1:n relationships in reporting

Hi experts,

for a long time I'm struggeling with the following problem, without having found a suitable solution.

Maybe you can show me the way, since it should be a very common topic.

The scenario is as follows:

I'd like to see production information.

I'd like to know which quantitiy of product P has been produced and

which according amount of components C have been used for that production.

Example:

ProdOrder Product Component

4711 P1 C1

C2

ProdOrder Product Component #Products produced #Components used 4711 P1 C1 10 10 C2 10 16 4712 P2 C1 15 15 C2 15 30 C3 15 20 P1 Result wanted: 10, is: 20 (=2*10) 26 P2 Result wanted: 15, is: 45 (=3*15) 65

My problem is the result for keyfigure "#Products produced". Since there are n records for the same product (one for each component), the result-row shows: n*(#products) - which is in fact wrong. P1 was produced 10 times, but used 2 components (e.g. "bolts" and "nuts") - anyway: That doesn't extend the produced amount, it's still 10, not 20.

So, to sum it up a little bit: I want each product-record (the fist? the last? the average?) only exactly once as part of my result-sum.

I hope I could make my problem understandable.

The above example is based on an infoset (where every record contains both keyfigures).

Using a multiprovider (where each record contains either the one or the other keyfigure) would be another option, but it would lead to the issue that as soon as a filter con components is set, the product-quantity-records would be excluded.

I tried a lot with exception aggregation, nested exception aggregation, constant selection, etc. but did not find a suitable way.

Would be great if anybody of you could show up the way.

Thanks a lot in advance.
Marco