Design optimization of merchant ships at an early stage using genetic algorithms

Currently, engineering designers are working to automate the design process. Using tools to support decision-making, engineers can explore a wider range of options and reduce the time to an initial design. An area which is prominent in the literature is the dimensioning of ships in the early stages. This is a complex part of naval design. Because shipbuilding is a competitive market, the construction cost of a ship is a vital part of the sale price. The objective of this work is to assess the main dimensions of various types of merchant ships by minimizing construction costs with computational methods. The difficulty presented by ship dimensioning is that it is a restricted optimization system of a non-linear type. In this work, to solve it, two mathematical methods have been applied that adapt to this type of problem; the genetic algorithm (GA) method and the sequential quadratic programming (SQP) method. A practical application of the calculation of the main dimensions of a new ship to be designed through the utilization of the procedures described in this work. The advantages of using genetic algorithms for ship dimensions are shown. A graphical interface has been developed to solve the problem of optimal dimensioning of a ship using AppDesigner integrated with Matlab.