Creating a Pareto from an unpivoted table

[MiusYan]

Hello all, I'm creating a dashboard to track output/yield and pareto chart,

This is the original set of data that I'm connected:

Based on that data, I created some slicers in my report to filter by Manufactruring date -> Product Description -> Lot Number -> Process. This filters help me generated a "genaral table" just for information, an output/yield Line and clustered chart and a cumulative output/yield cart. 

Then to generate the Pareto chart I created a duplicate to the original table and unpivote it (this was needed because each defect was listed in an individual column and I needed to have them like in the format below):

Then in the model view, I create a relationship between both of the tables (Process - Process, Many to Many, single cross filter direction).

Then I generate the pareto chart using the duplicated table (Y axis is the sum of the values of each individual defect and X axis is the defect name). 

When I start filtering the slicers to get the papreto chart for an specific dates, an specific product, lot number and process I noticed that the information shown in the pareto is the sum of the values of each individual defect in that time frame, but is not taking into account the "lot number" slicer selection.

How can I solve this?

Thanks a lot! 

[MiusYan]

I will add more context, this is an example of the original table at the data base I'm using:

And this is the unpivotted table (duplicated) I created base on the original table:

Then I created the relationship between them, but so far I dont have the result I need. that is a pareto chart that can be filtered by each lot number in an specific time frame.

Thanks

[amitchandak]

@MiusYan , if you are creating a Pareto chart, make sure the X-axis column is coming from dim, so that you can easily use all and allselected

Cumm Based on Measure = CALCULATE([Your Measure], Window(1,ABS,0,REL, ALLSELECTED('defect'[defect name ]),ORDERBY([Your Measure],Desc)))

or

Cumm Based on Measure = CALCULATE([Your Measure], Window(1,ABS,0,REL, ALL('defect'[defect name ]),ORDERBY([Your Measure],Desc)))

Also, make sure all filtering dimensions are joined with your new unpivoted table

Pareto Analysis Again, 80% of sales, Order by Measure when REL position is used: https://youtu.be/GpoITi_tRIw