An intelligent and engineering-oriented optimization framework for ship pipe collaborative layout design in realistic 3D compartment environments

Ship piping system layout in complex compartments or confined spaces is a critical task in the preliminary stage of ship and offshore engineering design, as its quality directly affects system safety, operational efficiency, and maintainability. However, most existing approaches rely on simplified geometric models or idealized regular environments, making it difficult to simultaneously satisfy strict engineering constraints and large-scale multi-pipe collaboration in realistic 3D curved-surface compartments. To overcome these bottlenecks, this study proposes a multi-ray navigation–based intelligent optimization algorithm that integrates a bounding volume hierarchy (BVH) with the Möller–Trumbore (MT) ray-triangle intersection test (BM-MRN). Specifically, a unified parametric space is first constructed over a triangular-mesh representation, where boundary surfaces, equipment, and structural components are encoded into an integrated obstacle–energy spatial zoning model. Within this space, multi-trajectory guided encoding combined with hierarchical multi-ray navigation enables a structured optimization process consisting of interference localization, orthogonal offset, and path reconstruction. To further handle strong inter-pipe coupling, a system-level multi-pipe optimization framework is developed, integrating strategies such as collaborative guidance, distance constraints, computational acceleration, and multi-solution generation. In addition, a composite evaluation function is formulated to balance routing compactness and smoothness under engineering constraints. Case studies on realistic scenarios under different configurations show that BM-MRN demonstrates consistently competitive performance relative to representative methods in layout quality, interference control, computational performance, and engineering feasibility, while producing multiple implementable solutions with reduced reliance on manual route adjustment. The proposed framework provides a practical basis for intelligent layout optimization of 3D piping systems in complex shipboard environments.