Integration of evaluation and simulation methods for virtual prototypes

This article combines evaluation and selection of early design concepts into a methodology at different stages of the development process. A web-based evaluation environment to support this methodology has been developed including Kurtoglu's component taxonomy. In addition, the Taguchi's design of experiments approach has been used to virtually assess all feasible combinations generated from Zwicky matrix. This evaluation methodology is applied to the case study of an autonomous robot for Eurobot competition. The virtual environment was useful to provide a fast visual representation of the different robot concepts and their behaviours. This evaluation methodology was compared with AHP and Pareto optimality evaluation methods. The main contribution of this paper is to develop a methodology and a computer tool integrating virtual simulation associated with a design of experiments providing a visual representation of the concepts. These concepts can be simulated according to situations they would face in reality. This enables the evaluation of concepts according to non-subjective criteria and it could assist the decision making process at early design stage. Furthermore, this evaluation method enables preliminary selection in a concurrent manner during the phase of concept generation. The goal is to integrate the entire approach as a web-based platform supporting the early design process. This work is developed for pedagogical purpose in the context of an EU project.