Data Ingestion & Infrastructure

Can I get values from a referenced case? Hi, in my datamodel the cases are customer's orders. An order can get paused due to other order of the same customer. In this case the first order also has the CaseID of the blocking order as a reference in the CASE table. Now I want to show each CaseID in a column and a value from the referenced Case in a seperate column.  I.e. I want to say something like:"Give me the value "CASE"."ORDER_TYPE" of the case whose CASE_ID is referenced in THIS case's column "CASE"."BLOCKED_BY". All attempts with Pull-Up and Domain_Table statements failed.Any ideas? BR Florian

celonis data extraction from source system is pull based or push based ?

During Step 4: Does this step generate a flat file or is all data held in memory?

Hi guys! I’m trying to apply a particular filter to a dataset with many activities. The scenario is as follows. The raw event log contains activities done in a general hospital for over a year.If I want to get the process for a specific activity, I would normally just filter ‘Include only cases that contain ‘Knee Replacement Surgery’.  However, because there is so much going on in the hospital, such a filter is not enough. For example, if a person underwent a knee surgery is june 2021, the process map will also try to include other procedures that this person did in january 2021 or december 2021, even though these have nothing to do with the knee replacement surgery. For that reason, I’m trying to narrow the filter down as follows: “Provide an event log that only includes data of a maximum of 2 weeks in the past after the activity ‘Knee replacement surgery’ and a maximum of 2 weeks after this same activity.” Note that this is NOT the same as applying a general time filter on the event

I am curious how much data can we extract and store in celonis ems for both free training use and production use?

---------------------------------------------------------------------------HTTPError Traceback (most recent call last)~/.local/lib/python3.8/site-packages/pycelonis/celonis_api/base_objects/ibc.py in api_request(self, url, message, method, timeout, get_json, threaded, **extra) 273 try:--> 274 r.raise_for_status() 275 except exceptions.HTTPError as e: /opt/conda/lib/python3.8/site-packages/requests/models.py in raise_for_status(self) 952 if http_error_msg:--> 953 raise HTTPError(http_error_msg, response=self) 954  HTTPError: 404 Client Error: Not Found for url: https://dell.us-1.celonis.cloud/process-mining/analysis/v1.2/api/analysis/325c3e97-3a13-4c7a-b818-ac392eebd840/bookmarks/809dfc8f-86c4-4210-9818-a4b2734543c6 During handling of the above exception, another exception occurred: HTTPError Traceback (most recent call last)/tmp/ipykernel_16778/3654934662.py

Hello, everyone, I would like to add my KPI "SUM(CASE WHEN (LEFT("COEP"."KSTAR", 1) IN ('6','9')) THEN "COEP"."WTGBTR" ELSE NULL END)" to the component filter to hide the 0 values. What do I have to pay attention to here, because FILTER SUM(CASE WHEN (LEFT("COEP"."KSTAR", 1) IN ('6','9')) THEN "COEP".."WTGBTR" ELSE NULL END) != 0 does not work. Greetings

WINDOW_AVG (or alternatives) Hi all, I calculate the average cycle time from PO Item creation to Record Goods Receipt per material with the following query: PU_AVG ( DOMAIN_TABLE ( "EKPO"."MATNR" ) , DAYS_BETWEEN ( PU_FIRST ( "EKPO" , "_CEL_P2P_ACTIVITIES"."EVENTTIME" , "_CEL_P2P_ACTIVITIES"."ACTIVITY_EN" = 'Create Purchase Order Item' ) , PU_FIRST ( "EKPO" , "_CEL_P2P_ACTIVITIES"."EVENTTIME" , "_CEL_P2P_ACTIVITIES"."ACTIVITY_EN" = 'Record Goods Receipt' ) )) Now, I want to consider only the latest 10 PO items into the calculation based on the creation date. How can I do that? I tried the WINDOW_AVG function, but it does not work: WINDOW_AVG( <above query>, 0, 10, ORDER BY (PU_FIRST ( "EKPO" , "_CEL_P2P_ACTIVITIES"."EVENTTIME" , "_CEL_P2P_ACTIVITIES"."ACTIVITY_EN" = 'Create Purchase Order Ite

Hello PQL experts, Am looking for SAP Order Confirmation related KPIs ?  Eg : Number of Orders Vs Durations ( Days) .Any inputs is appreciated in Advance ! Br.Vivek

PQL question; I am trying to create a query that counts activities after a specific activity is captured.

