Authors: Monique Axelos, Jean-Dominique Daudin, Guy DellaValle, Nathalie Perrot, Catherine MGC Renard, Caroline Sautot, Jean-Louis Sebedio
This article presents an EU project, DREAM, started two years ago, that develops realistic, physical and mathematical food models to be used as standards that can be exploited across major food categories. We underline the need for physical models realistic enough to cope with the wide diversity of food products and we detail their selection criteria, keeping in mind that these models should be versatile enough to favour the development of common approaches to risk assessment and nutritional quality for food research and industry. We also summarize the different sources of knowledge (scientific from various disciplines, expert engineers ?) that have to be completed, the essential properties to be determined at every structural scale and the scientific models of different phenomena to be integrated. An example of reverse-engineering approach is also proposed to optimise technical pathways in food processing, as aimed to be promoted by DREAM project.