Comparison of Meta-Heuristic Optimization Algorithms in Ship Hull Cutting Plan Generation

The optimization of ship hull dismantling is crucial for advancing the efficiency and safety of the ship recycling industry, particularly in green facilities. This paper presents a comparative study of meta-heuristic optimization algorithms-Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Simulated Annealing (SA)-for generating an optimized cutting plan for end-of-life (EOL) vessels. The fitness function developed aims to minimize the cutting area while imposing constraints on block mass and center of mass displacement, ensuring safe and efficient dismantling. Our study utilized CAD models and Fusion 360 API for data derivation, ensuring precise optimization inputs. Results indicate that GA and SA achieved robust performance, effectively balancing exploration and exploitation, and converging to optimal solutions with cutting areas around <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$9.1 \times 10^{6} {cm}^2$</tex>. In contrast, PSO struggled with the high-dimensional search space, failing to converge effectively. The GA demonstrated efficiency with a weighted random initialization and tournament selection, while SA benefited from a high initial temperature and an exponential cooling schedule. This study underscores the potential of meta-heuristic algorithms in enhancing the automation and safety of ship dismantling processes.