Computational & Technology Resources
an online resource for computational,
engineering & technology publications |
|
Civil-Comp Proceedings
ISSN 1759-3433 CCP: 35
DEVELOPMENTS IN ARTIFICIAL INTELLIGENCE FOR CIVIL AND STRUCTURAL ENGINEERING Edited by: B.H.V. Topping
Paper III.1
The Knowledge-Use Level: A Hybrid Knowledge Engineering Methodology for Construction Site Layout Design M. Mateus*, I. Watson# and A. Grilo#
*Technical University of Lisbon, Lisbon, Portugal
M. Mateus, I. Watson, A. Grilo, "The Knowledge-Use Level: A Hybrid Knowledge Engineering Methodology for Construction Site Layout Design", in B.H.V. Topping, (Editor), "Developments in Artificial Intelligence for Civil and Structural Engineering", Civil-Comp Press, Edinburgh, UK, pp 25-34, 1995. doi:10.4203/ccp.35.3.1
Abstract
In the present research, construction site layout design was
analysed from the perspective of studying the knowledge,
information and activities involved. According to this, a
hybrid methodology for site layout knowledge acquisition,
modelling, representation and verification was developed.
The methodology is based on the "Componential Framework", Structured Analysis techniques, and Chandrasekaran's family of "propose - critique - modify" methods, from which some techniques are presented, such as decomposition, constraint satisfaction, plan-based, case-based, and model-based methods. The adoption of the "knowledge-use level" presented in this paper provides the means by which an integrated and deep knowledge-base system can be built, since it represents the deep knowledge and first principles of the site layout domain. The methodology provides a structured medium of expression and communication with site planners. As a result, the domain was modelled in an integrated way with construction planning, and the information was structured according to its nature and function. Further developments will flow towards the implementation level of the domain knowledge. This means that future research can concentrate on the identified subtasks, focusing on how the site planner executes them, using traditional knowledge representation formalisms such as rules, logic, frames, and objects, to achieve the implementation. purchase the full-text of this paper (price £20)
go to the previous paper |
|