Analysis of Generative AI tools and approaches for enterprise-oriented nuclear engineering
The SCK CEN IST Institute, in the framework of MYRRHA and soon in the SMR LFR project, has been accruing a considerable amount of knowledge that is being collected in a so-called Conceptual Design Report (CDR) – currently, a +3500 page living document – and in a significantly large knowledge base of requirements, design decisions, constraints hosted in Polarion spaces. Within SCK CEN it is felt that it would be beneficial to create an intelligent, human-oriented tool able to answer complex questions about such a vast and growing set of information. Therefore, it has been decided to explore whether an internal-only GPT (or a similar generative-AI service) could provide SCK CEN with a useful “expert system” able to provide engineers with sought information, or at least guide to guide them towards meaningful enterprise assets providing the sought information. Such an exploration step is the core idea of this master thesis. The expected outcome is the acquisition of a more clear view on the necessary tools and approaches for creating an enterprise generative-AI tool to assist our engineers in the design steps of nuclear physics components, as well as some preliminary steps in the application of those tools and approaches to the needs of SCK CEN.
The minimum diploma level of the candidate needs to be
- Academic bachelor
- Master of sciences
- Master of sciences in engineering
The candidate needs to have a background in
- Informatics
- Mathematics
- Other
- Physics