Forum Discussion
ALL function disabled when applying date slicer
Hi,
I'm new to power bi, and i'm trying to calculate the OEE in a company.
I'm supposed to calculate the availability for every equipment, then the performance and quality for every component produced and then the OEE.
In that way, i need several slicers to apply the specific filters and calculations i need.
So, basically, i need the availability to make the calculation regarding the equipment, but then ignore the selected component.
For that i use an All function im my measure to ignore component. The calculation is done correctly if i don't have the week interval selected(semana), has you can see below.
If i change the selected component the availability stays the same, as it should, but if i change the week slicer to make the calculation for a specific week or weeks of the year, then everytime i change component in the slicer the availability slicer changes as well.
I don't understand why, but for some reason the week slicer must be disabling the all function.
Here's my measure, you can ignore most of it. It just changes the calculation depending on que equipment selected.
Here's my tables with their relations:
4 Replies
- johnt75Super User
I suspect that the problem is because of the bi-directional relationships. As it is currently, a filter on any table will affect every other table. Unless you have very good reason to use bi-directional relationships you should always use single direction relationships with the one-side filtering the many-side.
- lmiguellfFrequent Visitor
The problem is that i want my filters to be filtering both tables, dados and paragens. Without the bi-directionality i would need to add the columns that i want to use in the slicers in the dim table dados_paragens. However i cannot do this because when i add them, the key column Nº Ordem that i was using to link the tables returns duplicate values and no longer works to link. Also if i create a new key column with the combinations of the slicer columns i don't achieve a unique column as well, so i have no key to link the tables.
- johnt75Super User
You could potentially combine the 2 dimension tables, e.g.
Combined Table = DISTINCT ( UNION ( DISTINCT ( Dados_Paragens ), DISTINCT ( Dados_Teoricos ) ) )and then link this single table to the fact tables, removing the other relationships. Then the relationships from the new combined table could be single direction.