Initial load of inventory level from csv - double datarows in query

[Former Member]

Hello everybody,

a query result shown in a web browser seems strange to me and I would be very glad, if anyone can give me some advice how to solve the problem. As I do not think that it is related to the query, I posted it into this forum.

The query refers to an InfoCube for inventory management with a single non-cumulative key figure and two other cumulative key figures for increase and decrease of inventory. The time reference characteristic is 0CALDAY. The initial load has been processed reading from a flat file (CSV), the structure looks like this:

Product group XXX

Day 20040101

Quantity 1000

Increase 0

Decrease 0

Unit ST

The initial load runs fine, the system fills all the record sets into the InfoCube. Unfortunately I do not know how to look at the records written into the cube, because only the cumulative key figures are shown in InfoCube-> Manage-> Contents.

Well, when executing the query, a really simple one, the result is just strange, because somehow there are now two rows for each product group with different dates, one with the 1st of January, 2004 and the other for the 31st of December, 2003 containing both 1000 units. The sum is 2000.

It became more confusing, when I loaded the data for increase and decrease: now the quantities and sums are correct, but the date of the initial load is a few days later than before, the data table in the query does not contain the 1st of January.

Does anybody know, what I did wrong or where there is information about how to perform an initial load of inventory from csv in a better way?

Kind regards

Peter

[Former Member · Accepted solution]

Hi,

just one question to be sure:

have you mapped your initial qty. column to the update rule for the "increase" key-figure or not? Otherwise the upload will in my opinion not work at all...

If so, the following steps are necessary:

- initial load

- compression of infocube with marker update

- re-load of hoistorical movements

- compression of infocube without marker update

That shall do it....

Andreas

[Former Member · Accepted solution]

Hello once more,

I would like to thank you both, Andreas and Aneesh. The compression of the InfoCube as Andreas described and as I found in the SAP PDF "How To Handle Inventory Management Scenarios in BW" (never thought compression would have to do anything with the problem in my case) solved the problem. Now everything is fine!

By the way, the system proposed to map the inital quantity column to the increase key-figure, so I did it as you supposed to.

Thanks again an have a nice weekend!

Peter

[Former Member]

Peter,

Inventory is not that straight forward to evaluate as it is non-cumulative. Basically it means that one KF is derived from one/two other KFs. You cannot see non-cumulative KFs in manage infocube.

Have you uploaded opening balances separately? If so, your data for 31st of december is explained.

In non-cumulative cubes, there need not be a posting for a particular day for a record to exist. For e.g. if you have stock as 10 units on 1st and then no posting for 2nd and 3rd and then increase 10 units on 4th, even for 2nd and 3rd, the non-cumulative KF will report as 10 units (stock on 1st rolled forward).

There is a how to...inventory management document on service market place that explains this quite nicely.

Cheers

Aneesh