Tekoäly rinnakkaissuunnittelun arvioinnissa : tapaustutkimus erään aluksen suunnittelusta verrattuna VDI 2221 -standardin ohjeistukseen
Abstract
Marine vessel design is often described as the “art of balancing compromises.” The complexity of naval engineering and challenging project timelines demand increasingly sophisticated project organization and deeply layered concurrent engineering. This research investigates whether the combination of artificial intelligence and VDI 2221 systematic design guidelines can be used to audit how a design engineering team proceeds in a project. Two research methodologies are compared: a traditional participant observation and a fully automated process using a Retrieval-Augmented Generation (RAG) -enhanced artificial intelligence. The case study being examined is the design engineering project of a water taxi boat destined for a major European city. The study reveals that the water taxi project is an example of a highly agile concurrent engineering process, but that the project also exhibits violations of the VDI 2221 guidelines. Most notable of these is the practice of “stage-jumping”, which means, for example, advancing into detailed design before the conceptual solution is mature. The study demonstrates that a suitably trained AI can create a high-level view of a project in a fraction of the time a human requires to do the same. It is plausible that AI could even be used for real-time project situational feedback. However, for that to be possible, a project should follow a systematic process such as VDI 2221 to provide a contextual framework for the AI analysis. In conclusion, the combination of AI and VDI 2221 could be an interesting foundation for the development of novel project management tools.
Fields of Study & Subjects
Product Development and Customization
Systems Engineering Methodologies and Applications
Design Education and Practice
Publication details
- Journal / source
- LUTPub (LUT University)
- Publication date
- 1 January 2026
- Language
- EN
- Document type
- Thesis