After successfully completing this course, you will be able to
- Apply process mining techniques to real-life event data to analyze processes (for example, healthcare trajectories, or the order-to-cash processes)
- Explain the essential concepts of process mining (event data, processes, activities) and apply some popular algorithms
- Apply and explain process mining analysis, including, process discovery, conformance checking (process anomaly detection), process performance analysis, process outcome prediction, robotic process automation.
- Apply the essential event-data preprocessing techniques (such as creating views, aggregating events, enriching event data, various filtering techniques) to real-world datasets
- Interpret popular quality measures (e.g., fitness (recall), precision, generalization, complexity) to evaluate the performance of algorithms
- Collaborate in a team to tackle a data science project focussing on process analysis
- Use some popular process mining tools and libraries (e.g., disco, ProM, pm4py, Celonis)
The overall grade for this course will be determined based on the following :
- An individual assignment (presentation) researching on a self-selected process mining topic (e.g., Process discovery technique, RPA, process compliance) (15% of the final mark)
- A group assignment analyzing on a real-life event log (40%)
- An individual exam (30%)
- Course participation (15%)
The following background information will help you understand the content of this course:
- data / object modeling (e.g. UML class diagrams)
- Process modeling (e.g., BPMN, Petri nets, Markov models)
- basic data mining or machine learning concepts (e.g., classification, clustering, decision tree, accuracy, recall/precision)
Event data are being unceasingly recorded during process executions, for example, when a patient is following a healthcare treatment trajectory, when sales orders are placed online and delivered, or when a mortgage loan application is submitted.|
Event data store information about how these processes are executed in real-life and who is executing which tasks and when.
To analyze such data and to improve processes, Process Mining emerged as a Data Science discipline that studies event data and process behavior and provides data scientists with powerful tools.
Process mining techniques focus on
(1) the automatic learning of process models using event data (aka Process Discovery),
(2) the detection of anomalies in processes (Conformance Checking),
(3) the prediction of future process activities and outcomes (Predictive Process Monitoring),
(4) the visual analysis of event data to improve process executions, and many other topics.
With the recent uptake of process mining, many companies have started making use of this novel technology to obtain insights into their process executions.
These valuable insights help them to derive fact-based recommendations for automating processes, for (re)designing business strategies, and for improving process quality and efficiency.
To show the high practical relevance of process mining, the IEEE process mining task force has reported many successful case studies in a wide range of application domains (such as healthcare, logistics, finance/banking, etc) [Link1].
An article was published in the Harvard Business Review explaining why process mining is essential [Link2].
According to Forbes, Celonis, a major process mining vendor, raised a billion dollars in New York [Link3].
If you are curious, here is also a two minutes YouTube video that briefly introduces you to process mining [Link4].
This course focus on both the theoretical foundations and the applications of process mining techniques to real-life data sets.
The overall aim of the course is to provide the students with theories, techniques, tools, and practical experience for applying Process Mining to tackle process-oriented data science problems.
The course consists of lectures and tutorial sessions.
During the lectures, you will learn the theories, the concepts, and the algorithms in Process Mining.
During the tutorials, you will gain hands-on experiences by applying algorithms and tools and working on exercises and a practical assignment with real-life event data.
Selected book chapters, articles, and lecture notes.
Competencies-Entry requirements |Required materials-Instructional formatsTests|