To path indicator To navigation To quick access To footer with other services
To content
Logo Enterprise Computing
Department of Computer Science
[Translate to English:] © Oliver Schaper​/​TU Dortmund
Enterprise Computing

Current Topics for Theses

The possibility of writing a Bachelor's or Master's thesis is explicitly not limited to the following topics. These merely give an indication of the range of topics.

Topic overview

  • Development of a BPM lab: installation of BPM systems, implementation of demo processes, documentation (Prof. Christian Janiesch).
  • Advanced analytics in a socio-technical context: Effects of predictive and prescriptive analysis results on the cognition and emotions of users of information systems can provide insights into the acceptance and diffusion of information systems. (Philip Stahmann)
  • AI Mindfulness, Human-AI Interaction and Hybrid Intelligence: People have to interact with AI systems in many places. This may require new processes and forms or patterns of collaboration. (Prof. Christian Janiesch)
  • Interaction system development: Human-machine interaction plays an increasing role in IIoT, and to ensure productive work, cognitive load must be minimized, among other things. The aim of the work is the formulation of design criteria and the prototypical creation of such a system for decision-makers in the context of IIoT, taking cognitive load into account. (Maximilian Nebel)
  • Biases in machine learning: Machine learning is almost always based on data created by humans. This means that this data is subjective and the learned knowledge of the "artificial intelligence" is biased. (Prof. Christian Janiesch)
  • Explainable AI: Processing and working with the most common XAI techniques and XAI tools. If necessary, implementation of visual components for new use cases such as process management. (Prof. Christian Janiesch)
  • Artificial intelligence in Industry 4.0: The application potential of artificial intelligence for optimizing operational production is broadly diversified, and practical implementations are constantly increasing, e.g. in robotics, process automation or machine communication. (Philip Stahmann)
  • Computer vision in logistics: As part of the work, AI-supported image recognition methods are to be used to identify anomalies in the material of the pallets. For this purpose, real image data is classified with the help of practical expert knowledge. (Maximilian Nebel)
  • Levels of automation in intelligent process automation: Intelligent systems are used in a range of automation levels - from robotic process automation to autonomous agents - depending on the complexity of the process and the use of AI. (Seyyid A. Ciftci)
  • Implementation of autonomous agents: AI-supported agents enable autonomous, efficient decisions in complex processes that react flexibly to dynamic changes and requirements. (Seyyid A. Ciftci)
  • Procedure model for process mining: Developing a model for the procedure in process mining projects based on standard procedure models such as CRISP-DM. (Prof. Christian Janiesch)
  • Development of a process mining lab: Installation of process mining systems, implementation of demo processes, documentation (Prof. Christian Janiesch).
  • Management of RPA: Practical relevance, profitability, selection decision, implementation, maturity models, success stories and technological debts of RPA. (Prof. Christian Janiesch)
  • Machine Learning and RPA: Development of RPA bots that bring flexibility to RPA application scenarios through the integration of machine learning processes and thus overcome existing limitations of RPA. (Prof. Christian Janiesch)
  • Development of an RPA lab: installation of RPA systems, implementation of demo bots, documentation (Prof. Christian Janiesch).
  • Implementation of RPA: Modeling and automation of a BPMN model to optimize human-machine interaction. (Seyyid A. Ciftci)
  • The development of sensor technology and the associated use in industrial production enable access to status data in real time. This creates potential in data analysis, such as the detection of anomalies without time delays. (Philip Stahmann)
  • Learning classifier systems are able to classify data on the basis of evolutionary algorithms. This results in potential for real-time classification, e.g. for anomaly detection. (Philip Stahmann)
  • The availability of high-quality training data is a common problem in machine learning in operational practice, including real-time data classification for anomaly detection. Few shot learning approaches try to achieve the best possible training results for artificially intelligent models with little data. (Philip Stahmann)
  • Status Quo Anomaly Detection in IIoT: As part of a systematic literature review, the role that anomaly detection plays and will play in the future for the practice of IIoT will be presented. (Maximilian Nebel)
  • Generation of an overview and collection of anomaly types in IIoT sensor data: In order to enable decision-makers in production companies to draw the right conclusions from detected anomalies, it is necessary to differentiate them. The aim is to identify and describe generalizable anomaly types in the literature and data. (Maximilian Nebel)

Theses can also be written in cooperation with Queenland University of Technology (QUT) in Brisbane, Australia. Thesis topics will be agreed upon between the candidate, Prof. Rosemann and Prof. Janiesch and will focus on the "algorithmic" or "trusted enterprise", emphasizing the interface between computer science and management, especially in the areas of digitalization and artificial intelligence. If interested, please contact Prof. Janiesch directly outside the usual thesis application cycle.

To top of page