KEEPING-QUALITY ASSESSMENT OF PASTEURIZED MILK BY MULTIVARIATE-ANALYSIS OF DYNAMIC HEADSPACE GAS-CHROMATOGRAPHIC DATA .1. SHELF-LIFE PREDICTION BY PRINCIPAL COMPONENT REGRESSION

An objective and rapid analytical system for milk shelf-life prediction was established. Volatiles as determined by dynamic headspace capillary gas chromatography (DH-GC), psychrotrophic bacterial counts, and sensory evaluation were monitored during refrigerated storage of pasteurized milk. High correlation between flavor-related shelf life and volatiles indicated that a good proportion of the total variation was explained by the mathematical model obtained by principal component regression (PCR). The resulting standard error of the estimate of less than 2 days was an excellent approximation in shelf-life studies. Using DH-GC, results are available in 20 h (18 h of preliminary incubation + 2 h of detection and data processing). Although other rapid instrumental techniques have been proposed, they are indirectly related to flavor-related shelf life. Milk quality deterioration and consequently the termination of shelf life are caused by the appearance of off-flavors mainly determined by a composite effect of spoilage volatiles. Therefore, multivariate interpretation of headspace gas chromatographic data and flavor-related shelf life as derived by PCR offers a direct and accurate approach to milk shelf-life prediction.