skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Explore Research Products in the PAR
It may take a few hours for recently added research products to appear in PAR search results.


Title: Data-driven sustainable ship design using Axiomatic Design and Bayesian Network Model
Abstract Environmental sustainability, as well as social and economic well-being, must be considered in every stage of a product lifecycle, from conceptual design to its retirement. Even though this sustainability-centric approach represents a critical driver for innovation, it also increases the design complexity. Nowadays, the maritime transport accounts for a large share of transport demand, and the importance of sustainable ship design is increasingly growing, not only for ethical and legislative but also for competitive reasons. The design of a sustainable ship considering all those aspects is a complex process in this regard. One way to manage the complexity is to identify and avoid the functional couplings at the early stage of the design process. This paper presents the conceptual design of a merchant ship's conventional propulsion system with a view to the Axiomatic Design framework and known sustainable engineering principles. We also explore the Bayesian machine learning interface to propose a data-driven method for calculating the probability of achieving specific sustainability-related functional requirements. Data-driven Bayesian reasoning can also be used to select the best design parameter among the proposed alternatives as well as to identify hidden design couplings that have not identified by the designers in the conceptual design stage.  more » « less
Award ID(s):
1854833
PAR ID:
10346637
Author(s) / Creator(s):
;
Date Published:
Journal Name:
IOP Conference Series: Materials Science and Engineering
Volume:
1174
Issue:
1
ISSN:
1757-8981
Page Range / eLocation ID:
012003
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; (, ASME)
    Na (Ed.)
    Engineering design is a continuous and iterative process, where early-stage decisions significantly impact subsequent design outcomes. This study investigates the influence of AIassistance during early stages of design on subsequent design stages and measures the change in both design outcomes and cognitive processing in the brain. Sixty undergraduate engineering students participated in a two-stage design task. Students were first asked to identify design constraints related to the sustainable redevelopment of a site on campus either using human imagination or utilizing generative AI to assist them. Students, in both groups, without the aid of generative AI, then developed conceptual design ideas for redevelopment. The results indicate that the AI-assisted group identified significantly more design constraints (p < 0.05) and subsequently without the aid of AI developed a greater number of design concepts related to environmental sustainability. Brain imaging analysis revealed that AI assistance reduced the neuro-cognitive effort during constraints identification and had a residual effect in reducing neuro-cognitive effort during the concept design phase, particularly in the right frontopolar prefrontal cortex – a region associated with complex, abstract thinking. These findings suggest that AI-assisted design can enhance design efficiency by optimizing reducing cognitive effort and improving early-stage design outcomes. Future research should explore human-AI collaboration strategies to maximize its benefits in engineering design workflows. 
    more » « less
  2. ; ; (, American Society of Mechanical Engineers)
    Abstract Engineering design is a continuous and iterative process, where early-stage decisions significantly impact subsequent design outcomes. This study investigates the influence of AI-assistance during early stages of design on subsequent design stages and measures the change in both design outcomes and cognitive processing in the brain. Sixty undergraduate engineering students participated in a two-stage design task. Students were first asked to identify design constraints related to the sustainable redevelopment of a site on campus either using human imagination or utilizing generative AI to assist them. Students, in both groups, without the aid of generative AI, then developed conceptual design ideas for redevelopment. The results indicate that the AI-assisted group identified significantly more design constraints (p < 0.05) and subsequently without the aid of AI developed a greater number of design concepts related to environmental sustainability. Brain imaging analysis revealed that AI assistance reduced the neuro-cognitive effort during constraints identification and had a residual effect in reducing neuro-cognitive effort during the concept design phase, particularly in the right frontopolar prefrontal cortex – a region associated with complex, abstract thinking. These findings suggest that AI-assisted design can enhance design efficiency by optimizing reducing cognitive effort and improving early-stage design outcomes. Future research should explore human-AI collaboration strategies to maximize its benefits in engineering design workflows. 
    more » « less
  3. ; ; ; ; (, Advanced Materials Technologies)
    Abstract Morphing matter that change shapes and properties in response to external stimuli have gained significant interests in material science, robotics, biomedical engineering, wearables, architecture, and design. Along with functional advances, there is growing pressure and interest in considering the environmental impact of morphing matter during its life cycle. The unique manufacturing and usage of morphing matter means that existing sustainable design frameworks and principles for general physical products may not apply directly. For example, manufacturing morphing matter often requires designing and predicting materials' behaviors over time, and using devices fabricated with morphing matter often involves harnessing renewable energy and self‐reconfiguration, which pose unique sustainability opportunities and challenges. This study reflects and summarizes the field's practice in sustainable manufacturing, transport, use, and end‐of‐life handling of morphing matter. The term “sustainable morphing matter” (SMM) is coined, suggesting that sustainability‐conscious factors can become an integral component of morphing matter. In addition, ways to apply sustainability‐conscious factors to augment the existing design pipeline of morphing matter are presented, and more quantitative and algorithmic‐level developments are needed to apply these factors rigorously to the design process. 
    more » « less
  4. ; ; ; ; (, Scientific Reports)
    Abstract Data-driven design shows the promise of accelerating materials discovery but is challenging due to the prohibitive cost of searching the vast design space of chemistry, structure, and synthesis methods. Bayesian optimization (BO) employs uncertainty-aware machine learning models to select promising designs to evaluate, hence reducing the cost. However, BO with mixed numerical and categorical variables, which is of particular interest in materials design, has not been well studied. In this work, we survey frequentist and Bayesian approaches to uncertainty quantification of machine learning with mixed variables. We then conduct a systematic comparative study of their performances in BO using a popular representative model from each group, the random forest-based Lolo model (frequentist) and the latent variable Gaussian process model (Bayesian). We examine the efficacy of the two models in the optimization of mathematical functions, as well as properties of structural and functional materials, where we observe performance differences as related to problem dimensionality and complexity. By investigating the machine learning models’ predictive and uncertainty estimation capabilities, we provide interpretations of the observed performance differences. Our results provide practical guidance on choosing between frequentist and Bayesian uncertainty-aware machine learning models for mixed-variable BO in materials design. 
    more » « less
  5. ; (, Conference on Systems Engineering Research)
    Despite the efforts to increase the pace of sustainable design adaptation in industries, several systemic barriers currently hinder this shift. The design for sustainability methods have been utilized in product design and development phases in many industries. However, they do not have a holistic approach that can capture these systemic drivers and barriers while considering all three pillars of sustainability: environmental, social, and economic. This research proposes a systems thinking approach toward sustainable design that can collectively consider different aspects of the production system in an attempt to resolve the multi-dimensional challenges within the design for sustainability. A reusable water bottle is selected as the case study to illustrate the applications and limitations of this approach. In addition, this case study also helped to define the boundaries and stakeholders involved in the system and reduce the abstractions. The results from this analysis are demonstrated as a causal loop diagram that could be implemented in a system dynamics model to quantitively identify the systematic forces and leverage points driving sustainable design in product development. The comprehensive understanding provided by this analysis revealed many improvement possibilities, trade-offs, and feedback loops within the system that can assist in realizing sustainable product design proliferation and associated positive sustainability outcomes. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email