A PROLOG BASED EXPERT SYSTEM FOR CRACK DIAGNOSIS AND REPAIR SUGGESTIONS IN BUILDINGS

Abstract

Throughout the history , men have always attempted to find techniques and methods to bring fellow humans under their control. Rhetoric was one of the first attempts to reach this goal by a formaulized process called Artificial Intelligence; AI is basically a software technique to emulate the human reasoning process. The two major trends in AI applications arc Generalized Problems Solvers COPS) and Knowledge Based Expert Systems (KBES). Since the former did not offer sufficient power of high performance, more emphasis has been laid on developing Knowledge Cased Export Systems. The solutions of currently generated problems in CIVIL ENGINEERING analysis and design arc based on idealized modelling and simulation. Knowledge Cased Expert System is nothing but the art of simulation of human reasoning process in solving a problem which often consists of many years of experience and intution. This dissertation addresses the problem of design, development and implementation of a production type of expG.rt system for crack diagnosis and repair suggestions in a building which has been developed on an experimental basis with knowledge domain of real visual cracking symptoms in a single storeyed building constructed with conventional building materials. This system is designed using TURDO PROLOG, implemented through about 100 rules, to run on personnel computers and contains consultation module for actual consultation with an in-built user interface and also help module to note down the correct symptoms for entering in to the system to make it function by displaying causes and repair suggestions.