Prototype of an Expert System for Aluminum Welding V. TSOUKALAS a * , M. KONTESIS b , E. BADOGIANNIS c , D. PAPACHRISTOS b , N. FRAGIADAKIS a a Engineering Department Merchant Marine Academy of Athens Paralia Aspropyrgou 19300, Aspropyrgos GREECE b Computer Science Department Technological Educational Institute of Piraeus P. Ralli &Thivon 250, 12244 Aigaleo GREECE c Marine Enginnering Department National Technical University of Athens Iroon Polytechniou 9, 15773 Zografou GREECE * Corresponding author vdtsoukalas@yahoo.com Abstract: - Increasing demand for high quality aluminum welding and the international competition has been leading the ship industry to adopt new advanced IT tools. Prototype knowledge based expert system named WELDES (WELDing Expert System) has been developed under the frame of the Archimedes project 259(03). The system is able to identify the aluminum welding defects, to correlate them with the welding parameters, which cause them, and to offer advice regarding the necessary corrective actions. The first step in the construction of this program was the systematic classification of the aluminum welding defects and the analysis of the reasons, which cause them. The WELDES system is aiming to be used as an industrial tool for the identification, assessment, and correction of aluminum welding defects. The system consists of two modules: the Diagnostic Module, and the Adviser Module. The structure of the program is ideally suited to the knowledge domain of the welding and the ship industry as it makes productive the previous knowledge and uses the acquiring experience of many years in the research and production of high quality welding. The next step is the expert system to be tested in real industrial environment. Key-Words: - Artificial Intelligence, Knowledge Engineering, Expert Systems, Welding Process, Aluminium 1 Introduction The increased requirements for high quality aluminum welding in combination with method complexity as well as the conditions of international competition have been leading the manufacturing industry towards the adoption of promoted calculating tools [1-4]. Thus, even if the world of industry is conservative and circumspect towards new technologies, the application of Artificial Intelligence (AI) in the field of manufacturing and naval industry and its important role to the quality improvement of produced products, contributed to its widest and most rapid spreading [4-7]. As a result, a great number of research institutions, universities, and industrial units started an important research effort that aimed towards the growth and application of AI in various applications. The reasons that led to the adoption of AI in the scientific field of mechanical manufactures are mainly the following [2]: • The lack of a complete descriptive, qualitative mathematic model of many mechanical Proceedings of the 5th WSEAS Int. Conf. on COMPUTATIONAL INTELLIGENCE, MAN-MACHINE SYSTEMS AND CYBERNETICS, Venice, Italy, November 20-22, 2006 78