Bachelor and Master Theses

This thesis proposes developing and evaluating tooling that uses LLMs for automated test augmentation aimed explicitly at industrial control software. With the increasing complexity of control systems and stringent requirements in industries like manufacturing and transportation, traditional manual testing of control software is often insufficient for detecting edge cases. This study aims to apply LLM-based methods to automatically generate new test cases for a specific type of control software, PLC-based systems, identifying missed edge cases, improving test coverage, and reducing manual testing. By integrating LLMs into testing workflows and conducting experiments on industrial control systems, this research seeks to validate the effectiveness of LLM-based test generation for improving manual testing. 

understanding of industrial control systems, familiarity with software testing, programming experience, knowledge of LLMs