Advancing Machine Learning Methods for Multivariate and Multiphase Time Series to Automate Test Cycle Optimization
 

at ��ҹ̽��, Germany. The DBIS team is pleased to congratulate Stefan on his successful PhD defense (Dr. rer. nat.) and wishes him all the best for his future career. 
Stefan Gaugel was supervised by Prof Manfred Reichert as part of his industrial doctoral project - the industrial partner was Bosch Rexroth AG. In his thesis he proposes a data-driven framework that leverages advanced machine learning to automate test cycle optimization in the context of industrial manufacturing, unlocking cost-saving potential. The goal was to enable faster, more cost-effective quality inspections while maintaining fault detection efficiency by identifying the minimum subset of essential test phases and sensors. The framework was validated using a real-world industrial data set, called Hydraulic Pump Testing (HyPT), which was collected by the candidate in a manufacturing plant. The main results of the cumulative dissertation have been published in high-calibre journals.
 

Members of the PhD committee
 

• Prof. Dr. Manfred Reichert (��ҹ̽��, Supervisor & Referee)
• Prof. Dr. Rüdiger Pryss (University of Wurzbürg, External Referee)
• Prof. Dr. Hans Kestler (��ҹ̽��, Committee Member)
• apl. Prof. Dr. Friedhelm Schwenker (��ҹ̽��, Committee Member)
• Prof. Dr. Matthias Tichy (��ҹ̽��, Chairman)

BPMNE4IoT - A Framework for IoT-Aware Business Processes and Rules
 

at ��ҹ̽��, Germany. The DBIS team is pleased to congratulate Yusuf on his successful PhD defense (Dr. rer. nat.) and wishes him all the best for his future career. 
Yusuf's doctoral supervisor is Prof Dr Manfred Reichert, who has closely supervised him in his dissertation project over the last 4 years. Yusuf worked on the DFG project ‘Service-based Abstractions and Programming Mechanisms for Bridging Business Process Models and IoT Big Data’ (RE 1402/6-1). In his dissertation, he dealt with an extension of the process modelling and process execution language BPMN 2.0 to include IoT-specific aspects (e. g. integration of sensors and actuators in process flows). To this end, he developed the BPMNE4IoT framework for the modelling, execution, monitoring and analysis of IoT-specific processes and evaluated it in several empirical studies (including the digitalisation of processes in a Fischertechnik factory). 
 

Members of the PhD committee

• Prof. Dr. Manfred Reichert (��ҹ̽��, Supervisor & Referee)

• Prof. Dr. Luise Pufahl (TU Munich, External Referee)

• Prof. Dr. Steffen Zimmermann (��ҹ̽��, Committee Member)

• Prof. Dr. Franz Hauck (��ҹ̽��, Committee Member)

• Prof. Dr. Matthias Tichy (��ҹ̽��, Chairman)

 
Improving Deep Learning Performance under Scarce Data Conditions in Medical Imaging
 

at ��ҹ̽��, Germany. The DBIS team is pleased to congratulate Daniel on his successful PhD defense (Dr. rer. nat.) and wishes him all the best for his future career. 
Daniel Schaudt was jointly supervised by Prof Manfred Reichert and Prof Reinhold von Schwerin (��ҹ̽�� of Applied Sciences) as part of a cooperative doctoral project. In his dissertation, he developed advanced methods for dealing with data scarcity when applying deep learning methods in medical image analysis. Daniel is currently working as a project manager at (Transfer Centre for Digitisation, Analytics and Data Science Ulm), of which his two supervisors Prof. Reichert and Prof. von Schwerin are also members of the board.  
Members of the PhD committee

• Prof. Dr. Manfred Reichert (��ҹ̽��, Co-Supervisor & Referee)

• Prof. Dr. Reinhold von Schwerin (University of Applied Sciences Ulm, Co-Supervisor & Referee)

• Prof. Dr. Björn Schuller (TU Munich, External Referee)

• Prof. Dr. Meinrad Beer (��ҹ̽��, Committee Member)

• Prof. Dr. Matthias Tichy (��ҹ̽��, Chairman)
   

As Chairman of the Board of the (DASU), Prof. Manfred Reichert, Director of the Institute for Databases and Information Systems (DBIS), was delighted with the active participation of DASU's committed donor companies at this year's Founders' Meeting in November. The event was hosted by HENSOLDT AG. Almost all founders accepted the invitation to the meeting, underlining the solidarity and commitment within the DASU family.

