Optimization Methods for the design and production of Naval Structures

In the past decades the marine industry has undergone significant evolution and the production of large passenger ships faced dramatic changes. The size, the complexity and the security standards of the ships have increased and the ship-owners have become less and less willing to wait once the order is placed. In the meantime, due to growing intense competition, the shipyards have had to improve their efficiency and master their production costs: they progressively moved from manufacture to automated production processes. As a consequence, the shipyards have now to meet this difficult challenge: produce more complex ships, cheaper and faster.

In the first part of the thesis we consider the structural optimization problem that arises in the early design phase of a project. Given a vessel overall dimensions and form, structural optimization consists in defining the scantling of the structure’s constitutive elements so as to minimize its total weight or cost, while taking weight, robustness and security issues into account. Designers have to make the most adequate choices within a very short period of time. The decisions made during the design phase have a major impact on the final structure and on its production cost. We propose new algorithms to compute near-optimal solutions of the discrete structural optimization problem.

In the second part of the thesis, we turn to the building process of large ships and we present new methods to improve production facility management of shipyards. The ship building process requires the production and the assembly of tens, or even hundreds of thousands of steel elements. We define a space and time allocation problem that arises in assembly halls producing large building blocks and we develop algorithms for its solution. A successful implementation of a flexible and robust application is nowadays in use at a shipyard.