Skip to main content

Hi all, we report the Automation Rate with Eventtime as time dimension.

We were expecting, that the Automation Rate therefore should be stable e. g. for the past year 2022 regardless whether we extract the data in March or in June. Unfortunately we have changes even though we did not change the measurement.

We have deviations in the number of activities as well as the number of cases.

 

Formula:

AVG(CASE WHEN "_CEL_O2C_ACTIVITIES"."ACTIVITY_<%= language %>" IN

(<%= whitelist %>)

THEN

NULL

WHEN ISNULL( "_CEL_O2C_ACTIVITIES"."USER_TYPE") = 1

THEN

NULL

WHEN "_CEL_O2C_ACTIVITIES"."USER_TYPE" IN (<%= auto_user_type %>)

THEN 1.0 ELSE 0.0

END)

 

Remark: We have not changed the whitelisted activities nor the user types included in the variable auto_user_type.

Any ideas? Thanks in advance

 

Hi Silke,

 

could you please provide me with the Formula of the dimension Eventtime? I assume maybe it is a Problem of referencing the wrong Eventtime, which could distort the automation rate


Hello Rashid,

 

the formula is ROUND_MONTH("_CEL_O2C_ACTIVITIES"."EVENTTIME")

 

It is a column of the activities table.

 

imageView from the attribute selection:

imageThanks

Silke


Hello Rashid,

 

the formula is ROUND_MONTH("_CEL_O2C_ACTIVITIES"."EVENTTIME")

 

It is a column of the activities table.

 

imageView from the attribute selection:

imageThanks

Silke

Hello Silke,

 

thanks!

The Problem with this Formula is, when we have a case that has some activities in one month (that don't have the automation criteria you provided: probably end of a month) and the rest on an another month(which have the automation criteria:probably beginning of a month) so celonis get the eventtime of the false activitiy (first part).

Solution: Try this PQL instead:

ROUND_MONTH(TIMESTAMP_COLUMN ( CASE WHEN

Condition 1 when is automated

AND

Condition 2 when is automated

.

.

.

THEN

"ACTIVITY_TABLE"."EVENTTIME" END ) )

 

I hope that helps :)


Reply