Data-Driven Vessel Design
Data-driven operational profiles used for design input for new HTV designs
F. Hartjes (TU Delft - Mechanical Engineering)
J.L. Gelling – Mentor (TU Delft - Mechanical Engineering)
E.B.H.J. van Hassel – Graduation committee member (TU Delft - Mechanical Engineering)
Ko Stroo – Mentor (Ulstein Design and Solutions B.V.)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
Early-stage vessel design is traditionally based on assumed operational profiles, introducing uncertainty and potentially leading to suboptimal design decisions. This study presents a data-driven approach to vessel design by deriving operational profiles from Automatic Identification System (AIS) data, with a specific focus on Heavy Transport Vessels (HTVs).
A structured methodology was developed to process, clean, and segment AIS data into operational activities, which were subsequently used to construct representative operational profiles. In contrast to existing studies, this research evaluates the impact of data cleaning and repairing procedures on both the resulting operational profiles and their impact on vessel design.
The results show that data cleaning and repairing can significantly influence operational profiles when analysing individual vessels. However, when multiple vessels are combined into fleet-level operational profiles, these effects decrease, indicating that the importance of cleaning and repairing is reduced at fleet level. Furthermore, the impact of data-derived operational profiles on early-stage vessel design is found to be limited. For HTVs in particular, key design parameters are primarily constrained by monopile dimensions, reducing the sensitivity of design outcomes to variations in operational profiles.
This study demonstrates that AIS-based operational profiling is a reliable method for supporting early-stage vessel design. It further shows that increasing dataset size is more effective in improving robustness than increasing data cleaning complexity, thereby providing practical guidance on the required level of data-processing precision. Overall, this research provides a framework for integrating AIS data into early-stage vessel design and demonstrates the influence of data cleaning and repairing at multiple levels, as well as the impact of data-derived operational profiles on vessel design.