Answers are grounded in this paper and cite their sources.
This study proposes an integrated digital design optimization framework for ship structural engineering that operates directly within an AVEVA Marine product model. The framework targets a compact set of design variables comprising plate-group net thicknesses, double-bottom girder thickness, web-frame spacing, and stiffener spacing. These variables are linked to classification-rule, geometric-compatibility, manufacturability, and allowable-stress constraints, while finite-element-derived second-order response surfaces provide rapid estimates of structural weight, production cost, and critical stress responses. A surrogate-assisted non-dominated sorting genetic algorithm II (NSGA-II) is used to minimise cargo-hold structural weight and estimated production cost simultaneously. The methodology is demonstrated on the cargo‑hold region of a multipurpose cargo ship using a parametric 3D AVEVA model and finite‑element‑based response surfaces. A 128-point Latin-hypercube design of experiments was used to construct the response surfaces, for which ten-fold cross-validation produced R^2values of 0.99 for structural weight, 0.98 for production cost, and 0.95-0.97 for the stress responses. The selected design reduced cargo-hold structural weight from 723.20 t to 684.80 t (5.3%) and estimated production cost from 3,171,305 to 2,710,012 USD (14.6%) relative to the baseline configuration. The feasible Pareto set is presented in the formal weight-cost objective space. The final design was selected using a constraint-screened engineering compromise rule rather than an automated knee-point method; critical stress utilization was used as a post-optimization decision indicator, not as a third objective. The selected configuration was transferred to mapped AVEVA Marine objects through scripted attribute updates, regenerated, and checked using a detailed finite-element model. The governing hatch-coaming stress utilization was 0.893, corresponding to a deterministic safety factor of approximately 1.12. The framework is intended to support preliminary production-oriented structural design improvement; final implementation requires project-specific verification of governing local details, including applicable fatigue, buckling, fabrication-tolerance, and classification-rule checks.
No takes yet. Share an insight, caveat, or question.
TianQi Fei (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: