Graduate Thesis Or Dissertation
 

Understanding Designer Mental Models to Support Computer Directed Analogical Design

Public Deposited

Contenu téléchargeable

Télécharger le fichier PDF
https://ir.library.oregonstate.edu/concern/graduate_thesis_or_dissertations/gq67jv14s

Descriptions

Attribute NameValues
Creator
Abstract
  • Analysis of alternative concepts has a significant impact on design project outcomes, and yet many design teams fail to consider a significantly broad range of conceptual solutions. Within the realm of conceptual design exists a technique called design by analogy (DBA) -- the practice of reapplying old solutions to new problems. DBA mitigates the effort required to generate a large field of candidate concepts by leveraging existing knowledge from a wide variety of domains, making it an attractive approach toward improving design outcomes. Unfortunately, DBA is challenging in the absence of expert knowledge. Designers need computational support in order to effectively identify a large number of high-quality analogical connections across a wide variety of domains. With this challenge in mind, the goal of this dissertation is to improve the body of knowledge regarding computational support for design by analogy. More specifically, this body of work includes five manuscripts. Manuscript 0 presents a review of several function-related design abstractions, including their impacts on education and industry. Manuscript 1 studies analogy retrieval in a novel design context and catalogs the types of abstract similarity (including function) commonly used to form analogies. Manuscript 2 examines a scalable approach to capturing analogy-relevant design knowledge to support large-scale analogy searching. Manuscripts 3 and 4 examine and modify a technique from de novo drug design for quickly indexing and retrieving design analogies. Manuscript 3 examines the domain independence of the technique, and manuscript 4 develops it as a large-scale design analogy search method. The body of work contributes to a greater understanding of (1) the abstractions used by designers during conceptual design, (2) the use of human computation to support conceptual design activities, and (3) large scale solution screening using a variety of mixed design abstractions. This understanding advances the creation of tools that enable designers to consider a wide range of conceptual solutions in spite of lacking domain expertise.
License
Resource Type
Date Available
Date Issued
Degree Level
Degree Name
Degree Field
Degree Grantor
Commencement Year
Advisor
Committee Member
Academic Affiliation
Non-Academic Affiliation
Subject
Déclaration de droits
Publisher
Peer Reviewed
Language
Replaces

Des relations

Parents:

This work has no parents.

Dans Collection:

Articles

La vignette Titre Date de téléchargement Visibilité actes
file details ArlittRyanM2015.pdf 2017-08-14 Public Télécharger