Hi all,
The data model we are creating is intended for self-serve BI, so I looking for advice on how to make the most intuitive for the users all the while avoiding that context filters be copied over and over in different measures.
We have a few equipement productivity measures that are filtered on different dimensions (material, equipment, time classification, etc.). To make it easy for the user, we created very specific measures such as "Productivity - Production Shovels - All material" vs "Productivity - All Shovels - Ore material". Now if I have many dimensions, it would mean I need to copy the context filter every time and if a change in business logic occurs, all must be edited.
Is this the only way to accomplish my goal? I'd be curious to hear what the thought process is when developping combinations of context filters for closely related to each other.
Example:
Productivity = units produced (here tonnage) / time
can be split by the following with different time filters
Gross Productivity = tonnage / operational time
Net Productivity = tonnage / active time
Here is a mock example to illustrate the filter combinations:
Gross Productivity - Material 1 - Equipment 1 = calculate(sum(tonnage)/time, filter(material = "material1" && equipmentType = "Equipment1" && time = "operational"))
I suspect I am not approaching this from the right angle and would appreciate any insights on the matter. Please let me know if my example is not clear enough, and I can create an example with mock data from each dimension.
Thank you,
Vero
