Calculate permutation

[turcelaygoy]

Hello! 

I'm trying to calculate an evolution of faults calculating the average of the variations between two not consecutive cars. But I want to do a permutation and calculate the variation of each car with all the rest. I don't know if this could be possible to do or not. I attach the code and the explanation of each variable I have so far so it can give some more informartion.

Explanation of the variables on the code:

The variable PROYECTO and AREA are chosen in the data segmentations that appear in the doccument.

PROYECTO: It's the project to which the car belongs to

AREA: Is the area we want to analyze( Examples: painting area or stuctures area)

Hoja1: The sheet with all the information

ORDEN COCHE PROYECTO: It's a variable that I have created that numbers the cars chronologically so in the generateseries I can go car by car seeing how many faults have been found on each car.

FALTAS TOTALES (QA): The number of faults that have been found in each car

Code:

Quality variation_COCHE2 =

AVERAGEX (

CROSSJOIN (

VALUES ( 'Tabla Proyecto'[PROYECTO] ),VALUES('Tabla Area'[AREA])

),

VAR T =

CALCULATETABLE ( Hoja1 )

RETURN

AVERAGEX (

GENERATESERIES ( MIN(Hoja1[ORDEN COCHE PROYECTO]), MAX(Hoja1[ORDEN COCHE PROYECTO]) - 1 , 1 ),

VAR CurrentFaults =

SUMX ( FILTER ( T, Hoja1[ORDEN COCHE PROYECTO]= [Value] ), Hoja1[FALTAS TOTALES (QA)] )

VAR NextFaults =

SUMX ( FILTER ( T, Hoja1[ORDEN COCHE PROYECTO]= [Value] + 1 ), Hoja1[FALTAS TOTALES (QA)] )

RETURN

DIVIDE ( NextFaults - CurrentFaults, CurrentFaults ) * 100

)

)

[turcelaygoy]

Quality variation_COCHE2 = 
AVERAGEX ( 
    CROSSJOIN ( 
        VALUES ( 'Tabla Proyecto'[PROYECTO] ),VALUES('Tabla Area'[AREA]) ), 
    VAR T = 
        CALCULATETABLE ( Hoja1 ) 
    RETURN 
        AVERAGEX ( 
           GENERATESERIES ( MIN(Hoja1[ORDEN COCHE PROYECTO]), MAX(Hoja1[ORDEN COCHE PROYECTO]) -1 , 1 ), 
           var Z= [Value] 
               AVERAGEX( 
                  GENERATESERIES ( Z , MAX(Hoja1[ORDEN COCHE PROYECTO]) - 1 , 1 ),
                    VAR W = [Value] 
                    VAR CurrentFaults = SUMX ( FILTER ( T, Hoja1[ORDEN COCHE PROYECTO]= Z), Hoja1[FALTAS TOTALES (QA)] )
                    VAR NextFaults = SUMX ( FILTER ( T, Hoja1[ORDEN COCHE PROYECTO]= W + 1 ), Hoja1[FALTAS TOTALES (QA)] ) 
    RETURN 
         DIVIDE ( NextFaults - CurrentFaults, CurrentFaults ) * 100 )
 ) 
)

I came up with this but I would like to know if there is any option of making the variables Z and W work. Right now the measure doesn't work due to those two variables.