Find period with last x incidents

[Run_VK]

Hey all

Been working with PBI a couple of years and can often find similar issues in here that helps me - but this time I'm stuck... So hope some one is able to help

I have three tables

Case-table

Absense-table

Calender

In the case-table I have a measure that counts the days that each case has been open. In the absense-table I have a similar measure that counts the days of absense

There are no relation between the calender-table and the two others

Date	ID	absense	case
01-01-2023	22		1
02-01-2023	22		1
03-01-2023	22
04-01-2023	22	1	1
05-01-2023	23	1	1
06-01-2023	23		1
07-01-2023	23
08-01-2023	23	1	1
09-01-2023	23		1

For each ID I need to find the period that contains the last 365 days where the case has been open. The period can be longer than one year, as there can be days where the case is not open. At the same time I need the sum of absense-days in the same period, and divide these.

For exampel: The period containing the last 365 case-days are 1/1-22 to 1/3-23 (425 days). In that period there has been 90 absense-days = 90/365... 25%

Any ideas how to solve this with Dax

Thanks

R_

[lbendlin]

Please provide sample data that covers your issue or question completely.
Please show the expected outcome based on the sample data you provided.

https://community.powerbi.com/t5/Desktop/How-to-Get-Your-Question-Answered-Quickly/m-p/1447523

[Run_VK]

I hope this exampel provides the nessesary information

Wetransfer 

[lbendlin]

The case days computation is relatively trivial

Case Days = 
var b = ADDCOLUMNS(Calender,"case",if(SELECTEDVALUE('case'[End date])>=[Date] && SELECTEDVALUE('case'[Start date])<=[Date],1,0))
var c = topn(365,filter(b,[case]=1),[Date],DESC)
return sumx(c,[case])

However the Absense days computation is tricky because DAX still does not have a UNIONX operator. (Please vote for the idea).  Still  thinking.

[Run_VK]

Thanks - the case days computation works. Keeping the post open and hope someone have an idea to solve the other part 🙂

[lbendlin]

Does it have to be DAX or would Power Query be acceptable?

[Run_VK]

Power Query could be an option

The case start date and case end date could also be a part of the absense-table - if that makes it possible solving it in dax?

[lbendlin]

No, that won't help.  Only UNIONX would help.

Here's the Power Query version.

let
    Kilde = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("i45WMlTSUTI00DUw0jUyMDICcoyAHBMYxzGpODWvOFUpVgei0sBQF6gYKgnSZgjiGAM5AYnFxZllEJVgSVMkY4AcQyM8ZkKMAXGMgRxjGAeuMhYA", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type nullable text) meta [Serialized.Text = true]) in type table [ID = _t, #"Start date" = _t, #"End date" = _t, Kolonne1 = _t]),
    #"Omdøbte kolonner" = Table.RenameColumns(Kilde,{{"Kolonne1", "Type"}}),
    #"Ændret type" = Table.TransformColumnTypes(#"Omdøbte kolonner",{{"Start date", type date}, {"End date", type date}}),
    #"Added Custom" = Table.AddColumn(#"Ændret type", "Range", (k)=> List.Generate(()=>Number.From(k[Start date]),each _<=Number.From(k[End date]),each _+1)),
    #"Expanded Range" = Table.ExpandListColumn(#"Added Custom", "Range"),
    res = List.Difference(Table.SelectRows(#"Expanded Range",each [Type]="Absense")[Range],Table.SelectRows(#"Expanded Range",each [Type]="Passive")[Range]),
    #"Converted to Table" = Table.FromList(res, Splitter.SplitByNothing(), null, null, ExtraValues.Error),
    #"Changed Type" = Table.TransformColumnTypes(#"Converted to Table",{{"Column1", type date}}),
    #"Renamed Columns" = Table.RenameColumns(#"Changed Type",{{"Column1", "Date"}})
in
    #"Renamed Columns"

[Run_VK]

Hmm - maybe I don't get it 😕

The query ends with one single date collum - how can I achieve the result per ID with these dates?

[lbendlin · Accepted solution]

you can modify the query to produce date lists per ID. Then you can use these to intersect with the case dates for each ID.

It was a proof of concept of implementing UNIONX in Power Query.

[Run_VK]

Aah - okay. I will give it a go

Thanks for all your help Ibendlin