Based on understanding of Vertica SQL/Postgre SQL, following piece of script I am implementing to enable primary key. But even after successful, error free execution of these statements, I can see Celonis allows INSERT with same values. Can you help me in understanding if I am falling short on something to enable primary key? CREATE TABLE _CEL_ACTIVITY_PRIMARY_TEST(    _case   varchar(25),    activity_name varchar(100) not null,    eventime DATETIME not null,    PRIMARY KEY(_case)); SELECT * FROM _CEL_ACTIVITY_PRIMARY_TEST; INSERT INTO _CEL_ACTIVITY_PRIMARY_TEST VALUES ('1234','act1','2022-08-22T09:46:46');

I get the following error message when I attempt to upload a flat file (csv/excel): "There was an error during file upload. Please check your file type and content". Any suggestion how to fix this?

We are trying to connect Celonis4.7.2 (onprem) to the GCP using the BigQuery Simba JDBC Driver supplied by Google. The .jar files have been unzipped into <installDir>\\lib but trying to connect from the frontend, we encounter the following errormessage: Has anyone successfully connected Celonis4.7 (or any other onprem version) to the GCP?  Thank you for your help! Cheers, Felix

Hi! I'm working with Celonis task mining, capturing data with filter by applications and URL, I have two different users, one can capture the URL and the other cannot. What could be the problem? Would it be firewall?

For example, in P2P data model we connect EKPO and EKKO using foreign key relationship, but then also we mention that in data jobs while creating EKKO. So my question is why we are doing the double work here? Also what is going to happen if I do not connect two tables that are required to be connected in the data model ?

Hello, I need your swarm knowledge. I need a combination of the following formulas.... TRIMMED_MEAN(CALC_THROUGHPUT(FIRST_OCCURRENCE[<%=Process_Start_1%>] TO LAST_OCCURRENCE[<%=Process_Start%>], REMAP_TIMESTAMPS("T_CEL_AT_HP5"."EVENTTIME", DAYS, MONDAY TUESDAY WEDNESDAY THURSDAY FRIDAY)), 5, 5) ...if the condition beforehand... "/RWEE/PMCP_TAUFK"."ARBPL_AN" LIKE 'S-%-S%' ...grips. Hope for help :-) thx

Hi,  I would like to extract a BPMN model from my Event Log but want to include additional attributes (like user role that is related to the activity, application in which the specific activity takes place, etc.) in order to generate a complete model of the process.  Is this possible, or can you only extract a simple BPMN model with the activities and connections?  Thanks in advance!

Hi there,I am looking for an efficient way to reduce the size of my data model: I already reduced the time period of the ingested data. As a second step I want to get rid of some unused table columns. To identify unused table columns I thought about doing the following:Export a .txt documentation of each analysis, then search for a specific table (ctrl+f) (i.e. LIPS) and document which columns are really in use in this analysis. I do this with all analyses and then I know which columns I can exclude in the extractions because they are not used anywhere.  It's not as much manual work as it sounds but still a lot. Do you know a better way to do this? How do you avoid such work in the future? Is there a standard procedure how to handle this problem or not run into it in the first place?  Thanks for your helpDaniel  

Hello community, we are trying to create a task mining use case but we struggle with the event processing rule  Could you please support with the following: Process start should be only be triggered when the URL = a specific URL Process end should be always when an email is sent When another URL than a specific URL is used then task mining should not capture that informationSee screenshots attached of how the event processing rules are being configured.  

Could you teach me will a two-way connection occur between the On-premise Extractor and the EMS Data Integration?

Dear Community, I'd like to make a OLAP table that have Activity count and Connection count.  Is there any way to make this OLAP table? Thank you. Best regards,Takayoshi Ichikawa 

On click on a drop down button it should show more drop downs like cascading. Is it possible to add like that?

We are trying to create a report which shows contract information with different validities. I can see the information in rows but ideally need them shown in columns.Layout in rows means I have 8 rows for 8 different validities but in columns this would reduce the number of rows massively

I am confused why we use Reload from cache, does it provide some kind of undo functionality? Also I am confusing it with Partial reload, are they same?

1) Raw file (scrubbed):Role=MEMBER; LOC=IL;CLMC=C;Role=MEMBER; LOC=IL;CLMC=C;Role=MEMBER; LOC=IL;CLMC=C;LOC=IL; CLMC=C;LOC=IL; CLMC=C;LOB=LOB1; LOC=IL;CLMC=C;CLR=Y;LGLSST=ASGN;LOB=LOB1; LOC=IL;CLMC=C;CLR=Y;LGLSST=ASGN;LOB=LOB1; LOC=IL;CLMC=C;CLR=Y;LGLSST=ASGN;LOB=LOB1; LOC=IL;CLMC=C;CLR=Y;LGLSST=ASGN;ID=1234,TYPE=PROP;ID=1234,TYPE=PROP;LOC=IL;CLMC=C;LOC=IL;CLMC=C; 2) Desired outcome: