Recent works have introduced the concept of “digital twin” in the maritime industry as a system that integrates physics-based simulation models with artificial intelligence techniques and data gathered from field sensors. Reduced order modeling (ROM) is a technique used to decrease the computational time of high-fidelity simulation models when integrated with machine learning. This presentation will delve into the significance and efficacy of artificial intelligence within the maritime domain, illustrating an industrial application of ROM for the fast prediction of vessel resistance by leveraging hydromechanics data. Additionally, it will highlight how businesses can extract value from machine learning models and how maritime engineers utilize these models to expedite engineering calculations. Finally, strategies for mitigating risks while ensuring privacy and security through cloud computing will be discussed.