Hi Steven,
There are multiple solutions to calculate the throughput times. The first solution calculates the throughput time between events and is explained here:
https://help.celonis.cloud/help/x/7IQeAg
.
Having a look at your data (especially the EBAN table), your proposed solution would be better. The formula then becomes:
DATEDIFF(dd, PU_FIRST(<parent table>, EBAN.Lieferdatum), PU_LAST(<parent table>, Actiontable.EVENTTIME, Actiontable.Anderungskennzeichen LIKE "Bestellnummer"))
You see that there is a need for a parent table, I expect that that will be the name of your case table in this case. Hopefully this does the trick.
Bests,
Jan-peter
Thank you very much.
Unfortunately, it still doesnt work properly.
You are right. My Case Table and so my parent table is EBAN.
It seems that the Problem is the PU_FIRST statement.
Here my real formula:
DATEDIFF(dd, PU_FIRST(EBAN,EBAN.FRGDT), PU_LAST(EBAN,T_CEL_AT_HP3.EVENTTIME,T_CEL_AT_HP3.Kurzbeschreibung LIKE Bestellnummer))
If I separate the formulas I get an error for the first part PU_FIRST (EBAN, EBAN. FRGDT).
image489619 8.89 KB
mhhhh
Hi StevenB, The following approach may work for you. I was able to apply it in an invoice analysis using Invoice Header and Event Log. It is similar to the formula you shared, except that there should be no need for the PU_FIRST on EBAN.
DATEDIFF(DD, EBAN.FRGDT, PU_LAST(EBAN, T_CEL_AT_HP3.EVENTTIME, T_CEL_AT_HP3.Kurzbeschreibung=Bestellnummer))
Hello. Thank you for your answer.
I think we need PU_FIRST here.
If you look at my screenshot above, then I have several start dates per BANF in the EBAN table. But I need the first one. In my example 01.01.2020