General Product Q&A

I have 2 columns in which I am looking to find the difference for. Since some values of each column contain a null they are not reflected accurately for exampleColumn 1 Column2 Diff2 1 14 NULL NULL (should be 4)NULL 5 NULL (should be -5) I have the formulas associated with getting values in Column 1 and Column 2 within saved formulas as "KPI(.......)". My initial idea was to substitute NULL as 0, but when I surround KPI(...) with a CASE WHEN it fails saying that CASE is not supported with INT or STRING Any ideas?

I have a skill that add data table, run ML script that update some other tabls and reload the data modal but updates on other table are not persisted in data modal. My assumption is that action engine do not wait for ML script to complete rather it just start it and move to subsequent steps.  Any thoughts, suggestions on this scenerio ? 

Please see the screenshot showing sample data, we want to show waste transfer movements which go out of a source storage unit (Movement type 319) to a destination (Movement type 0) and then return. I can make the PU function work well when there is one waste movement however there are many times that the stock is moved out and back, and out and back and I can't make Celonis line up the 2nd out and back movements

Hello everyone, iam wondering if there is any way to sort the entries in a button drop down? i.e. when loading dates, displaying the most recent dates first instead of the oldes dates. Many thanks!

I have a formula stored in a variable that is giving me an accurate result of what I need to get when I check via static value based on a date filter I am putting in. However, when I pull it to my OLAP table it is showing me another value even though the same date filter is applied? Example: 83000 is stored in <%=Var_1%>. When I apply the date filter is changes to 17000 (which is correct), but when I try to store the same value as a KPI in my OLAP table it still shows 83000

Column 1 has values where the KPI is as such for a company 1 (after clicking calculate shares in percentage checkbox):SUM(Days related to certain case numbers)/COUNT(DISTINCT (No. of cases))  Column 2 has values where the KPI is as such for a certain company 2 (after clicking calculate shares in percentage checkbox):SUM(Days related to certain case numbers)/COUNT(DISTINCT (No. of cases))  Column 3 has the difference Based on what I saw the values for column 1 and 2 are read in correctly, but the difference in column three is wrong Any suggestions on this would really be appreciated

I am trying to get the percentage value of the no. of cases in a happy path by dividing by the total no. of cases for that process and display in a single KPI. Any suggestions/formulas on how I can do that?

I followed the SAP Extractor documentation. It is very helpful and detailed. However, there is one thing I noticed isn’t mentioned: While installing the SAP Extractor there is a screen: Configure the Agent UplinkIn there are two fields:App KeyAgent Name These aren’t mentioned in the Docs. Are these specific to my Celonis team or to the SAP installation?

As we know, Process Mining has the tremendous advantage of providing an objective and realistic view on the processes, as opposed to the traditional BPM analysis, which is very dependent on human judgement and prone to bias.However, building a process model from an event log is not a straightforward task. It requires an algorithm, and algorithms are based on assumptions have to somehow deal with things like noise. Conformance checking can also be done in different ways. Therefore, to better understand my results, I would be very interested in learning the algorithms used by Celonis. Thanks!

Hello everyone, the goal for my master thesis is to analyse and simulate different process variations. After identifying weak points, the process should be adapted and a simulation should take place. Is it possible to simulate different process variants in the academic version of celonis?  Thanks in advance!

Hello all,I'm looking for a way to search for certain activities without having to show all the variations and then read over them with my eye. An example from P2P process:I search "Delete Purchase Order Item". But I don't know from which variant it becomes visible in the explorer. How can I search/find something like that? 

Based on the example table below:I am getting the values in 'Column_2' by using PU_COUNTHow would I go by converting each value in 'Column_2' to a percentage amount?

Hello everyone, I wonder that when I create a skill with Manual Sensor, can i schedule trigger of this skill ? For example I want to trigger this skill once in a month automatically . If I can't is there any differenw way to do that Thank you 

Is it possible to use SQL Procedure in data transformations? If not, is it on the roadmap? Thanks!

