Just recently, I shared my first blog post about Process Mining. In my previous post, I shared the first step to get started with Process Mining. In this post, I will go one step further and present a walkthrough of how to create a new process using an event log data file.
Create a new process with an event log
Before you can create a new process, you would need to have an event log file. An event log contains at least a Case-ID, an Activity, and a Timestamp. It can have more attributes for more in-depth analysis and filtering. In my next post, I will elaborate on the data requirements of an event log file and will show what happens with the minimum or more attributes. For this blog, you can download an example from Microsoft Learn. This example is the same as the process you get when choosing to get started with a finance process. You will find the download on this page: Get started with the process mining capability – Power Automate | Microsoft Learn. Note that this process does not have any relation with Dynamics 365. On the same page, you can learn more, like the options on how to look at the generated reports.
To create your own process providing an event log, you can navigate to the Process Mining page on Power Automate and click the tile Start here.
A form will pop up where you can specify a Process name and Description.
As mentioned in the introduction blog, there is a difference between process mining and task mining. Power Automate supports both. I will focus for now on process mining. In that scenario, choose Import data and click the button Continue. The data source Recordings is intended for task mining.
There are 4 steps to complete before you can view the report for the process. Firstly, you would need to upload the file using a data source.
Like Power Automate Flow, there are different connectors available as a data source. When you click on View more ->, you will see a complete list of connectors available for Process Mining.
At this moment of writing, there are 51 connectors available, This is far less than the connectors for flows, but for sure enough to create your processes. Only connectors providing tabular data are needed. The MSN Weather, for example, is not a connector that can be used for process mining. If you browse the list, you will notice that there is no connector for Dynamics 365 Finance and Operations and Dynamics 365 Business Central. Depending on the scenario, like the number of records and what data needs to be analyzed, you can consider using Azure Data Lake, Text/CSV, DataVerse, or a Web API to analyze data from your ERP. In a future blog, I will talk about extracting and transforming data from transactional tables into event logs.
For this walkthrough, you can choose the data source Text/CSV and upload your file.
In this second step, the service will try to use your credentials to get access to the file via OneDrive or SharePoint. In case your credentials do not match the storage security permissions, you would need to manually provide different credentials that have access to the file.
Click Next to go to the next screen, where you can validate the contents of the file.
In case you expected different contents or would like to make some changes, you can return to the previous step. To continue, click Next which will bring you to the third step where you can filter data, remove columns, and transform data using Power Query.
Changes you make, are traceable on the Query settings side panel. Click Next to go to the last step before the process mining report can be generated.
In this step, you need to link the file columns with attribute types. There are three mandatory attributes. The first column will show the headers from the source. You can change the attribute types to meet at least the mandatory attributes as mentioned earlier in this blog. There will be a check if these three are actually set up as attribute types. Note that this screenshot is different from the settings shown on Microsoft Learn. At this moment, I only focussed on the mandatory columns for generating the report. You can also change the attribute type to the meaning of e.g. a resource or case-level attribute. I will talk about other attribute types in another blog.
Once you set up the attribute type, click Save and analyze which will start processing the data and building the report.
The preparation of the report can take several minutes. You can wait for the report to be completed or come back later and open the report. You will get an email notification in case the report is ready.
Once the report is available, you can select different views, like looking at the duration times. If you do the same as I did, the report will not have duration times for the activities as I didn’t map the end time attribute. I will elaborate on attributes in my next blog when I will explain the data requirements for an event log with some examples.
YouTube demonstration video
A video demonstrating the creation of a new process using an event log is embedded in this post. On my YouTube channel, I added a playlist for The F&O Twist on Process Mining. More videos will be added along with more blog posts on Process Mining.
There is more…
Still after two blogs, there are more topics to discuss. In my next posts, I will for example write about:
- What is an event log with process mining? What attributes are mandatory and what is the effect of using some non-mandatory attributes?
- Process mining integration, available in Dynamics 365 Supply Chain Management
- How to create event logs from Dynamics 365 F&O transactional data
If you want to learn more about Process Mining, you can explore the documentation on Microsoft Learn: Overview of process mining and task mining in Power Automate – Power Automate | Microsoft Learn
I do hope you liked this post and will add value for you in your daily work as a professional. If you have related questions or feedback, don’t hesitate to use the Comment feature below.
That’s all for now. Till next time!