Sum At A Lower Level Of Granularity

Hi all,
I have a big problem with the design and the queries of a couple of tables. I have to calculate the weighted average of the number of days between an invoice and the sum of its payments (rateal payments), weighting this number with the invoice's amount, and only when the sum of the invoice's payments equal invoice amount.
I have designed two FactTables to accomplish this:

Example:
data in the fist table (FactInvoice) looks like this:

CustomerId, InvoiceDate, InvoiceNumber, InvoiceAmount
1234567 2008-03-10 123 1000.00
1234567 2008-04-10 150 2000.00

and data in the second one (FactPayments) looks as below:


CustomerId, PaymentDate, InvoiceNumber, PaymentsAmount
1234567 2008-03-10 123 500
1234567 2008-03-15 123 500
1234567 2008-04-20 150 800
1234567 2008-04-23 150 1200




Let's suppose that I'm querying the cube for the customer 1234567, at the date 2008-05-01. I need to sum the payments for the first invoice multiplicating it for the count of days elapsed, then divide the number by the InvoiceAmount (or PaymentsAmount, is the same) : (0 days * 500‚¬ + 5 days * 500‚¬ + 10 days * 800‚¬ + 13 days * 1200)/3000‚¬ = 8,7 days of weighted average to cash an invoice.

How can I do this with Analisys Services ? Is it too complicated ? Here, in the southern Italy, the problem of customer's debit is heavily felt, and total like this are really important!
Any help or suggestion will be really appreciated.

Marco