Collaborative SGER: Computational Support for Semantically Rich Design Problems - Addressing High Risk Elements

点击翻译按钮获取中文摘要

The objective of this SGER proposal is to explore a conceptual basis and empirical justification for representing engineering semantic information in a computational framework. This is high risk, yet high reward, in that modeling semantic information has been elusive whereas facilitating reasoning on both geometric and semantic information would radically change how engineers design. This research consists of three tasks: conducting case studies on industry use of semantic information, survey of existing approaches to representing this information, and analysis of computational support for these representations. Three case studies, including a small manufacturing firm, a large automotive OEM, and senior capstone design projects, will be conducted to determine how and when engineers use, manipulate, and document the semantic information. The case studies will be conducted using standard historical and active protocol analysis techniques and will serve as the guide for surveying and evaluating appropriate semantic modeling techniques.If successful this understanding of the role of semantic information in engineering design will provide justification for developing theory of affordances, borrowed from perceptual psychology, into a computational environment as supported by exemplar technology. Furthermore, if a suitable structure for capturing semantic information were available and it could be mapped to a computational representation this would greatly impact mechanical engineering design. As CAD tools have revolutionized design, semantically rich tools have the potential to radically impact practice, providing an environment where multiple facets of a project can be considered, resulting in reduced time-to-market, fewer design iterations, fewer costly mistakes, and improved quality.