Review of the past year

After a welcome address by the host, Dr. Jürgen Bestle, and the Chairman of the Founders' Assembly, Heinz Klemm, Manfred Reichert and Reinhold von Schwerin, both members of the DASU Management Board, took over and presented a detailed review of the past financial year and an outlook for the future. The focus was on the further development of the foundation, the report on the RegioWin funding project (EU / state of BW) and the successful projects that underline the growing relevance of DASU's work.

Presentations by Boehringer Ingelheim and Liebherr Group Digital Development Center

A particular highlight were the specialist presentations by Jan Seidel, who addressed the question “AI in clinical development: a new era in medicine?” and Heinz Klemm, who reported on his experiences and the prospects of an AI strategy in the company.

Mining Object Lifecycle Processes

Challenges, Concepts, Use Cases

at ��ҹ̽��, Germany. The DBIS team is pleased to congratulate Marius on his successful PhD defense (Dr. rer. nat.) and wishes him all the best for his future career. 

Marius was a doctoral student at the Institute for Databases and Information Systems (DBIS). Over the past few years, he has co-supervised the institute's export modules. He has also played a leading role in the two BMBF projects SoftProc  and ProcMape. His dissertation dealt with the mining of object life cycles in the context of object-centric business processes.

Members of the PhD committee

• Prof. Dr. Matthias Tichy (��ҹ̽��, Chairman)
• Prof. Dr. Manfred Reichert (��ҹ̽��, Supervisor & Referee)
• Prof. Dr. Stefanie Rinderle-Ma (TU Munich, External Referee)
• Prof. Dr. Mathias Klier (��ҹ̽��, Committee Member)
• Prof. Dr.  Birte Glimm (��ҹ̽��, Committee Member)

Background to the doctoral program “Data Science & Analytics”

The use of data science methods, in particular artificial intelligence and machine learning, is now standard in many areas of science, business, medicine and society. Advancing digitalization and the digital transformation are reinforcing this trend. Data science methods and the associated analysis processes (analytics) are now increasingly being used for decision-making processes. Systems and algorithms are taking over decisions, including those that have far-reaching consequences or are even irreversible. This shows the urgent need to put data science methods, procedures and algorithms on a solid foundation. Terms such as explainability, traceability, transparency and non-discrimination are being discussed intensively in the field of AI, among others. The Cooperative Doctoral Program (KPK) will contribute to application-oriented research in the field of Data Science & Analytics by going beyond transparency and explainability.

Trust in data science methods naturally presupposes explainability and traceability. However, these terms are aimed more at the period after the application of data science methods. However, trusting a method usually requires guarantees about the expected behaviour of the method before and during its application. This requires a quantitative investigation of the methods, the specification of key influencing factors, sensitivities and dependencies. This also allows the control of procedures and algorithms and - depending on the application - their monitoring and control , possibly again by procedures and algorithms.

Evaluating XAI methods in the Context of Predictive Process Monitoring

at ��ҹ̽��, Germany. The DBIS team is pleased to congratulate Ghada on her successful PhD defense (Dr. rer. nat.) and wishes her all the best for her future career.

Ghada was a doctoral student at the Institute of Databases and Information Systems in the cooperative doctoral program ‘Cognitive Computing in Socio-Technical Systems’ and was supervised by Prof. Manfred Reichert. In her dissertation, she focused on the evaluation of methods of explainability in the context of predictive process monitoring.

