Hi to everybody.
I explain you this big trouble:
I have a really big dataset with a lot data.
After loading the script I have a really big table that has different fields.
The relevant for what I am not able to do and give me problems are the following:
BigTable:
ID
DateA
TimeA
DateB
TypeOfDefect
...Other fields...
But this table results after a join in the script between the followings:
Table1:
ID
DateA
TimeA
TypeOfProduct
...Other fields...
Table2:
ID
DateB
TypeOfDefect
... Other Fields ...
Table 2 is a subset of Table 1 in terms of ID: i.e. all IDs in table 2 are also in Table 1 and not viceversa.
After exhausting trials and errors, I could not be able to construct some table charts, using Set Analysis, doing the following:
When select one or more DateB, I need to count all IDs whose DateA is equal to DateB.
I explain better: DateB represents the date when a unit is classified as defective, while DateA is the date when the unit is subjected to process A.
I need to count, for each TypeOfProduct and for each TypeOfDefect (in two different table charts of course), the number of defectives (i.e. when TypeOfDefect is not null)/the number of ID which was subjected to process A (i.e. I suppose count Time A or Date A).
I tried with set analysis, but it was impossible for me to find a way to compute this, when I select as filter a DateB, I need that the second part of the ratio counts the IDs subjected to process A, but obviously this date must be DateA=DateB. The problem is DateB in null when the unit is not defective. So the result is that it counts only the A-processed which are defectives, not the good ones.
Given that it is a KPI ratio, I need for example, if I select DateB=12th January 2017, I need to count all the defectives defined in that date, divided by all the IDs subjected to process A on the 12th of Jan. 2017.
Thank you very much for the help.
It is really important.
Bye
