Proefschrift
Efficient constraint multi-objective optimization with applications in ship design
Constraint multi-objective optimization with a limited budget for function evaluations is challenging. This thesis tackles this problem by proposing new optimization algorithms. These algorithms are applied on holistic ship design problems. This helps naval architects balance objectives like cost, efficiency, environmental impact, and safety while meeting physical and regulatory constraints.
- Auteur
- R. de Winter
- Datum
- 08 oktober 2024
- Links
- Thesis in Leiden Repository
The main focus of this research is on developing efficient algorithms to identify the feasible Pareto frontier for computationally expensive problems. These novel algorithms use surrogates for expensive functions and evaluate inexpensive functions directly. The innovative multi-point acquisition function moreover allows parallel evaluation of solutions. The algorithms preformance are validated on benchmark problems, after which they are applied to ship design optimization problems.
Emphasizing the importance of early design decisions, the thesis argues for integrating multi-objective optimization techniques early in the process to achieve cost-effective, efficient, and environmentally friendly vessels, aligning with industry sustainability trends. Five different case studies confirm the power of the algorithms in real-world scenarios, bridging the gap between theory and practice.
In conclusion, this research represents a contribution to the field of multi-objective optimization and naval architecture, paving the way for innovative, sustainable, and cost-effective design solutions.