Ram D. Sriram
Beyond Geometry: Issues in Product Representation,
The early part of this millennium has witnessed the emergence of an Internet-based engineering marketplace, where engineers, designers, and manufacturers from small and large companies are collaborating through the Internet to participate in various product development and marketing activities. This will be further enhanced by the next generation manufacturing environment, which will consist of a network of engineering applications, where state of the art multi-media tools and techniques will enhance closer collaboration between geographically distributed applications, virtual reality tools will allow visualization and simulation in a synthetic environment, and information exchange standards will facilitate seamless interoperation of heterogeneous applications. To support such an environment we need a product representation scheme that goes beyond the geometric-centric approach of traditional CAD systems. In this talk, I will describe two major efforts that I have been involved in over the last two decades in achieving such a goal.
The first effort was the MIT DICE project, which was one of the first efforts on developing a computer supported collaborative design environment. In this project, we developed a shared object product model which provided support for: multiple levels of functional abstractions; multiple levels of geometric abstractions; multiple functional views; and representation and management of multiple levels of constraints. This model used an integrated multi-level and multi-dimensional geometric representation scheme that is general enough to capture geometry through the entire design life cycle, i.e., from qualitative spatial relationships to detailed specifications of geometry at various levels of abstraction.
The second effort is the interoperability framework that is currently being developed at NIST. This framework focuses on the exchange of part and assembly information between heterogeneous modeling systems for collaborative design and manufacturing. The framework has the following key attributes: (1) it is based on formal semantics, and will eventually be supported by an appropriate ontology to permit automated reasoning; (2) it is generic: it deals with conceptual entities such as artifacts and features, and not specific artifacts such as motors, pumps or gears; (3) it is to serve as a repository of a rich variety of information about products, including aspects of product description that are not currently incorporated; (4) it is intended to foster the development of novel applications and processes that were not feasible in less information-rich environments; (5) it incorporates the explicit representation of design rationale, considered to be as important as that of the product description itself; and (6) there are provisions for converting and/or interfacing the generic representation schemes into a production-level interoperability framework.
Location: Visualization, and Engineering, First International Conference, Spain, Sept. 2004
We are grateful to members who have shared presentations with the Washington Academy of Sciences. If you have content of this kind that you would be willing to share, please contact us.