Methodology for Optimizing Commonality Decisions in Multiple Classes of Ships.

Commonality;Ship Design;Evolutionary Algorithms;Platforms;Multicriterion Optimization;Platforms;Naval Architecture and Marine Engineering;Engineering;Naval Architecture & Marine Engineering

Corl, Michael J.Ann Arbor ;

University of Michigan

关键词: Commonality;    Ship Design;    Evolutionary Algorithms;    Platforms;    Multicriterion Optimization;    Platforms;    Naval Architecture and Marine Engineering;    Engineering;    Naval Architecture & Marine Engineering;   

Others  :  https://deepblue.lib.umich.edu/bitstream/handle/2027.42/55684/mcorl_1.pdf?sequence=2&isAllowed=y

瑞士|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

A methodology is presented for optimizing commonality decisions in multiple classes of ships with consideration of mission effectiveness, construction costs, and the savings that will result from the use of commonality.Currently the automobile and consumer product industries apply commonality methodologies in developing families of products that use a common platform. From the use of common components, a manufacturer can realize significant savings in a given line of products. If properly applied, the shipbuilding industry could also see significant cost savings from the use of commonality.An approach to the optimal use of commonality has been developed for the design of different classes of ships using a platform of common components.Unlike previous methodologies, this optimization process takes into explicit consideration the savings associated with the use of commonality. A ship synthesis model was adapted to create ship designs from independent design variables in order to predict the mission performance and calculate construction costs for each design. This was linked with a fleet commonality savings model reflecting savings from bulk purchases of equipment and the shipbuilding learning curve. The commonality items considered were the cruise engines, ship service generators, weapons, superstructure, and midship hull blocks. A multicriterion evolutionary algorithm is utilized to efficiently search the design space for feasible designs of multiple classes of ships. The criteria studied are the mission performance of ship class A/average cost, performance of class B/average cost, and net fleet savings through commonality.Through the use of dominance sorting and genetic operators, offspring solutions are developed in order maximize multiple design objectives from one generation to the next. The resulting solutions form a discrete Pareto front that allows the designer to choose a suitable set of designs to best meet the mission needs of multiple ship classes.The design methodology is demonstrated and tested through a problem modeled on the U.S. Coast Guard Deepwater High and Medium Endurance Fleet missions to enable interpretation of results.Results demonstrate that the methodology presented will prove to be a valuable tool in making good commonality decisions.