Process Mining Café

When Linda took over the process improvement team, she realized that meeting with domain experts for two hours per week on a project did not lead anywhere. After she switched to full-day workshops, everyone noticed that it was significantly better. They have gained focus and obtained much better results. Linda shares step by step how she prepares and runs these workshops.

Healthcare processes are complex due to the individuality of each patient, so process mining is particularly useful for understanding them. Process maps Owen's research group discovered at a UK hospital illustrated how often patients were misdiagnosed and sent around. Owen talks about their studies as well as their methodological framework for process mining in healthcare.

Data protection teams, the works council, and process owners need to trust that you will act responsibly in your process mining project. How do you address their concerns? We talk with Daniel about the organizational side (who to involve when and how) as well as technical steps you can take, such as removing unnecessary information, anonymization, and pseudonymization.

How do you determine what you should compare in a process? Hanna discovered a process at her bank for three different regions. She analyzed the case durations but also rework, parallel vs. sequential flows, and long waiting times. After understanding the different process characteristics in each region, Hanna identified best practices for improving them.

Clickstreams are the digital traces that visitors leave when they navigate through a website. You can then analyze this data with process mining to make the website more effective or user-friendly. Irene shares her experience of a recent clickstream analysis project. We discuss why the funnels of traditional web analytics are not enough to understand behavior.

Vinicius, Rudi, and Anne talk about the common challenges and best practices for process mining. They discuss the importance and unimportance of data quality, data privacy, process mining methodologies, and domain expertise. What must you do to ensure your process mining initiative is successful? Watch the café to find out!

When you import your data set, you first have to look around and get an overview about your process. Where does it start and where does it end? Can you recognize the expected process? Is there a lot or little variation? Do you see rework or other unexpected patterns? Join us as we discover the loan application process from the BPI 2016 dataset.

Some preprocessing tasks can be easily done in Disco. Others require the use of data transformation tools. Xixi has classified the common preprocessing tasks for process mining into six categories: Enriching, Integration, Filtering, Transformation, Reduction, and Abstraction. For each category, we discuss concrete examples from practice.

Many companies use or plan to use automation techniques to reduce manual labor. But which parts of a process can be automated, and which parts should be better kept human? Erik and Lloyd share how they use Disco to analyze the performance of their RPA robots, identify new automation opportunities, and analyze their training processes.

Auditors have different goals than process improvement teams. For example, internal auditors verify the efficiency of the process and whether it is under control. They analyze the segregation of duties, variants, and cases. Mieke talks about the benefits of using process mining as an auditor and presents two recent research works by her and her team.

We review what we learned at Process Mining Camp 2024. At camp, we immersed ourselves for two intense days of exchanging knowledge with our fellow process miners. Bhawana was a first-time attendee, and Lieven was a speaker at this year’s camp. They share their takeaways, and together, we retrace the steps of these two days.

Variants are sequences of activities, and they are their own dimension in process mining. They complement the process map by providing a scenario-based view of the process. Looking at the variants can give you many insights. But the crux is ensuring your variants are meaningful before you analyze them. Rudi and Anne show how to elevate them to the right level.

Finding your way to your business case as a process miner is a bit of a chicken-and-egg problem: You first have to do the analysis to determine the improvement potential. Rudi and Anne recommend keeping it simple, and Miguel shares how they increased revenue by 6 million Pesos (350,000 USD) without hiring additional people.

Andrea and Daniel discover process flows for the bus lines in Montevideo using open data from the city. They then export the XML process maps from Disco to display the discovered processes in the spatial city map. We discuss their project and Andrea and Daniel give you all the information that you need to reproduce their steps.

There is a lot of variability in Human Resources (HR) processes. This is only natural as people are different! And there is different legislation in different countries, different types of benefits, etc. We show two approaches for dealing with this complexity: One bottom-up approach and one top-down approach using the APQC standard process framework.

Hugo walks us through the analysis of the complaints process at the Dutch health insurance provider VGZ. As the data analyst, Hugo could use process mining to bring a fact-based view of the process to the subject matter experts. But he also needed to protect them from jumping to conclusions too early when the data was not good enough yet.

In Activity-Based Costing, indirect costs such as salaries and utilities are tied back to products and services. If you combine Activity-Based Costing with with process mining, then you can allocate costs to activities in your processes and evaluate “buckets of cases” according to their costs for the organization. Willem and Anne show how this works based on a concrete example.

What do you do if you want to do process mining but you don’t have any data? Tobias found an elegant solution: He created a system that allows employees to collect data about process executions in a lightweight and system-independent way. One advantage is that this internal system has privacy considerations built-in.

The data you find in information systems was not created for process mining purposes. So, often, you need to shape the data to make it suitable. Furthermore, different analysis questions may require particular views on the data. Together with Scott, we look at two concrete examples to illustrate the steps in such transformations.

Rudi and Anne talk about what it takes to make a good project plan. We first discuss the criteria for selecting a process for your process mining initiative and then go through the five ingredients of your project plan. Rather than just a big, scary document, the project plan is there for you and helps you manage the risks and communication points in the team.

If you are creative, you can find process mining data in unexpected places. Nuno shows us how he turned static data from an agricultural government funding program into a process view. After creating timestamps and activities from implicit information, he could analyze how farmers requested more or less aid over the years in the context of the development of their farms.

Getting the data is often one of the first challenges you have at the beginning of your process mining project. Rudi and Anne answer the most frequent questions around process mining data. We also show you two common data transformations based on a practical example: One for merging two data sets and one for transposing a data set from columns to rows.

Jan and Muthu have applied process mining to their quality management processes. On top of that, they have also extended their analyses with the so-called process capability index (Cpk). We show, based on a concrete process mining example, how this works. And you can even try the approach yourself by using their Excel template.

Magy gives an overview of the different types of games and traditional game analytics. We look at examples, where her group has used process mining to understand the flow of games. We also talk about the ethics of using game analytics to make games more addictive and discuss the challenge of abstracting the data to the right level.

Process mining does not replace the traditional audit approach. However, it requires some changes and a conscious effort to fit process mining into the existing way of working. In some places, more work is needed. In other places, things get easier. Together with Jasmine and Andreas, we talk about their data-driven audit approach with process mining step by step.