Master thesis Master's Thesis: Capturing Implicit Process Knowledge Using AI

The KIProWahl research project focuses on AI-based process analysis for the selection of business software in SMEs. The aim is to reduce the time required for process elicitation using an LLM-based chatbot. However, a large part of process knowledge in production environments consists of tacit experience. In this thesis, you will develop and evaluate LLM-based interview strategies (e.g., storytelling, problem-centered approaches) to systematically and role-specifically externalize this hidden knowledge from shopfloor employees.

Your tasks

• Identification and analysis of existing interview methods for knowledge extraction
• Development of specific LLM prompt strategies, e.g., by integrating linguistic approaches to interpret sentence semantics
• Conception of role- and use-case-specific question generation for the chatbot
• Implementation and testing of the strategies in the Python-based software demonstrator
• Evaluation of the methods regarding the quantity and quality of the captured tacit knowledge

Your profile

You are studying one of the following subjects:

• Computer Science, Business Informatics
• Mechanical Engineering, Industrial Engineering
• or comparable

You are interested in Artificial Intelligence, Process Management, and Human-Machine Interaction. Additionally, you have initial experience in Prompt Engineering and basic knowledge in software development (ideally Python). Good written and spoken German and English skills are required.

We offer

• independent work
• flexible working hours
• well-equipped workplaces
• home office by arrangement
• execution of experiments