Prediction and Simulation in Process Mining
Language: The language of the course is English; therefore, all meetings and written reports will be in English.
Process mining brings transparency into the processes using historical event data recorded by information systems. For instance, process discovery techniques reveal the real running process inside an organization, conformance checking techniques determine whether the running instances of the process are in line with the discovered or planned process model, and process enhancement techniques provide a variety of techniques to improve the process in performance aspects. Most of the techniques provide insights into the past execution of processes. An important aspect of process mining to enhance the processes is providing future analysis of processes. Simulation techniques are able to address this aspect, however, creating an accurate model is not an easy task. Using the power of process mining an accurate and close to reality simulation model can be generated, instead of using assumptions such as how processes look like or what will be the average service of each activity,
In this seminar, we focus on the forward-looking techniques in process mining which use the provided insights by process mining to generate a close to the reality simulation model. These created models are exploited to play-out the behavior of the processes and running what-if analysis. Many techniques are introduced to re-generate a process for the purpose of performing different scenarios, e.g., discrete event simulation, business process simulation, and queue theory. Process mining techniques fill the gap between the real processes and the simulation models. In practice, event logs are the start points of process mining techniques including process model discovery, conformance checking, and performance analysis. These event logs and the provided insights are the input of future analysis techniques in process mining.
In this seminar, students will practice understanding research papers, writing a research paper, and based on the gained knowledge providing some research ideas. These steps will be followed by a presentation to support presenting research topics clear and understandable to the audience.
Each student starts with a specific research paper and will perform a literature survey as the first step. The next step is to provide an outline of their research paper which is followed by writing their research paper for the given topic. They conclude their work by presenting the results and discussing them with other students.
The registration is carried out via the central registration process on the website SUPRA maintained by the Computer Science Department. Please state your qualifications, experiences, and all grades in your enrolled study as detailed as possible. Please give clearly why you are a suitable candidate for this seminar.
You will be informed about the first meeting in the weeks after SUPRA announces the final participation lists.
The course is expected to start in February 2023.
Prior knowledge of the fields of Process Mining. Ideal prerequisite includes:
- Participation in the Introduction to Data Science (IDS) course
- Participation in the Business Process Intelligence (BPI) course
- Participation in the "Process Mining: Data Science in Action" MOOC on Coursera
- Having read the book "Process Mining: Data Science in Action" by Wil van der Aalst
All deadlines for submission of the outline, term paper, presentation slides will be announced in the kick-off meeting. Presentations will take place on probably two or three days and this will also be announced in the kick-off meeting.