Forum Discussion

manjirit's avatar
manjirit
Helper I
8 years ago

Need Help with Rankx

I have a problem and can't seem to find whats going wrong!   I am using Rankx() function to rank the rows in the resultset table.   The RankX function gives two distinct ranks to the rows viz. 1 ...
Need Help
manjirit's avatar
manjirit
Helper I
8 years ago

Thanks!

 

Here are the pictures of resultsets. I have masked some columns for confidentiality reasons. 

 

I am giving here an example of rank based on column called "Unmonitored Assets".

 

Here is how Unmonitored Assets (a measure) is calculated:

 

Unmonitored Assets = CALCULATE(DISTINCTCOUNT(TableName [AssetName]), TableName [Excepted_Assets] = 0) + 0 --> filtering on column Excepted_assets = 0 and  getting the distinct count for those assets.

 

The rank measure is as follows: 

Rank = RANKX (all(TableName),[Unmonitored Assets])  -- > I want to rank the table based on number of "Unmonitored Assets".

 

Here is what I see: 

 

Rank = 1 for all Unmonitored Assets >0 

Rank = 160001 for all Unmonitored Assets =0 

 

What is going on, anyone?

 

 

Greg_Deckler's avatar
Greg_Deckler
Community Champion
8 years ago

Need a sense of your source data. In that table, do [AssetName]'s appear once or multiple times? I really think that you need to do a SUMMARIZE of your table based on some grouping of which I am not clear about yet and then do an ADDCOLUMNS to add a rank column as shown in previous reply.