Members of the PhD committee

• Prof. Dr. Matthias Tichy (��ҹ̽��, Chairman)
• Prof. Dr. Manfred Reichert (��ҹ̽��, Supervisor & Referee)
• Prof. Dr. Mervat Abu-Elkheir (German University of Cairo, External Referee)
• Prof. Dr. Rüdiger Pryss (University of Wuerzburg, External Referee)
• Prof. Dr. Mathias Klier (��ҹ̽��, Committee Member)
• Prof. Dr. Hans Kestler (��ҹ̽��, Committee Member)

DBIS doctoral student Tobias Jahn, together with Philipp Hühn and Maximilian Förster from the Institute for Business Analytics, received the Best Student Paper Award at the 19th International Conference on Design Science Research in Information Systems and Technology (DESRIST 2024) for his paper "Wasn't Expecting that - Using Abnormality as a Key to Design a Novel User-Centric Explainable AI Method":

Tobias Jahn, Philipp Hühn, Maximilian Förster

19th International Conference on Design Science Research in Information Systems and Technology (DESRIST 2024)

About the content of the award-winning work: The opacity of AI systems is a major obstacle to their use. Explainable AI (XAI) methods that automatically generate counterfactual explanations for AI decisions can increase user confidence in AI systems. Here, abnormality of explanations is an essential property that is not yet sufficiently taken into account by existing XAI methods. Against this background, a new approach for generating counterfactual explanations based on the abnormality of features was developed as part of this research work. In this paper, the approach was demonstrated in a realistic environment and its effectiveness was subjected to a user test. The results show that the developed approach generates explanations that are both more trustworthy and more helpful in explaining the factual situation.

DESRIST, which was held this year in Trollhättan, Sweden, is both renowned and rich in tradition in the field of business informatics. Different disciplines, united by a focus on design science research, are dealt with in different tracks.

Burkard completed his doctorate at DBIS on December 22, 2022 on the topic of "Data Understanding in Predictive Maintenance Applications in the Age of Industry 4.0". He was supervised by Prof. Dr. Manfred Reichert and .

A total of eight former doctoral students of Prof. Reichert now hold a professorship at a university or university of applied sciences.

Marius Breitmayer, Lisa Arnold, Marko Pejic, Manfred Reichert: . M. Weske, J. Michael (Hrsg.): Modellierung 2024, Lecture Notes in Informatics LNI 83, März 2024.

The Modeling Conference is the central scientific event of the Modeling Committee of the German Informatics Society. This community has been meeting for more than twenty years to discuss modeling issues across the entire spectrum of computer science. These include fundamentals, methods, techniques, tools, domains and applications.

Combining Ecological Momentary Assessment and Mobile Crowdsensing in eHealth and mHealth

at ��ҹ̽��, Germany. The DBIS team is pleased to congratulate Robin on his successful PhD defense (Dr. rer. nat.) and wishes him all the best for his future career.

Robin was a joint doctoral student of the Institute for Databases and Information Systems (Director: Prof. Dr. Manfred Reichert) and the Institute of Psychology / Department of Clinical Psychology and Psychotherapy. His dissertation is anchored in the Digital Health focus area of both institutions. His academic supervisors were Prof. Dr. Manfred Reichert, , and Prof. Dr. Harald Baumeister.

Robin Kraft's interdisciplinary work is a very good example of the close cooperation between the Departments of Computer Science and Psychology at the University of Ulm.

