Automated requirement and test case generation with Generative AI for MBSE in the context of Formula Student
- Subject:Testing of CPS Variants and Versions
- Type:BA
- Date:ab 05 / 2025
- Tutor:
Automated requirement and test case generation with Generative AI for MBSE in the context of Formula Student
Context
Specifications and regulations are key sources of information in the development of complex technical systems. However, these documents are often long and difficult to keep track of. In the context of Formula Student, numerous technical requirements must be derived from the extensive set of rules and checked. Modern generative AI models, in particular transformer-based models such as GPT, offer new possibilities for automatically evaluating such documents and deriving requirements and test cases from them. The aim is to integrate this approach into an existing Model-Based Systems Engineering (MBSE) tool such as PREEvision.
Objectives
- Analysis of a set of rules (Formula Student) with a generative AI model
- Automated extraction of technical requirements from the set of rules
- Generation of suitable test cases to validate the requirements
- Automatic transfer of the requirements and test cases to an MBSE tool (PREEvision)
Prerequisites
- Interest in generative AI and NLP (Natural Language Processing)
- Basic knowledge of Python, ideally experience with Transformers (e.g. HuggingFace)
- Understanding of technical systems and specifications