Several optimization techniques have been employed to solve ship design problems. The development of Genetic Algorithms (GA) technique with variables uncertainty consideration has also been recently applied. This article discusses the techniques of G.A. and treatment of variables with a degree of uncertainty in some ship design problems. Two case studies are developed. One showing a double hull tanker midship section design and the second case study is a catamaran preliminary design.

Genetic algorithms (GAs) are adaptive methods which may be used to solve search and optimization problems. They are based on the genetic processes of biological organisms. A genetic algorithm allows a population of possible solutions composed of many individuals to develop, under specified rules of selection, a state that minimizes the cost function. The selective mechanisms achieve the changes that determine the evolution of a population across generations. Such changes may occur due to interactions between individuals or due to environmental influences on the individual. It derives three basic mechanisms, crossing or recombination, reproduction and mutation, called genetic operators, to carry out the development of the algorithm. The application of these operators is preceded by a process of selection of individuals best adapted, which uses a function to evaluate the individuals named function fitness or, function setting.

The variable uncertainty is included in the mathematical model and the simulation using the Monte Carlo approach is used simultaneously with the genetic algorithm optimization procedure.

ModeFrontier software is used in both case study and results indicates the method applicability.