Members of the PhD committee:

• Prof. Dr. Matthias Tichy (��ҹ̽��, Chairman)
• Prof. Dr. Manfred Reichert (��ҹ̽��, Supervisor & Referee)
• Prof. Dr. Rüdiger Pryss (University of Wuerzburg, Co-Supervisor & Referee)
• Prof. Dr. Oliver Amft (University of Freiburg, External Referee)
• Prof. Dr. Fabian Prasser (Charité Berlin, External Referee)
• Prof. Dr. Harald Baumeister (Universität Ulm, Committee Member)
• Prof. Dr. Franz Hauck (Universität Ulm, Committee Member)

Legacy software systems are often used to support business processes, which are outdated and hide the process logic in the code. Vital process adaptations to changing circumstances are generally not adequately supported. The aim of the ProcMAPE project is to enable (partially) autonomous, data-driven adaptations of object-centric processes in software systems. In order to realize this flexibility with the required robustness, efficiency and traceability for users, a holistic approach is required, which is summarized in the project under the acronym MAPE: (1) Monitoring, (2) Analysis, (3) Planning and (4) Execution. The aim is to achieve (1) continuous monitoring and (2) the analysis of running processes, (3) the derivation, evaluation and planning of necessary process adaptations and (4) the technical system implementation of these adaptations (e.g. by adapting program code, database schemas, and electronic user forms). The overall goal is the development and implementation of an automated solution approach for monitoring and analyzing ongoing processes as well as the planning and technical implementation (i.e. execution) of necessary process and code adaptations.

Recently, with SoftProc the two partners successfully completed another project in the SME Innovative program with SoftProc.

The ZEUS Workshop focuses on discussing fresh research ideas in computer science, presenting ongoing work, and establishing a scientific network for young researchers from the DACH region

At ZEUS'24, there were 13 presentations on topics such as process automation, process mining & monitoring, artificial intelligence, cloud computing, and privacy & security. The Institute of Databases and Information Systems (DBIS) contributed with 4 presentations and a keynote at the workshop.

The 18 participants discussed ideas from different thematic areas and developed novel solutions for emerging challenges.

A big thanks also goes to this year's , without whom the event would have been impossible.

Context-based Probabilistic Prediction of the Driving Behavior 
of Surrounding Traffic Participants

at ��ҹ̽��, Germany. The DBIS team is pleased to congratulate Florian on his successful PhD defense (Dr. rer. nat.) and wishes him all the best for his future career. 
Florian was an external doctoral student of Mercedes-Benz AG at the Institute of Databases and Information Systems (DBIS). Florian’s PhD advisor and promotor was Prof. Dr. Manfred Reichert. The external reviewer for the doctoral colloquium was  from the  at TU Darmstadt. The dissertation was also co-supervised by Dr. Jochen Hipp and Dr. Joachim Herbst from Mercedes-Benz AG.
Florian is the sixth doctoral student at Mercedes-Benz AG to successfully complete his doctorate at DBIS under the supervision of Prof. Reichert.

Members of the PhD committee: 
•    Prof. Dr. Frank Kargl (��ҹ̽��, Chairman)
•    Prof. Dr. Manfred Reichert (��ҹ̽��, Supervisor & Referee)
•    Prof. Dr. Steven Peters (TU Darmstadt, External Referee)
•    Prof. Dr. Hans Kestler (��ҹ̽��, Committee Member)
•    Prof. Dr. Friedhelm Schwenker (��ҹ̽��, Committee Member)

Engineering Process Enactment and Coordination in Data-Centric Process Management Systems

at ��ҹ̽��, Germany. The DBIS team is pleased to congratulate Sebastian on his successful PhD defense (Dr. rer. nat.) and wishes him all the best for his future career.

Sebastian’s PhD advisor and promotor was Prof. Dr. Manfred Reichert. (, University of Potsdam) acted as external examiner at the doctoral colloquium.  Until 2022 Prof. Weske was chairman of the steering committee of the BPM conference series.

Sebastian carried out his research work in the PHILharmonicFlows. He has also made a significant contribution to setting up the DBIS Business Process Modelling Lab and has contributed to several third-party funded projects (BMWK, State of Baden-Württemberg).

