A Fuzzy Expert System for Optimal Material Selection
Kamel Mehdi11 Dridi Sofien, Ali Zghal
Mechanical Laboratory of Solids, Structures and Technological Development of the ESSTT
Superior School of Sciences and Techniques of Tunis, 5 avenue Taha Hassen B.P. 56 Bab Mnara
1008 Tunis – Tunisia.
The material selection process operates in two distinct areas of engineering application. Firstly, it is an indispensable part of any new product development project and has an important role in the review and re-evaluation of existing products and their performance with regard to cost and reliability. Secondly, it has to ensure compliance with design specifications and proven service requirements. An improperly chosen material can lead not only to the failure of the part in service, but also to unnecessary cost. Historically, materials were selected from handbooks with limited choice and on the basis of limited property data. Computer based material selection software are now used, since the list of possible materials and processes has become too large.
This paper presents a fuzzy expert system for optimising materials selection. The system is based, firstly on the definition of the different material classes with object oriented language and secondly on a knowledge base in form of IF-THEN rules for multiple decisions according to the desired objective. The expert system utilized forward chain for its reasoning mechanism and fuzzy data sets and crisp sets for material characteristics representation. The expert system is developed according the object-oriented environment of Kappa-PC.
KEYWORDS: Artificial Intelligence, Fuzzy expert system, Object oriented system, Design Methodology, Material selection.
do so for
2000-2010 Professor F.R. Hall & Dr I. Oraifige, University of