I tried something along the line of( PU_COUNT(target_table, source_table.column))/SUM(PU_COUNT(target_table, source_table.column)) But I got an error saying:"The aggregation function SUM cannot be used together with a dimension function input. Please check that there are no aggregations and dimensions used together as function inputs"

Hello.I'm currently testing the ServiceNow Connector for IT Service Management - Incident Management. Unfortunately, the model cannot be loaded: Data model load failed during table export: Data model tables: _CEL_INC_ACTIVITIES, sys_user_group, BACKLOG, incident contain more rows than the allowed maximum per table of 0. Please reduce the number of rows. The data jobs ran errorless.Any idea what is wrong?Best regards,Heiko

Hi!! I'm interested in filtering an activity out of an analysis. When I try to filter out the activity from the attribute selection, the activity still shows in the Process Explorer. Is there a way to do a crop selection type filter in the load script?

When user click on this single number KPI it should lead to another tab?

According to the documentation the API GET /api/v1/data-push/{poolId}/jobs/  Should return the data jobs contained in the Data Pool And according to the below screenshot the data Job ID should be in the position of the 2nd box in the image attached:   Source: Help resources, section: Data Job Execution API  When we run the API above, it returns Extraction tables, not Data Jobs, and the data in the response has nothing to do with the actual data jobs we have in our team.   Can you please explain what the API above returns? Also, when we try to delete a data job it drops in the database and doesn’t show in the GET requests sent after that. However, this doesn’t reflect in the interface on the system.  Please Advise.     

As dimension I have: PU_FIRST("case_table", "activity_table"."activity_time")The KPIs I am plotting/I want to plot are:throughput time: CALC_THROUGHPUT(ALL_OCCURRENCE['Process Start'] TO ALL_OCCURRENCE['Process End'], REMAP_TIMESTAMPS("activity_table"."activity_time", MINUTES))AVG of throughput times as a constant function:If I use here AVG(... throughput time ....), then (not so surprisingly) Celonis computes these for the single cases in the "dimension set", i.e. returns the single throughput times --> is not what we wantclearly the same happens if I set up a variable with text/replacement "AVG(... throughput time ....)" and call this when defining the KPI in the plotif I set up a variable with a static value being the PQL query AVG(... throughput time ....), then I get a constant function, but it does not filter with selections, i. e. the value of the constant function stays the same independent

I have the following case:A Variable Input component is configured to write to a variable called "Target"AN OLAP table with one of the KPIs calculating the difference in days between the actual throughput time and the "Target" inputThe formula I am using is: TRIMMED_MEAN(CALC_THROUGHPUT(CASE_START TO CASE_END, REMAP_TIMESTAMPS(""."", DAYS)), 5, 5) - <%=Target%>)However, the table output will fail if the user wipes out the value from the Variable Input component (which for sure can be the case).  I've tried with different operators to handle CASE WHEN the variable is empty (blank): IS NULL, != '', !="" - it's not working (e.g.  CASE WHEN <%=Target%> IS NOT NULLTHEN(TRIMMED_MEAN(CALC_THROUGHPUT(CASE_START TO CASE_END, REMAP_TIMESTAMPS("GTO_Application_Workflow_csv"."CHANGE DATE", DAYS)), 5, 5)- <%=Target%>)ELSE NULLEND) The only solution I was able to found in documentations deals with

It seems that date picker filter is not applicated in the same way for:the KPI in the high side of the sheet (Sales Order Items value and Number)the KPI in the table that shows year by year Sales Order Items value and Numberthe values (Sales Order Items value and Number) for a specific customer if we add also this filterBecause for some customers each of them shows different values ..

Hi,We have Excel files with few columns are always/sometimes blank. When we try to upload the excel, Celonis automatically identify and select the datatype for each column based on the first x records . Even though column has null values Celonis considered them as INT and then during the data model load this gets failed. We have many columns and manually checking data type for each column is very time taking. Is there any solution to this?

Hi, I am trying to filter the data using CASE WHEN statement in the component filter section but it is not working.Scenario: when DATE_1 is not NULL then show data only till yesterday using DATE_1 (DATE_1 < TODAY()) else show data till yesterday using DATE_2 (DATE_2 < TODAY()). This works file if I execute single statement like FILTER DATE_1 < TODAY() but it is not working when I combine into CASE WHEN like below: FILTER CASE WHEN ISNULL(DATE_1)=0 THEN DATE_1 < TODAY() ELSE DATE_2 < TODAY(); Any help is really appreciated. Note: We are using Celonis on-prem 4.5

Can we add comments against activities to cite the reason as to why the activity occurred? If so how to do it?