Designing ships involves intricate layouts and multifaceted systems—ranging from mechanical to operational—that must be interdependent and thus precisely arranged. Traditional automated tools, though effective, are often too resource-intensive to be feasibly employed during the critical early stages of design. This paper builds on prior work that introduced an innovative solution: a network-based, architecturally normalized current representation, which offers a computational method to predict system arrangements in two dimensions without generating detailed vessel models. Our method’s advantage lies in its ability to guide early stage design decisions, thereby optimizing the use of subsequent, more resource-intensive design tools. This study extends the method to a three-dimensional framework, capturing more nuanced system-to-system interactions and yielding more realistic ship arrangements. A methodology was proposed to support this three-dimensional extension and demonstrate its applicability through a case study focused on the conceptual design of a naval frigate.
Read full abstract