Skip to Content

SAP Lumira - fix data set with weekly sales data in columns ( time dimension "hidden" in columns).

Mar 07 at 04:53 PM


avatar image

Well, I thought I just registered for the community discussion board. Maybe not. So sorry if this is the wrong place. Also, I am pretty new to SAP Lumira, but I have good experience with relational databases

I have 3 very large data sets. The are the same structure but one is for unit price, unit cost, and unit sales. There is no date dimension as you are used to in Lumira. The column headings are essentially the date dimension, but I don't think I can tell Lumira that. For example, the column labeled 1 is for week one unit cost, price, or sales data for over 180,000 items that I can group by product, store, or category. So on for the column labeled 2. Alternatively, I could label these columns 1/4/14 and 1/11/14, but I don't think that helped any.

So there is still no real date dimension. I tried to get around this by simply dragging all columns 1 through 52 as y-axis measures. This workaround makes it look just like, say unit sales by whatever across time. But then if I join the price dataset to get actual revenue it appears that I would then have to make 52 custom calculated fields. That's too much. And I would have the same problem for any other calculations.

So, I hate to have to reinvent the wheel here. Although this was not a very good way for whoever to store data like this, it is what I have. And, It seems likely that this is not an uncommon way to record sales unit data.

I have searched hundreds of links and youtube and all that I find are examples of nice and clean data sets that already have a a true date dimension.

So any assistance on preferably how to fix this data set from within SAP Lumira would be great. Thanks.

Edit: Help -> About says Lumira 1.31.6

I guess it is desktop version. It doesn't say "desktop", but it is installed on a university computer.

Sample of data set is:

ID, Category, Item, Store, 1, 2, 3, 4

AL-0-0, DAIRY, 0, 0, 2, 4, 5, 8

AL-0-1, DAIRY, 0, 1, 2, 1, 2, 2

This is unit sales. Again, field "1" represents sales for week ending 1/4/14. Then you just keep going a week.

I don't necessarily need someone to do all the work for me. Just any tips on how to manipulate / re-structure data in Lumira would be helpful. I could do this easily with code in Python or something, but was hoping to do it in Lumira. Thanks.

10 |10000 characters needed characters left characters exceeded

Hi James, are you using Lumira Desktop, Discovery, or Designer? Which version? What is your data source?

* Please Login or Register to Answer, Follow or Comment.

0 Answers