Multi-objective crashworthiness optimization design of ship side structures considering collision-parameter uncertainty

Ship collision processes are affected by environmental disturbances and human operations, resulting in random fluctuations in collision angle, impact velocity, and vertical impact location. Conventional structural crashworthiness optimization is generally based on deterministic collision analyses, which may lead to insufficient reliability under realistic and complex collision scenarios. To address this issue, this study develops a rapid damage-assessment and structural-optimization framework that explicitly accounts for collision-parameter uncertainty. Collision angle, impact velocity, and vertical impact location are selected as the key random variables, and the effects of representative probability distributions on the probabilistic damage assessment of the ship-side structure are investigated. Single-variable and multivariable coupled uncertainty analyses are performed to reveal the effects of distribution characteristics on indentation, striking-bow intrusion volume, component damage volume, and dissipated energy, thereby demonstrating the necessity of considering parameter uncertainty in ship collision damage assessment. Based on the probabilistic damage-assessment model, a multi-objective crashworthiness optimization of the side structure is conducted by integrating the uncertainty-aware rapid prediction method with multi-objective particle swarm optimization. The dimensional parameters of shell plating, deck plating, and stiffeners are selected as design variables, whereas structural weight and a composite expected-damage index are adopted as the optimization objectives, subject to manufacturing and crashworthiness constraints. The resulting Pareto-optimal solution set characterizes the trade-off between structural weight and crashworthiness. The TOPSIS-selected compromise design reduces structural weight by 3.61% while maintaining the composite expected-damage index at a level comparable to that of the baseline design. At an indentation of 4 m, the dissipated energy of the optimized structure is approximately 6.53% higher than that of the baseline design. Finite element simulations further verify the feasibility and effectiveness of the proposed optimization framework. The present study provides a new approach for the safety-oriented design and crashworthiness enhancement of ship structures considering collision-parameter uncertainty.