Please use this identifier to cite or link to this item: http://localhost:8081/xmlui/handle/123456789/6290
Title: EXPERT SYSTEM FOR THE DIAGNOSIS AND CURE OF DEFECTS IN BUILDINGS
Authors: Agarwal, Sanjay Kumar
Keywords: CIVIL ENGINEERING;EXPERT SYSTEM;|BUILDING DEFECT;DIAGNOSIS
Issue Date: 1995
Abstract: An expert system is AI based computer software that encapsulates specialist knowledge about a particular domain of expertise and is capable of making intelligent decision within that domain. Expert systems help to conserve expertise and are used to make expertise more widely, easily and quickly available for assistance in the decision-making process and research. A number of problems arise when an expert system is built for practical use. Management of uncertainty is one problem which plays an important role in the design of expert systems. This dissertation attempts to develop a knowledge-based expert system for the diagnosis and cure of commonly occuring defects in buildings, named as BUILDEX, on an experimental basis Most buildings, though designed adequately for the applied loads, develop a variety of defects over the years. Diagnosis of commonly occuring defects in building and suggesting an appropriate cure demand a high level of expertise. BUILDEX attempts to capture this expertise. The system domain include two most commonly occuring defects in buildings dampness and cracking. The system knowledge has been represented in the form of rules and it has been implemented in Turbo Prolog. BUILDEX addresses the problem of uncertainty management using fuzzy set theory. Fuzzy set theory has been used for demonstrational purpose to represent uncertainty in the rule base of BUILDEX and method of approximate reasoning has been employed to draw inference.
URI: http://hdl.handle.net/123456789/6290
Other Identifiers: M.Tech
Research Supervisor/ Guide: Godbole, P. N.
metadata.dc.type: M.Tech Dessertation
Appears in Collections:MASTERS' DISSERTATIONS (Civil Engg)

Files in This Item:
File Description SizeFormat 
246756CE.pdf1.48 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.