A TRIZ-based CAI Framework to guide Engineering Students towards a Broad-spectrum Investigation of Inventive Technical Problems

Nowadays, a growing consensus is attributed to conceptual design in the perspective of developing effective and successful products; as a consequence, major efforts should be dedicated within the Computer-Aided Innovation field to correctly support this task. A particular line of evolution of these systems concern computer coaches, i.e. software applications capable to assist users along each step of design activities. In such perspective the authors have developed a dialogue-based system supporting a natural language questioning procedure to investigate technical problems through TRIZ way of thinking.Anemerging deficiency of its first version concerns limited capabilities in providing a broad screening of the issues and features relevant within the encountered inventive problems. The integration of a further analysis module supporting the logic of the TRIZ System Operator has allowed the individuation of effective resources for breaking technical conflicts inherent to the investigated systems or formalizing contradictions in TRIZ terms. The paper provides further support about the need of developing broad thinking skills in engineering education. The new generation of Computer-Aided systems is extending its scope, by supporting the user already from the early design stages, where creativity and cognitive processes are of paramount importance. In such context, the authors have developed a natural language dialogue-based framework capable to coach designers in investigating non-routine technical problems through a TRIZ way of thinking. The first version of the questioning procedure has showed limitations in terms of the capability to widen the designer's perspective, resulting in the overlooking of aspects relevant for the problem. The new version of the dialogue-based system integrates a further analysis module, here presented, supporting the logic of the TRIZ System Operator. New tests demonstrate that the modification allows to improve the quality of the analyses, especially in terms of the identification of features to be advantageously redesigned in order to solve the addressed problems. © 2013 TEMPUS Publications.