Multivariate accelerated shelf-life testing: a novel approach for determining the shelf-life of foods

In order to meet consumers' expectations for high-quality products, food industries must conduct shelf-life studies that many times include the assessment of several analytical and sensory properties. However, whenever a new product is to be launched onto the market, defining which are the most relevant properties to monitor, as well as their cut-off criterion, is the subject of strong debate. Besides, for products with long estimated shelf-lives, accelerated studies have to be conducted and a third parameter has to be estimated: the acceleration factor which defines the correlation between the different storage conditions. In this study we propose a new approach for determining the shelf-life of industrialised food products, the Multivariate Accelerated Shelf-Life Test (MASLT), in which a Principal Component Analysis (PCA) is performed and the scores of the time-related components are taken for estimating the multivariate rate constants (k(m)), the multivariate acceleration factor (alpha(m)) and the multivariate activation energy (Ea(m)). The method was successfully applied to a single-concentrated industrialised tomato product for which the actual shelf-life was estimated to be 28 months. Pseudo-zero-order kinetics resulted in the multivariate parameters alpha(m)(35,25) = 2.7 and Ea(m) = 150 kJ center dot mol(-1). Copyright (C) 2006 John Wiley & Sons, Ltd.