Résumé
The kinetics and duration of winemaking fermentations may vary considerably as a function of must composition and thermal conditions. This makes it difficult to manage tanks and power use in wine production. Modern decision support tools may help, but are difficult to use, as the underlying optimization problem is complex and its solution is time-consuming. We developed a decision support software based on a physiological model of yeast activity, including a mathematical model of heat transfer. The integrated decision support module uses fuzzy logic and a heuristic approach to provide efficient, rapid solutions to facilitate winery management at an industrial scale in various winemaking conditions. The applications of this software are illustrated using various realistic cases. (c) 2008 Elsevier Ltd. All rights reserved