Kraiem, Z M (1988) DISCON: An expert system for construction contract disputes. Unpublished PhD thesis, University of Colorado at Boulder, USA.
Abstract
Artificial Intelligence (AI) techniques that enable computers to help analyze difficult problems and make decisions are proving their merit in various commercial applications. Expert systems combine expert human knowledge with AI methods for the programming of reasoning. This research developed an expert system application which aids in the analysis of a construction contract dispute. In construction contracts, disputes and claims stem from a wide variety of origins. Construction problems present not only engineering problems but legal problems as well. DISCON is an artificially intelligent expert system designed to give legal advice to design and construction engineers as well as contractors and owners concerning Differing Site Conditions claims. DISCON assesses the contractor's chance of being awarded the extra expenses he claims are due to the occurrence of conditions differing materially from what was expected or indicated by information available at the time the contract was bid and awarded. This research demonstrates that computer-based legal consultants are possible aids for legal analysis. Concepts of acquiring DISCON's knowledge from domain experts and the knowledge representation aspects of a construction contract claim analysis are presented. DISCON has been developed within the M.1 environment. The ability of a knowledge base to explain its reasoning and to inform the user of the contents of a knowledge base is central to effective use of the expert system. A C language program has also been developed to enhance the explanation features of the system. The interaction of the M.1 shell and the C code offered DISCON the advantages of adding no overhead to the expert system program no matter how long the explanation, as well as allowing the user to control the depth and degree of an explanation. In analyzing a construction contract claim, uncertainties may arise from different sources, including inaccuracy of findings, inconsistency in judgment, difficulty in proving facts, and incompleteness of data. DISCON checks all uncertainties and reacts according to their source as well as to their effects on the analysis. At the end of an analysis, DISCON shows the contractor's chance for recovery weighted by the certainty factor that is estimated throughout the analysis. (Abstract shortened with permission of author.)
Item Type: | Thesis (Doctoral) |
---|---|
Thesis advisor: | Diekmann, J E |
Uncontrolled Keywords: | reasoning; artificial intelligence; disputes; legal analysis; programming; expert system; consultant; owner |
Date Deposited: | 15 Apr 2025 07:36 |
Last Modified: | 15 Apr 2025 07:36 |