Early-stage design assessment of loads such as vertical bending moments can be a critical quantity of interest for design exploration. Traditionally, classification societies’ rules are used to calculate such loads. However, relying solely on these rules for designing new vessels may be insufficient, and conducting direct analyses of a large number of designs to support design exploration is computationally infeasible. Currently, key factors such as wave-induced loads are typically evaluated only in later design stages, where a limited number of promising designs are under consideration. This research explores the potential of harnessing multi-fidelity models for early-stage predictions of wave-induced loads, with a specific focus on wave-induced vertical bending moments. As an initial step in this direction, the vertical bending moment analysis was simplified to consider responses in a regular sea state, where the wavelength matches the vessel’s length. The assessed multi-fidelity models include the application of both linear and nonlinear Gaussian processes and compositional kernels to improve predictions of wave-induced loads, specifically focusing on wave-induced vertical bending moments. The case study focuses on the early-stage exploration of the AXE frigates. Multi-fidelity models were constructed using both frequency- and time-domain methods to evaluate the vertical bending moments experienced by the hull. Finally, a critical reflection is provided on how traditional early-stage design processes can be enhanced by integrating multi-fidelity models.

Early-stage design is the most critical phase in a vessel’s development, as this is when many of the major decisions are made and locked in for the remainder of the design process. Most early-stage design frameworks are tailored to conventional vessels, aiming to explore a broad design space—essentially assessing numerous design variations. However, to evaluate such a wide range of design solutions, these frameworks often sacrifice accuracy in their analysis methods to allow for more design evaluations. Consequently, low-fidelity tools are typically used for early-stage design exploration.
For novel vessel designs, low-fidelity analysis methods are insufficient for accurately assessing performance, as they often fail to capture the new and sometimes complex physics involved. While increasing the fidelity of analysis methods leads to more accurate performance assessment, it also raises computational costs, making it impractical to evaluate a large number of design variations. Multi-fidelity models, which combine lower-fidelity methods with a high-fidelity analysis method, offer a promising solution for enabling higher-fidelity assessments earlier in the design process. Thus, this dissertation builds the architecture of a multi-design architectural framework for early-stage design of novel vessels...

Early-stage design of complex systems is considered by many to be one of the most critical design phases because that is where many of the major decisions are made. The design process typically starts with low-fidelity tools, such as simplified models and reference data, but these prove insufficient for novel designs, necessitating the introduction of high-fidelity tools. This challenge can be tackled through the incorporation of multifidelity models. The application of multifidelity (MF) models in the context of design optimization problems represents a developing area of research. This study proposes incorporating compositional kernels into the autoregressive scheme (AR1) of multifidelity Gaussian processes, aiming to enhance the predictive accuracy and reduce uncertainty in design space estimation. The effectiveness of this method is assessed by applying it to five benchmark problems and a simplified design scenario of a cantilever beam. The results demonstrate significant improvement in the prediction accuracy and a reduction in the prediction uncertainty. Additionally, the article offers a critical reflection on scaling up the method and its applicability in early-stage design of complex engineering systems, providing insights into its practical implementation and potential benefits.

Early-stage design exploration is crucial since most of the major design decision are locked-in and only small design modifications are possible at later stages. To assess the performance of the various design candidates while performing design exploration, there are available methods and tools of various fidelities. These methods can be combined to form a multi-fidelity (MF) framework that guarantees accuracy through the high-fidelity model and achieves faster computational speeds through low-fidelity models. The present study proposes the adoption of information-theoretic entropy to improve a MF design framework based on Gaussian Processes (GPs). Entropy quantifies the uncertainty associated with the prediction of the design space. We propose using this uncertainty metric both as a criterion to determine whether further designs should be sampled to construct a reliable approximation of the design space and as a criterion to establish in which optimization step the optimization of the covariance matrix for the MF-GPs should be performed. The approach was tested to benchmark analytical functions and to a ship design problem of an AXEfrigate. The approach holds potential in practical applications, as it aids in the determination of whether additional resources should be allocated for high-fidelity analysis to support early-stage exploration.

This paper proposes a grey-box modelling approach to predict marine biofouling growth and its effects on ship performance. The approach combines empirical or experimental-based white-box models with data-driven black-box models. First, a white-box model is built to predict ship resistance considering a bare hull. This prediction is based on calm water resistance, wind, waves, and temperature differences. Subsequently, marine biofouling growth is predicted using an experimental model that estimates the level of roughness on the ship hull. Finally, a deep extreme learning machine is used as a black-box model, employing a feedforward neural network technique. To test the approach, a superyacht case study was selected as a category of vessel heavily exposed to fouling. The study used a 2-year dataset obtained through a collaboration with Feadship. Results showed that the black-box approach outperforms the white-box approach in predictive capabilities. However, when the knowledge encapsulated in the white-box model is included in the grey-box approach, the model shows the highest prediction accuracy achieved by leveraging less historical data. This study demonstrates the potential of the proposed grey-box approach to accurately predict marine biofouling growth and its effects on ship performance, which can benefit ship operators and designers in improving operational efficiency and reducing maintenance costs.

The aim of this paper is to discuss the challenges associated with the early-stage design of novel and reliable vessels, and discuss some of the expected benefits of the application of multi-fidelity models in addressing some of their early-stage design problems. Traditionally, early-stage design tools are computationally cheap, but lack in accuracy. However, for the design of novel vessels, these tools are not sufficient. The first part of the paper discusses the challenges associated with the design of novel vessels. The second part of the paper focuses on a literature review on the application of the multi-fidelity models to the design of complex engineering systems. Finally, the most promising methods are identified and discussed.

This paper presents and demonstrates a new design thinking framework for early stage complex ship design, called the Design Knowledge Management Square (DKMS) framework. The DKMS framework provides a structure that explicitly incorporates the collaborative nature of complex ship design, contrary to other models or frameworks that primarily focus on the technical integration of tools and methods to describe early stage complex ship design. The DKMS framework is applied to three case studies: 1) multi-disciplinary early stage design of complex ships, 2) the integration of concept design generation and analysis methods, and 3) the application of design rationale to support collaborative design decision-making. The case studies show that the DKMS framework provides added value by explicitly describing both the collaborative and technical nature of complex ship design. Thereby the framework helps to analyse, support, and understand complex ship design.

This paper describes two new modular ship design activities for graduate education at Delft University of Technology that have been developed during COVID. First, a new 2-hour hybrid format (in-person and virtual participation) game was designed to teach students modular design for offshore support vessels (OSVs). Second, an 8-week MSc-level ship design project was redeveloped to cover the design of a small fleet of modular OSVs for offshore wind. The paper discusses the drivers behind these new design educational activities, the details of the activities themselves, and concludes with lessons learned focused on improving graduate education for masters students studying ship design.

This paper proposes a parametric modelling method based on Knowledge Based Engineering (KBE) for an LNG bunkering vessel (LNGBV). Parametric models aim to define the geometry and main systems configuration of the vessel starting from the principle that the vessel should be able to perform its mission successfully. The adoption of KBE in combination with parametric modelling is expected to improve the current practice by automating the generation of the parametric models based on the design requirements. The results showed that different design alternatives can be rapidly generated which in turn gives the possibility to the designer to perform a wide exploration of the preliminary design space.