Computational & Technology Resources
an online resource for computational,
engineering & technology publications |
|
Civil-Comp Proceedings
ISSN 1759-3433 CCP: 13
ARTIFICIAL INTELLIGENCE AND STRUCTURAL ENGINEERING Edited by: B.H.V. Topping
Paper IV.1
Reasoning About Constructibility: Representing Construction Knowledge and Project Data M. Fischer
Center for Integrated Facility Engineering, Stanford University, Stanford, USA M. Fischer, "Reasoning About Constructibility: Representing Construction Knowledge and Project Data", in B.H.V. Topping, (Editor), "Artificial Intelligence and Structural Engineering", Civil-Comp Press, Edinburgh, UK, pp 105-112, 1991. doi:10.4203/ccp.13.4.1
Abstract
Due to the fragmentation in the construction industry, structural engineers rarely use explicit
constructibility knowledge when making decisions about the layout and dimensioning of structural
elements. An on-line construction expert offers the opportunity to allow use of efficient
construction methods on a project by interfacing with a CAD representation of the design and by
providing decision support in the form of construction knowledge. I developed such an expert
system called COKE (Construction Knowledge Expert) by acquiring constructibility knowledge
from field experts, formalizing and organizing this knowledge base in a way suitable for input to
the design process, specifying a symbolic model of a reinforced concrete structure that supports
constructibility reasoning, and implementing a software prototype that links a CAD system to an
expert system. When providing constructibility feedback, COKE first uses high level application
heuristics to determine whether a construction method is applicable or not. It then compares the
project data in the symbolic model with specific constraints about the layout and dimensioning of
structural elements, and alerts the designer about mismatches between these constraints and the
structure. This research shows that software technology can be used to integrate downstream
project constraints into the work environment of design engineers. Object-oriented programming
offers a computing paradigm to support this cross-functional integration. Once implemented, such
integrated intelligent tools offer the opportunity to solve problems in early project stages at a
fraction of the cost of making changes in the construction or operation phases of a facility.
purchase the full-text of this paper (price £20)
go to the previous paper |
|