Members of the PhD committee:

• Prof. Dr. Matthias Tichy (��ҹ̽��, Chairman)
• Prof. Dr. Manfred Reichert (��ҹ̽��, Promotor & Referee)
• Prof. Dr. Mathias Weske (Hasso-Plattner-Institute, University of Potsdam, External Referee)
• Prof. Dr. Birte Glimm (��ҹ̽��, Committee Member)
• Prof. Dr. Mathias Klier (��ҹ̽��, Committee Member)

The goal of the project was the development of software-supported procedures that "observe" users during the execution of a process with the software and write this information in fine granularity in an event stream or an event log. With the resulting online process analysis, dedicated processes are determined retrogradely from a running application and the use of the application is documented across different users, locations or points in time.

Scientific and technical starting point

The common approaches to obtaining models for business processes have in common that the software systems to be analyzed want to transfer as accurate an event stream as possible to a process mining tool. In 2016, the IEEE defined the XES (Extensible Event Stream) standard format for event streams. This standard is supported by common process mining tools and allows, among other things, free extensibility using custom extensions to map specifics of the company or its business processes, as well as specifics of the software systems and to transfer the event streams to the process mining tool. For the large standard systems with constant data models (e.g. SAP, Microsoft Dynamics, Oracle E-Business Suite), there are already preconfigured export routes that extract all the necessary event stream information from the application for a classic procurement process, for example.

However, there are no simple export routes for standard applications or industry solutions, let alone for individual software. Since in most cases the individual software solution does not necessarily use a workflow or process engine in the background that is capable of providing helpful audit data with just a few extensions, the only option in many cases is to look at the changes in the data model. However, as discussed above, this involves a lot of time wasted on interviews, analyses and assumptions.

Key results

As part of the research and development work, it became clear that the greatest innovation within the work packages lies in the identification of processes based on user interactions, identification of the "real" data model and the DataActivityLog. As long as a process description with BPMN or PHILharmonicFlows can be created from this, the further innovations in the area of conformity and compliance were only minor. For this reason, the focus in the further course of the project was much more on high quality in the semi-automatic recognition of the business processes implemented in the application.

PITSS

The results and findings achieved in the project will be incorporated into the upcoming release of the PITSS.CON product and will therefore be available to the customer base as an update or for new customers as a stand-alone product.

DBIS

DBIS supported PITSS in the project with the latest scientific and research findings and contributed its expertise in the field of business process management to the project. It also designed and evaluated the algorithms and procedures for determining the conformity and compliance of implemented processes with a given target process model or with existing compliance rules. DBIS has also published the particularly original results of its research work in the SoftProc project in prestigious publications and presented them at 12 international conferences and workshops.

On 22 December 2022, Burkhard Hoppenstedt successfully defended his PhD thesis entitled

Data understanding in predictive maintenance applications
in the age of Industry 4.0

at ��ҹ̽��, Germany. The DBIS team is pleased to congratulate Burkhard on his very successful PhD defence (Dr. rer. nat.) and wishes him all the best for his future career.

Burkhard's supervisors and referees were and Prof. Dr. Manfred Reichert.  

from the Research Lab "Knowledge Management & Discovery" at Otto von Guericke University Magdeburg participated in the doctoral colloquium as an external referee.

The dissertation project was carried out in cooperation with .

Members of the PhD committee:

• Prof. Dr. Hans Kestler (Universität Ulm, Vorsitz und Protokoll)
• Prof. Dr. Manfred Reichert (Universität Ulm, Betreuer & Gutachter)
• Prof. Dr. Rüdiger Pryss (Universität Würzburg, Betreuer & Gutachter)
• Prof. Myra Spiliopoulou (Otto-von-Guericke-Universität Magdeburg, Externe Gutachterin)
• Prof. Dr. Matthias Tichy (Universität Ulm, Wahlmitglied)
• Prof. Dr. Franz Hauck (Universität Ulm, Wahlmitglied)