Int. J. Computer Applications in Technology, Vol. 30, Nos. 1/2, 2007 113 Copyright © 2007 Inderscience Enterprises Ltd. A substance-field ontology to support the TRIZ thinking approach Alexis Bultey*, François de Bertrand de Beuvron and François Rousselot INSA Strasbourg (Graduated School of Science and Technology), LGECO (Design Engineering Laboratory), 24 bvd de la victoire, 67084 Strasbourg, France E-mail: alexis.bultey@insa-strasbourg.fr E-mail: francois.debeuvron@insa-strasbourg.fr E-mail: francois.rousselot@insa-strasbourg.fr *Corresponding author Abstract: An ideal TRIZ reasoning environment should support TRIZ fundamental concepts and simulate its thinking process. In this paper, an advanced TRIZ methodology is analysed in this perspective: the substance-field analysis. Previously, it has been shown that the TRIZ knowledge can be modelled and managed in an object-oriented ontology for computer aided problem formulation. A new ontological model based on this previous work is proposed in order to simulate the TRIZ problem solving stage: from the generation of a general solution to the interpretation phase linking the abstract field of general solution to the real field of physic. Keywords: TRIZ; substance-field analysis; SFA; ontology; description logics knowledge representation system; DL-KRS. Reference to this paper should be made as follows: Bultey, A., de Bertrand de Beuvron, F. and Rousselot, F. (2007) ‘A substance-field ontology to support the TRIZ thinking approach’, Int. J. Computer Applications in Technology, Vol. 30, Nos. 1/2, pp.113–124. Biographical notes: Alexis Bultey is currently a PhD student of F. Rousselot. He is a graduate Electrical Engineer of INSA, Strasbourg, and he followed the Advance Master in Innovative Design of INSA, Strasbourg. His current research interest is interdisciplinary, including both engineering design and knowledge representation. François de Bertrand de Beuvron is an Associate Professor of Computer Science at INSA, Strasbourg, France. His current research focuses on knowledge representation system, particularly the DL-KRS. François Rousselot is a Professor of Computer Science at Marc Bloch University of Strasbourg, France. His current research deals with knowledge extraction, representation and management. 1 Introduction Structuring the inventive thinking is the core approach of TRIZ; it is based on the studies of 200,000 patents and it has outlined typical general problems and their general solutions. Altshuller has developed TRIZ since 1949 in the former USSR (Orloff, 2003). Theoretically speaking, even if TRIZ seems to be a promising opportunity in technological problem-solving, it is difficult to implement it in the designer practice. There are many reasons for this situation. First, the TRIZ comprehension suffers from the occidental translation, 95% of about all the TRIZ physical literature is written in Russian and most of the TRIZ fundamental concepts have suffered from translators’ interpretation because of their own design knowledge (Cavallucci et al., 2002). Second, between the 1970s and 1980s, the TRIZ theoretical materials have increased to give the designer a complete tool set but it led to complicate the practical application of TRIZ methods (Mann, 2002). Third, a final factor is the need of Artificial Intelligence (AI) for supporting the TRIZ activity. In fact, designers use themselves as intellectual interpreters; they translate their own abstract solutions according to their own knowledge of technological and physical sciences. It implies that every designer must know or have access to an amount of knowledge, which increases day by day with the physical and technological discovery. To overcome this dilemma, the AI assistance seems to be the key. A similar idea has been followed in the TRIZ development with the ARIZ appearance, the algorithm of inventive problem-solving. ARIZ expresses the idea that the TRIZ process can be generalised by some heuristics, linking each step of the solving process to the TRIZ tools or the TRIZ principles (Altshuller, 1999).