Hi, I’m Niklas!

Welcome to my personal website! I am presenting my research and work on this website. I was born and raised around Aachen and am currently doing my Ph.D. under the supervision of Wil van der Aalst at the Chair of Process and Data Science, RWTH Aachen University.

My Research

(Object-Centric) Process Mining

While processes are being executed in any kind of real-life system, they leave data traces in the information systems that support their execution. We can extract discrete events from the information system, telling us what happened at which time for which execution of the process. This so-called event log can be analyzed to uncover the underlying process, check for deviations, improve performance, or provide other means of operational support. Traditionally, each execution of the process is assumed to be a sequence of events. However, this assumption is limiting to the true nature of processes: A process execution can contain different objects for which different events occur. Examples of such processes are supply chains or production processes, where different parts are being produced and assembled. My core research deals with translating process mining techniques to this more realistic setting and assessing the improvements to process mining results yielded by this new generation of process mining algorithms. These advancements in object-centric process mining help increase the accuracy of process mining insights.

Research Areas

Object-Centric Process Mining

Moving from one process to multiple dependent processes & from sequences to graphs

Fundamentals of OCPM

Object-centric process executions cannot be captured using the traditional, sequential case concept. We generalize this concept to a graph-based concept that can accurately capture the data. Based on this, we adapt process mining algorithms by proposing object-centric variants, performance analysis, quality metrics, and a predictive monitoring framework.

Stay Tuned…

Future research will be placed here 🙂

Tooling and Code

We provide extensive support to apply and extend our contributions to object-centric process mining. The tool OCπ lets a user discover an object-centric process model along with the object-centric variants and allows for filtering the event data. The python library ocpa is an open-source solution that contains all of the proposed object-centric process mining algorithms, along with data sets and example scripts for an easy extension.

Concept Drift

Detecting and explaining significant changes in processes

Explainable Concept Drift

While detecting significant changes in a process already gives useful insights one might also be interested in potential reasons for the change, uncovering root-causes and potential improvement points for a process. You can find our paper introducing explainable concept drift here.

Evaluating Process Concept Drift Detection

A plethora of process concept drift detection algorithms has been proposed over the past. However, a systematic evaluation of strengths and weaknesses has so far been missing. Find our experimental evaluation here.


ᴏᴄᴘᴀ Object-Centric Process Analysis


A Python library for analyzing object-centric event data. The library comprises functionalities to load and process object-centric event logs, perform process discovery, evaluate process models, analyze process performance, and predictive process monitoring.

Together with Gyunam Park. The documentation is available at readthedocs. Code on Github.

OCπObject-Centric Process Insights


A standalone tool to explore object-centric event data. OCπ allows users to filter and tailor object-centric event logs to visualize the process model and variants of process executions.

Tool and corresponding publications are available at ocpi.ai

PADS Excellence

Founder and Head

The honors class (top 5% of process and data science students) and alumni network for the PADS (Process and Data Science) Chair at RWTH Aachen University. Currently, PADS Excellence has more than 100 members who are affiliated with more than 50 universities, more than 30 companies, and of which more than 30 have co-authored a peer-reviewed scientific publication.

Together with Harry Beyel and Christopher Schwanen. Find some information here.


Journal Papers


Jan Niklas Adams, Gyunam Park, Wil M. P. van der Aalst:
Preserving Complex Object-Centric Graph Structures to Improve Machine Learning Tasks in Process Mining.
Engineering Applications of Artificial Intelligence, 125: 106764

Jan Niklas Adams, Sebastiaan J. van Zelst, Thomas Rose, Wil M. P. van der Aalst:
Explainable Concept Drift in Process Mining. Information Systems, 114: 102177


Jan Niklas Adams, Gyunam Park, Wil M. P. van der Aalst:
OCPA: A Python Library for Object-Centric Process Analysis. Software Impacts, 14: 100438

Conference Papers


Jan Niklas Adams, Cameron Pitsch, Tobias Brockhoff, Wil M. P. van der Aalst:
An Experimental Evaluation of Process Concept Drift Detection. PVLDB, 16 (8): 1856-1869

Lukas Liss, Jan Niklas Adams, Wil M. P. van der Aalst:
Object-Centric Alignments. ER 2023, TBP.


Jan Niklas Adams, Gyunam Park, Sergej Levich, Daniel Schuster, Wil M. P. van der Aalst:
A Framework for Extracting and Encoding Features from Object-Centric Event Data. ICSOC 2022: 36-53

Jan Niklas Adams, Daniel Schuster, Seth Schmitz, Günther Schuh, Wil M. P. van der Aalst:
Defining Cases and Variants for Object-Centric Event Data. ICPM 2022: 128-135

Gyunam Park, Jan Niklas Adams, Wil M. P. van der Aalst:
OPerA: Object-Centric Performance Analysis. ER 2022: 281-292

Jan Niklas Adams, Wil M. P. van der Aalst:
OCπ: Object-Centric Process Insights. Petri Nets 2022: 139-150


Jan Niklas Adams, Wil M. P. van der Aalst:
Precision and Fitness in Object-Centric Process Mining. ICPM 2021: 128-135

Jan Niklas Adams, Sebastiaan J. van Zelst, Lara Quack, Kathrin Hausmann, Wil M. P. van der Aalst, Thomas Rose:
A Framework for Explainable Concept Drift Detection in Process Mining. BPM 2021: 400-416

Workshop Papers


Jan Niklas Adams, Wil M. P. van der Aalst:
Addressing Convergence, Divergence, and Deficiency Issues. OBJECTS@BPM2023: TBP (Accepted)


Harvard BST 209

Instructor/TA (2023)

How to use data science to derive knowledge about healthcare? We investigate this question in the Collaborative Data Science in Healthcare course at Harvard SPH. My task was giving a class on process mining and mentoring students on data science.

Introduction to Data Science

Instructor (2021/22, 2022/23)

How to use data to make informed decisions? This question is investigated in IDS. As an instructor, my responsibilities included hosting instructions for the 800 students participating in the course, and creating assignments and exams. My main topics are supervised learning, e.g., Regression, SVM, and neural networks. The main target is to equip students with the necessary knowledge to conduct a sound data science pipeline for making decisions.

Business Process Intelligence

Instructor (2021, 2022)

How to gain data-driven insights for processes based on collected event data? This question is investigated in BPI. As an instructor, my responsibilities included hosting instructions for the 450 students participating in the course, creating assignments and exams, and giving a Celonis guest lecture.

Advanced Process Mining

Instructor (2021)

What are state-of-the-art process mining problems and their solutions? These questions are investigated in APM. As an instructor, my responsibilities were the creation of assignments and exam questions and providing a guest lecture on concept drift in process mining.


Lecturer (2022)

How to conduct science? The proseminar is a small mandatory bachelor’s course aiming to equip students with the essential tools of the scientific method. My responsibility is compelling writing, i.e., how to communicate your research efficiently and convincingly.


In 80 Schnitzels Around the World