الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 110 |  |  |
| | | from | 110 | | |
Abstract
The need for new ships of diverse typesis rising day after another, and whether it be commercial or combatant ships, every newly built ship has to fulfill some distinct purposes and satisfy all stakeholders’ objectives when building that ship, where these objectives usually conflict with each other, as in order to realize certain objective, another may be violated. Furthermore, after the choice of a concept design for the ship and carrying on with the following design stages, some changes may be decided in order to improve specific aspects or even one of the objectives since this was not considered earlier. Therefore, the process of an optimization-based conceptual design, regarding the structural arrangement design of the ship based on the objectives provided by the stakeholders and imposed criteria related to the maximum hull girder stresses, is presented in this thesis, in an attempt to give a simplified overview of how such process can be applied in that early stage, and how it is able to help the decision-making during that ambiguous stage where none of the involved individuals has an appropriate visualization of this concept design. This optimization process is presented herein, where it depends on two major procedures, these are the parametric modeling using ‘ANSYS APDL’ parametric design language and the utilization of a ‘Particle Swarm Multi-Objective Optimization’ algorithm using ‘MATLAB.’ The combination of these notions, parametric modeling, and optimization, gives the capability to quickly alter the features’ variables affecting the structural arrangement of the concept design quickly without the need for exhaustive remodeling. In addition to that, by introducing the required stakeholders’ objectives into the optimization algorithm, it keeps altering the parametric model’s variables by its turn, aiming to find create a ‘Pareto Front,’ which is a set of the achieved optimized alternatives for the concept design, where each alternative gives a specific level of compromise to each objective. Accordingly, the thesis walks through the main tasks performed in order to reach that output, eventually. At first, the parametric model is built by ANSYS parametric design language using scripting instead of the user interface, starting with the definition main design, then passing by the construction of the finite element model; material, sections, element types, meshing, loading case, and boundary conditions. Afterward, the algorithm is established along with its different features along with the implementation of the constraint (criteria) function. A point to note is that this procedure is just a decision-making support approach where certainly, detailed analysis is required after opting for a concept design. This thesis presents the study of such a procedure along with presumed assumptions and simplifications and presents some suggestions for potential future work in order to improve what was done here.