Forum Discussion

Pr0ton's avatar
Pr0ton
Frequent Visitor
2 years ago
Solved

Production Time Analysis - Calculate Interval between sequenced Transactions.

I would like to calculate the interval between Transactions. See the example table below. I'm looking for some trends on 'intervals' or 'waiting times' , and would like to look at a years wort...
Need Help
  • amitchandak's avatar
    amitchandak
    2 years ago

    Pr0ton ,

    Create the following column

    Rank = Rankx(filter(Table, Table[UID] = earlier([UID]) ), [Start Time])

     

    diff = if([Rank] <>1, datediff(Maxx(filter(Table, Table[UID] = earlier([UID]) && Table[Rank] = earlier([Rank]) -1), [Start Time]) , [Start Time], second) )

     

amitchandak's avatar
amitchandak
Super User
2 years ago

Pr0ton ,

Create the following column

Rank = Rankx(filter(Table, Table[UID] = earlier([UID]) ), [Start Time])

 

diff = if([Rank] <>1, datediff(Maxx(filter(Table, Table[UID] = earlier([UID]) && Table[Rank] = earlier([Rank]) -1), [Start Time]) , [Start Time], second) )

 

  • Pr0ton's avatar
    Pr0ton
    Frequent Visitor
    2 years ago

    Thanks All,

     

    I think this is now  resolved. 😀

    Thanks to amitchandak 

    1. Removed :- Schoolboy error in the data query which had unwanted 'almost duplicate' lines which prevented the Rank from counting  correctly.

     

    2. Tweaked the Datediff so that it uses   [Completed Time] as Date1 , and  Ranked-1 Earlier  [Start Time] as Date 2 (which is actually the subsequent transaction as the ranking increments as it gets younger)

     

    @gmsamborn  Unfortunatley the Measure was too resource hungry for this dataset, but yes, on the sample you generated, works . Much appreciated for your time.

     

    Many thanks again.