Read Blog Post

Read Blog Post

Contributed on May 6, 2021

1

487

Sequences of Stuff

In last week’s Process Mining Café, we talked about finding events in unexpected places with Léonard Studer, an internal consultant for business process management at the City of Lausanne in Switzerland. You can now watch the recording here.

We talked about how looking for process mining data is essentially looking for “sequences of stuff”. Once you realize what the minimum requirements for process mining are, you start seeing possible process mining data everywhere.¹

Léonard gave three examples of sequences of stuff that he has analyzed with process mining.

Sequences of circulations

In the first example, Léonard took data related to a building permit process from the ERP system at the city. Because the ERP data had no real activity name (just a free-text field), he started to analyze the data with a text mining tool. He ultimately broke down the process into circulations of files and used his understanding from the text mining to categorize the communications into ambiguous vs. non-ambiguous texts, which had a clear correlation to the duration of a case in the process.

Sequences of people

In the second example, there was no information system supporting the process at all. But Léonard still wanted to help the opera in Lausanne with their program management process. So, he looked at the creation and access of files via the Network Attached Storage (NAS) access log. From these access logs, the implicit communication flows between people – and the activity levels of the program creation process – could be analyzed.

Sequences of speeches

As a third example, Léonard looked at debates in the local parliament as a process: The activities were the speeches (or the parties); the cases were the topics that they debated.

Léonard also shared many practical tips about how to put people at ease when you start analyzing their work with process mining. For example, he first makes sure that they understand that the goal is to analyze the process – not the people. Then, he schedules a confidential session with the workers in the process without management. In this meeting, nothing leaves the room (like you would expect from a priest in a confession).

He has seen that, once the right steps are taken, process mining really helps the people to speak the truth about their process.

Thanks again to Léonard, and to all of you, for joining us!

Links

Here are the links that we mentioned during the session:

• Data suitability checklist for process mining

• OpenRefine is the text mining tool that Léonard used

• Léonard’s Process Mining Camp talk from 2015, where he shows how he works with OpenRefine

• The Ethical Charter that Léonard contributed as an example for the process mining book