Regression Models for Estimating Cruise Ship Main Dimensions in the Early Stages of Design

Cruise ships are complex object to design, requiring the knowledge of multiple aspects of naval architecture and marine engineering. During the concept phase of design, the first step consists in estimating the ship main dimensions as a function of the desiderata of the ship owners. Therefore, it is necessary to provide designers with suitable tools for a fast estimation of the principal ship parameters. The literature reports the availability of simple regression methods and the application of modern machine learning techniques for assessing this issue. The present paper focuses on the traditional regression methods, providing for the first time on cruise ships multiple linear regressions that could be directly applied by designers in the early stages of design. Starting from a database of modern cruise vessels, models have been developed not only for the main dimensions but also for other relevant preliminary attributes like the deadweight, displacement and installed power. The results highlights how the multiple linear regressions provide a good estimate for all the considered parameters, giving to designers an instrument easy to apply.