Title: Design tool selecting aid: a neuronal approach

Authors: Khaled Benfriha; Améziane Aoussat; Marc Le Coq

Addresses: Laboratoire CPI, Ecole Nationale Supérieure d'Arts et Métiers, 151 Boulevard de l'Hôpital, 75013 Paris, France ' Laboratoire CPI, Ecole Nationale Supérieure d'Arts et Métiers, 151 Boulevard de l'Hôpital, 75013 Paris, France ' Laboratoire CPI, Ecole Nationale Supérieure d'Arts et Métiers, 151 Boulevard de l'Hôpital, 75013 Paris, France

Abstract: The objective of this paper is to prove that it is possible to integrate knowledge engineering in the development of computational design process models. The main contribution of our research is to propose a parametric conceptual model for the design process, adapting neural networks to the design context, and finally an algorithm to aid design tool choice. This algorithm is modular and allows the integration of other functionalities. We were particularly interested in the problem of design tool selection aid and we developed an algorithm that integrates a learning modelling approach (multilayer neural networks). The application stemming from this algorithm gives a computer tool which offers the designers, according to their design problems, a set of adequate design tools. This computer-based design tool selection aid gives very encouraging results.

Keywords: design process; computer-based selection; CAD; optimisation; design tools; design tool selection; modularisation; neural networks; learning modelling; knowledge engineering.

DOI: 10.1504/IJPD.2015.069324

International Journal of Product Development, 2015 Vol.20 No.3, pp.197 - 220

Published online: 12 May 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article