The research objective of this award is to improve understanding and capabilities in concept generation through design by analogy methods. The proposed approach, a collaboration between the disciplines of cognitive psychology, engineering design and computer science, is to provide new tools for design based on a representation that associates functional and geometric information, combining a linguistic search for functional similarity with a multi-level search for geometric similarity to automatically identify and present analogies to the designer. The initial application for the Verrocchio Project is the design of prosthetic and orthotic devices for persons with disabilities, a domain that is ripe for innovation. The initial search space is the USPTO utility patent repository. Deliverables will include: (1) means to more effectively generalize design problems through functional descriptions; (2) the ability to search for analogical solutions with alternative functional representations; (3) ways to search for geometric similarities across a set of functional analogies; and (4) the ability to produce a tractable set of analogies for use by the designer.
If successful, the results of this research will have five broad and transformative impacts. The first impact arises from the key product of the research: a method for systematically and automatically identifying analogies. This tool will have broad applicability across many product domains, and will improve the efficiency of the search for innovative products in those domains. Second, the research will advance the foundation for use of analogies in engineering design methods. Third, because the research will result in a computer implementation of the method, the work will add to the growing cyber infrastructure of the country. Fourth, the interdisciplinary nature of the research ensures cross-fertilization of theory and research techniques in each of the collaborators? disciplines. This interdisciplinarity will enhance the probability of broad adoption of the approach. Fifth, instructional materials will be developed for training students in the approach. These instructional materials will be freely available to instructors at other institutions, encouraging broader adoption of the approach.
A methodology was developed that allows for the discovery of deep structural relationships among design repository data, enabling the basis for automatic finding of inspirational stimuli for design by analogy practice. These structures were validated with further cognitive studies to test their use in finding analogical stimuli, and comparison with expert representations of the space, described in the next three sections. The method uses different stages of textural similarity matching and Bayesian structure creation. The structures created with this methodology yield spaces of patents that are meaningfully arranged into labeled clusters, and labeled regions, based on their functional similarity or surface content similarity. Examples showed that cross-domain associations, and perhaps transfer of knowledge, based on functional similarity can be extracted from the function based structures, and even from the surface content based structures as well. The comparison of different structural form types is shown to yield different insights into the arrangement of the space, the interrelationships between the patents, and the information within the patents that is attended to – enabling multiple representations of the same space to be easily accessible for design inspiration purposes. In addition, the placement of a design problem in the space effectively points to the most relevant cluster of patents in the space as an effective starting point of stimulation. The work illustrated that in general near patents can stimulate more similar concepts while more distant ones can stimulate more novel concepts. However there is such a thing as too near or too far where too near may be too similar and too far may be irrelevant to the design conceptualization task. Instead a "sweet spot" is proposed that balances the search for novel solutions with the practicality of the design task. Cognitive studies verified the use of the tool to inspire design, and also the relevance for experts as well as novices.