Forum Discussion

Anonymous's avatar
Anonymous
Not applicable
5 years ago
Solved

Pareto chart with tie values

Hello Community  -  

 

I have a data set where I am tracking the number of late lines, per product.   And doing a pareto of these items.   A number of items have the same amount of late lines.   Meaning, the order line the product was on shipped late.    So for example, in the chart below you can see that a number of different items have the same amount of late shipped lines...resulting in a flat line where the ties occur.     I guess my question is whether or not this is acceptable in the world of pareto charts?    Or should a pareto Always have the traditional curve to it?  

 

The conditional formatting (grey color) kicks in at the 80% value on the pareto blue line.  

 

Need Help
  • Fowmy's avatar
    Fowmy
    5 years ago

    Anonymous 

    It seems your data is not the same granularity as mine, I added a dimension table for the Products and repeated the fact table lines. Please the attached file. Created a new table and added new measures. Please check.

     

15 Replies

  • Fowmy's avatar
    Fowmy
    Icon for Super User rankSuper User
    5 years ago

    Anonymous 

    Ideally, in a Pareto chart, each value should get added and that should be calculated as a % of the total. I think you need to look at your calculation to avoid getting similar percentages when they the same.


    • Anonymous's avatar
      Anonymous
      Not applicable
      5 years ago

      Fowmy     Makes sense, but where you have repeated values, what I am getting is the same cumulative percent.   For example.   So you are saying I should have a different cumulative percent even if I have the same value for the late shipped lines?

       

      # of late lines      cumulative percent 

       

      6                              33.85%

      6                              33.85%

      6                              33.85%

      7                              28.69%

      7                              28.69%

       

       

       

      • Fowmy's avatar
        Fowmy
        Icon for Super User rankSuper User
        5 years ago

        Anonymous 

        When it comes to ties, you need to incorporate another column to break them, I did a sample file with Pareto using this technique. Please find attached the file below my signature. Product B and C have the same values but still adds up.

         

        Pareto =
VAR __CurrentRank =
    RANKX (
        ALL ( Table1[Product] ),
        CALCULATE ( [Total Sales] + UNICODE ( MIN ( Table1[Product] ) ) / 1000 ), , ,
        DENSE
    )
VAR __Total =
    CALCULATE ( [Total Sales], ALLSELECTED () )
VAR __Cummulative =
    SUMX (
        FILTER (
            ADDCOLUMNS (
                ALLSELECTED ( Table1[Product] ),
                "rank",
                    RANKX (
                        ALL ( Table1[Product] ),
                        CALCULATE ( [Total Sales] + UNICODE ( MIN ( Table1[Product] ) ) / 1000 ), ,,
                        DENSE
                    )
            ),
            [rank] <= __CurrentRank
        ),
        [Total Sales]
    )
RETURN
    DIVIDE ( __Cummulative, __Total )
  • Syndicate_Admin's avatar
    Syndicate_Admin
    Icon for Administrator rankAdministrator
    3 years ago

    I was able to reproduce your solution, but the data zeroed out (which has values >1 in other months) is showing up on the graph. I use filters for different months and periods (quarter and semester) to generate periodic reports, and I intend to use this same base. How can I hide them?