Rankx with two subcategories

please try 

Rankx with 2 categories =
CALCULATE(
RANKX(
CALCULATETABLE(
ALL(a_notifi_list),
a_notifi_list[format date_yyyy.mm] = SELECTEDVALUE(a_notifi_list[format date_yyyy.mm]),
a_notifi_list[Plant for material /plant number] = SELECTEDVALUE(a_notifi_list[Plant for material /plant number])
),
a_notifi_list[Defective (external)],
,DESC
),
ALL(a_notifi_list)
)

I try to describe my case a bit:

I would like to rank an amount of defects within 3 columns: month, plant and vendor column.

That means to show the rank by defects for all vendors for each plant and each month. Can't find a working solution for this for a long time. Is it possible to do it in a custom column? Because i need to count all the vendors with rank 1-5.

My idea is to create an additional column and set for every rank 1-5 the value "1" and for all the other ranks the value "0" and then just sum all the "1" values for every vendor. Just to know how often a vendor was in the top 5 ranking in every plant for every month.

Now, i would like to rank all the suppliers by amount of defects for every month and plant (in a common measure or column).

something like this:

The final result is to know often each supplier was in top five ranking (in the picture is top3) in the last 12 months and how often is every factory affected by this.

final report visual is this:

y-axis : vendors listed

x-axis amount of being present in top five ranking last 12 months (in every month)

legend: factory A, factory B, and so on

I hope it's clear a bit more now.