MULTI-OBJECTIVE STRUCTURAL OPTIMIZATION - A REVIEW OF THE GENETIC ALGORITHM METHODS

There are numerous means of enhancing ship structure in an attempt to satisfy both the shipyard and the ship owner requirements. These means are addressed as a ship's attributes and they are regularly contrary in meaning, such that the improvement in one cannot be achieved without making some other worse. Typically these are e.g. cost, weight, reliability, safety and stability, which can be minimized or maximized to improve the final design. A compromise obviously has to be made between them, depending upon the importance of each attribute. Evolutionary algorithms have proven their worthiness in a great variety of practical engineering problems, handling many variables, constraints and objectives. However, the need to improve them is always present. This has recently been done by hybridization, combining different approaches in order to get the best of them. To devise an algorithm which will be applicable to all problems is impossible. In this review, for the purpose of directing the experts entering in the field of structural optimization, only a few of the most important approaches and multi-objective algorithms have been presented.