - in columns '0 to 3 days', '4
to 7 days' and '7+ days' from the shipment date.

My questions are:

1) how do I create a filter which would give me only the shipments for which
payments are _not_ overdue
2) how do I break up the shipments in two columns

I thought that each column should be a calculated member summing up the
data, something like:

with member '0-3' as "Sum ( LastPeriods( 3, Time.CurrentMember. ),
[Measures].[Shipment Amount] )"
with member '4-7' as "Sum ( LastPeriods( 3, Time.CurrentMember.Lag(3) ),
[Measures].[Shipment Amount])"
with member '7+' as "Sum ( PeriodsToDate( Year, Time.CurrentMember.Lag(7) ),
[Measures].[Shipment Amount])"

But I am having a hard time creating filter statement to separate shipments
with overdue and non-overdue payments.

Any ideas would be higly appreciated!

Thanks,

Denys Kotseba

the easy way is to create a dimension which is "delay" linked to your
payment term fact table column.

This new dimension will be:
Overdue Group Key
No 0 0
Yes 1-3 1
Yes 1-3 2
Yes 1-3 3
Yes 4-7 4
etc...


"Denys Kotseba" <kdl2001 @ ukr.net> a écrit dans le message de news:
1094903004.42171 (AT) embrayer (DOT) merlin.net.ua...