Data Ingestion & Infrastructure

Automation Rate, Simple Ratio formula is as follows: KPI("Ratio", "_CEL_AP_ACTIVITIES"."USER_TYPE" = 'B') -- result: 64.07% We are trying to replicate the same simple ratio with a standalone query instead of visual editor to understand the mechanism of the Simple Ratio. We have tried two formulas and both of them gave us different answers. Using PU_COUNTSUM(PU_COUNT("BSEG" , "_CEL_AP_ACTIVITIES"."USER_TYPE" , "_CEL_AP_ACTIVITIES"."USER_TYPE" = 'B')) / SUM(PU_COUNT("BSEG" , "_CEL_AP_ACTIVITIES"."USER_TYPE")) -- result: 65.45%  Using CASE WHEN statementAVG(CASE WHEN "_CEL_AP_ACTIVITIES"."USER_TYPE" = 'B' THEN 1.0 ELSE 0.0 END) -- result: 65.51% As it can be seen from above, that each query gave us different results and we were not able to replicate the Simple ratio as well. What is the actual backend formula used for Simple Ratio? A

Hello, for a complex formula, I need to substract two percentages in order to multiply it afterwards However, the formula including a filter is successful and consistent for a Count but not for an Avg (the result is inconsistent, far from the to-be result) :   Also, I tried to put the filter in a Case When where I obtain the percentage thanks to a division between the ‘Change Items’ (= Count) and the ‘PO Items Not Auto’ (= Count_Table’) but I have the error :    Please note that my client has the OnPremise Celonis, therefore some functions available on the Cloud may be not available here. Any help would be the most welcome, thank you and have a good day ! Best regards 

We need a Celonis extraction connector and/or app for Microsoft Dynamcis365 FO (Finance and Operations) and CE (Customer Engagement). For IT service management, ServiceNow is used. Same need there. Any existing Celonis assets for these applications?

How to check any dimension (like Vendor etc) wise Violation Rate?

Hi All,I have basic question on PQL. How can I explore PQL for process mining? Do we have any start up guide? Should I need to have full license version to explore PQL ? ThanksDurai

Hi all, I have a basic question regarding activity table column configuration. We have the luxury of having start AND end datetimes for each activity. I remember this was possible at some point to specify the date columns as such with the effect that for each activity the duration was calculated and shown in the process explorer (performance view).  Is this still possible, and if so, how? Thank you!

can you please make a video how to integrate the Uplink Integrations also how to create a data connection 

It also cannot add unique ID number as a column the table which i have created before. question1. is there another way to create uniqueID for each row? question 2. how can i fix this error?    

I am not able to interpret why trigger tables are used. Data that will be extracted on real time, doesn't it automatically be refreshed in all tables?

can you please make a video how to integrate the Uplink Integrations also how to create a data connectioni try to many time but every time i fail  

System requirements of an on-premise Extractor serverThe on-premise server will be running both Celonis SAP Extractor and Celonis on-Prem Agent. For this we recommend the following system requirements:Hardware requirementsVirtual machine or physical serverCPU: Min. Intel Xeon processor with 4 CoresRAM: min. 20 GBi don't have a 20 GB RAM . so its true that i can't connect the data connection please solve my issue i want to try to make data connection but i can't   

Hi @janpeter.van.d​ / @Celopeers .​  I need to query Table A and look for all distinct Machinename . After that I need to update Table B, If Machinename exists so ok. If not, need to create new lines in Table B with 'No ID' field at location 

Team - we are supporting a user community that has been using the platform for over two years. As such, they have many apps that have been loaded over the period of time and the majority are not being used. We are looking for a report that shows the list of all the applications and what space they reside. Also, who published it and when and who last used it and when. This could give us a clear picture of what we can archive. Any suggestions or do you have a report to re-use? Thanks.

Hi @janpeter.van.d​ / @Celopeers .​  I'm trying to compare two dates from different columns in order to find out about a delay.  For this I need to (1) compare the highlighted dates in the red box (2) if the ETA is earlier than the Timestamp of the Activity "Ready for Pickup" I need to mark it with the value "1" in the respective column and line (as highlighted) [Note that the columns "ETA" and "Delay" are two manually created attributes within Celonis and were not part of the extracted data.] I guess I need to use a PU function here, but as I'm not very familiar with it I'd very much appreciate any help.  Thanks in advance anf brgds, Finn

I read the following message before: We have the AVG time cycle for a set of case and we want to obtain the % of cases with TC below de average and above average.I tried to do it in this way: CASE WHENCALC_THROUGHPUT(ALL_OCCURRENCE['Process Start'] TO ALL_OCCURRENCE['Cla. Closed'],REMAP_TIMESTAMPS( "AT_CLAIM_L1"."_TIMESTAMP" , days))>AVG(CALC_THROUGHPUT(ALL_OCCURRENCE['Process Start'] TO ALL_OCCURRENCE['Cla. Closed'],REMAP_TIMESTAMPS( "AT_CLAIM_L1"."_TIMESTAMP" , days)))then 'TC > AVG'else 'TC <= AVG'END but I get the following message:The aggregation function AVG cannot be used together with a dimension function input. Please check that there are no aggregations and dimensions used together as function inputs. Which would be the correct way to have the information we need?  AND this was the answer:  Save AVG(CALC_THROUGHPUT(ALL_OCCURRENCE['Process Start'] TO ALL_OCCURRENCE['Cla

 After creating an Attribute, to have a specific column in my table in celonis, I cant access this column in the KPIs.How can I access this new column when working with KPIs? Thanks in advance!

