Graduate Thesis Or Dissertation
 

Theory and application of vector space similarity measures in computer assisted conceptual design

Public Deposited

Downloadable Content

Download PDF
https://ir.library.oregonstate.edu/concern/graduate_thesis_or_dissertations/1v53k137g

Descriptions

Attribute NameValues
Creator
Abstract
  • A number of computational tools now exist to aid in developing conceptual solutions based on a functional description of a design problem. A key limitation of these tools is the way results are organized for presentation to the user. In general, results are an undifferentiated mass of potential solutions. Analysis using a novel concept clustering tool shows concept generator output represents permutations of a set of a few solution archetypes. This provides an initial solution to organizing and presenting the results. More efficient solutions are sought by adopting a generate-evaluate-guide framework from the computational design synthesis literature. Specifically, the concept generation approach is altered so that each generated solution maximizes the variety it adds to the set of solutions. To achieve this, suitable similarity measures must first be developed. Current techniques for similarity assessment in the design literature tend to be ad hoc and highly specialized to particular tasks. Prior work from the field of information retrieval is applied and extended to create a generalized approach to similarity assessment for vector space design data. These techniques are validated against an existing design by analogy methodology. A new tool for locating functional analogies within a database of existing products is developed as a result. Improved similarity measures are combined with the proposed computational synthesis framework from literature to modify an existing concept generation tool. The resulting tool efficiently locates the few novel solutions in the set of possible results, and is a key step in the continued evolution of this class of computational design tools.
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
Rights Statement
Publisher
Peer Reviewed
Language
Replaces

Relationships

Parents:

This work has no parents.

In Collection:

Items

Thumbnail Title Date Uploaded Visibility Actions
file details Poppa_dissertation_final.pdf 2017-08-19 Public Download