Global optimization algorithms in naval hydrodynamics

Real life optimization problems are much more complicated than the analytical test functions usually adopted for estimating the performances of new optimization algorithms. For a numerical optimization procedure to be useful in teh design spiral, it is essential to understand the peculiar features of the design problem and chose a suuitable numerical technique among a multitude of methodologies. Unfortunately, although the numner of papers related to numerical ship design is rapidly growing, there is still the need of testing well assessed algorithms in the specific ship design applications, expecially in the context of global optimization techniques. The last decade has seen a proliferation of papers on ship optimization, ranging from building cost of the ship to their hydrodynamic and structual characteristics. All these problems are highly constrained and complex. Local optimization techniques tend to get trapped al local minima without finding usually a much better global optimum. Global optimization (GO) algorithm aim instead at finding the global optimum of the objective function, indipendent from the chosen starting point.