UNABLE TO UPLOAD FILE ON EMS !!!I have a csv file that is correctly uploaded on a trainning team but is being uploaded on my EMS CUSTOMER. Is Trainning better than EMS?

How to find the Cluster Details in the Training environment? Please Help!

Hi, I am trying to create a new connection to the on-prem MS SQL Server Database on my personal system for training purposes with database type as Microsoft SQL Server(native) but getting an errorTelnet connection to the host MSSQLSERVER2019:1433 failed. Telnet failure indicates that either host/port combination is incorrect or the extractor is not allowed to access the database due to firewall or access permissions.Please see the screenshot below.Port Number: 1433 Please help to resolve the issue. 

We are trying to do a user analysis to understand the usage of Celonis applications and were able to process the audit logs. Is there a way to understand how many Users were created along with user groups and Date of creation of User id etc. Swathi

Hello everyone,here is my case: I am using SAP table CRMD_ORDERADM_H as case table. I am using PU_SUM to have an OLAP table column with the sum of CRMD_ORDERADM_I.NET_VALUE grouped by CRMD_ORDERADM_I.HEADER (case id). I am using a dropdown to filter for CRMD_ORDERADM_I.ORDERED_PROD. It works in so far that the table only shows cases, where at least one item has the selected PROD, but the sum stays the same.I would like the filter to be applied to the sum.In case you are not familiar with SAP tables - I added a mock up of what I am trying to do.Have you got any ideas how I could accomplish that?Thanks in advance!

Dear Celopeers, May I ask you to dispell a potential misunderstanding on my part regarding the Calc_throughput function. I have three consecutive activities : A, B and C I would expect that: CALC_THROUGHPUT(FIRST_OCCURRENCE['A'] TO LAST_OCCURRENCE['B']) + CALC_THROUGHPUT(FIRST_OCCURRENCE['B'] TO LAST_OCCURRENCE['C']) = CALC_THROUGHPUT(FIRST_OCCURRENCE['A'] TO LAST_OCCURRENCE['C'])   Unfortunately, this is not what I'm observing at the moment when doing this on aggregates (say... all Purchase orders processed in the last 6 months). Calc_throughput A->C reports a much shorter throughput time than i would expect from the above calculation. What could be the reason for this?  Best Regards,

Hi AllAm very new to celonis data connection and events including PQLI have a data set for Requisition PR (EBAN), Procurement PO (EKKO and EKPO)I also have the finance data invoices BSEG.BSCHL = 31Is there a column or a query l can use to directly link the invoices (BSEG) to POs(EKPO) without the table EKBEEventually, l have to compare PR, PO and Invoice value?Thanks

Hi,Imagine a table of sales data with each row containing the sales date and the amount sold. My goal is to calculate the standard deviation of monthly sales (and use this value in other formulas). Getting the monthly sales is easy:DIM: ROUND_MONTH(SALES.DATE), KPI: SUM(SALES.AMOUNT) In order to STDEV over those monthly values, I thought about doing the monthly sum within a PU function like so: PU_SUM(DOMAIN_TABLE(ROUND_MONTH(SALES.DATE)), SALES.AMOUNT)) In my understanding of PU functions, I would've thought that this results in the same values as DIM: ROUND_MONTH(SALES.DATE), KPI: SUM(SALES.AMOUNT).  But it does not, the values are roughly 10 times as high, each "month" is higher than the sum of all amounts together.  In the end I want to be able to use a formula to get the STDEV and use this to calculate other KPIs Thanks in advance

I have some KPIs that are calculated based on the unfiltered population and some that are filtered by selections, but now I want the population to be flexible and since I have multiple filters that are applicable I need those KPIs to be flexible too. It does not work if I set the variables to not static because then some KPIs are calculated incorrectly with the sheet filters and selections even if i calculate them with the GLOBAL function. You can think of it as two populations. The first population is the basis for the main KPIs which is filtered by more generic filters but not with selections. The second population is the filtered first population which is the basis for more KPIs and visuals which can then be selected.  If anyone could think of a better way to do this I am all ears too because if I can assign multiple values to a static variable then I will have to implemented x Case When statements to check for which filters are active.