I've done a bunch of Google searches and can't seem to find a question that mirrors what I'm trying to do (if it's even possible).
I have a bunch of inventory snapshots that are stitched together in the folder data source with many more to come. Each file has a column with the date of the export repeated through each line which typically serves as the X-axis so I can trend each data report.
Where I got stuck is how the inventory reports I'm dealing with come out and the data I need to represent. Within the report there are six different columns that each represent different inventory segments (things like in-transit between warehouses, off-quality goods, etc...). The problems comes because they can have inventory in multiple buckets, so I can't just do some logic to look at those columns and say "Product A is in transit". Product A may have some inventory that is being put into a shipment, some that's off-quality, some that's in-transit, etc... So bottomline, the columns need to remain and I can't just add them all together and say "this is the total inventory for Product A". I'd like to trend this so we can see as a whole how our good inventory is trending, how our inventory being put into shipments is trending, etc... I thought the perfect tool for that would be a stacked area chart, but that wants one dimension for the magnitude of the bar and one dimension to bucket the row into each category of the legend. What I'd like to do is basically have each column be an element of the legend with the value of the column being a component of the magnitude of that bar.
I hope I conveyed what I'm trying to do.
Thank you for any help!
