Hi
I have fact table with info concerning a helpdesk (service request - SR). I want to analise the helpdesk service performance issues
I want to be able to define a objective each year like : my response target will be to respond in less than 2 hours, and i want to hit that value 90% of the times. These objectives can change over time

I have defined a a dimension for the "request date", a measure for "the time to respond", another for the "total of requests" (all measures with sum as agregate function)
What kind of measure or dimension do i need to create to calculate if each request was done "within time"!!?

If i could have a measure with the sum of "requests within time" that would be suficient.
But if i do that in sql(before the cube processing) and if i change my goal say for (i need to respond in less that 1 hour) all my historic data will be lost

Can anyone help? what is the best practice for this kind of situations

thank you very much for the time spent
Best regards,
Rui Barbosa

What is the size of the fact table, and of the dimensions? In theory,
you could roll up the service level from the individual fact records (if
there was a dimension that drilled down to the individual ticket); but
this wouldn't work well with large data sets, unless you use 64-bit AS.


- Deepak

*** Sent via Developersdex http://www.developersdex.com ***
Don't just participate in USENET...get rewarded for it!