Computational & Technology Resources
an online resource for computational,
engineering & technology publications
Civil-Comp Proceedings
ISSN 1759-3433
CCP: 17
KNOWLEDGE BASED SYSTEMS FOR CIVIL & STRUCTURAL ENGINEERING
Edited by: B.H.V. Topping
Paper XI.3

Application of Classical AI Planning to Structural Design

D.T. Ndumu, D. Lloyd Smith and B.A. Izzuddin

Department of Civil Engineering, Imperial College of Science, Technology & Medicine, London, England

Full Bibliographic Reference for this paper
D.T. Ndumu, D. Lloyd Smith, B.A. Izzuddin, "Application of Classical AI Planning to Structural Design", in B.H.V. Topping, (Editor), "Knowledge Based Systems for Civil & Structural Engineering", Civil-Comp Press, Edinburgh, UK, pp 265-276, 1993. doi:10.4203/ccp.17.11.3
Abstract
In this paper, we describe the elements of a methodology for design founded on classical artificial intelligence (AI) planning. Essentially, we view design as composed of twin recursive and mutually constraining subprocesses of formulation and synthesis-analysis-evaluation. Formulation establishes the goals and constraints of a design problem, while synthesis-analysis-evaluation attempts to achieve these goals by utilising available means within the bounds of the imposed constraints. By focusing on this view of the design process that emphasises its goal-, constraint- and means-centred nature, we have developed a methodology for design based on planning. Formulation is performed by plan recognition, and synthesis-analysis-evaluation is performed by plan generation, execution and monitoring, coupled with inference from a causal representation of the application domain. We provide an example of the functioning of our methodology through a prototype system for the design of beams to satisfy lateral torsional buckling requirements.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £60 +P&P)