Документ взят из кэша поисковой машины. Адрес
оригинального документа
: http://www.stsci.edu/~rdouglas/publications/thesis/section2_6_5.html
Дата изменения: Tue May 13 19:35:11 1997 Дата индексирования: Sat Dec 22 19:13:45 2007 Кодировка: Поисковые слова: п п п п п п п п п п п п п п п п п п п п п п п п п п п п |
~ ~ There were two forms of knowledge acquisition that had to be done to develop SNEAKERS. The first concerned concurrent engineering. The second concerned the choice of a domain for the project and knowledge about design within that domain.
Knowledge about concurrent engineering was gathered though several sources.
A survey of the literature, summarized in Chapter ,
provided a basic understanding of the topic. Professors Bausch and Zenger
of the WPI Manufacturing Engineering department also sparked ideas
and guided the search for information. After this, a meeting with
Paul Posco, A. J. Overton, John Hamer, and David Meeker
of the Competitive Product Development Institute at DEC offered some
other suggestions.
One point of that meeting was to determine how close our analysis of the literature was to what they were saying about concurrent engineering. They confirmed the points gathered through the literature. The other point to the meeting was to attempt to determine the important features that should be included in SNEAKERS.
The choice for the domain is detailed in Chapter . Professor
Zenger helped to determine the domain, as his expertise was needed
to choose a domain that would meet the requirements of being intuitive and
multi-disciplinary.
Once the domain was selected, Professor Zenger helped provide knowledge for
the agents. He provided sample rules and acted as an expert in tower building.
Knowledge about how to design a tower, and when to invoke other agents,
was gained by carefully stepping through the process of designing several
towers. This knowledge was captured in part through the storyboarding
described in Section .