preliminary approach to predictive modelling of extra virgin olive oil stability

The aim of this work was to set up a phenomenological model to predict the stability of extra virgin olive oil based on combined stability/instability indices. A screening of stability/instability indices was carried out by multivariate statistical analyses on company data for virgin olive oil. Screened indices were acidity, oleic acid content and bitter taste. The ability of these indices to predict stability was checked by measuring the following degradation parameters on 11 oil samples of different compositions planned by a fractional factorial design (FFD): peroxide value, UV spectroscopic indices, minor polar component content, oxidative status of fatty acids, antioxidant activity and sensory descriptors of aroma, taste and flavour. Experimental data were processed by multivariate statistical analyses. A combination of acidity, oleic acid content and bitter taste values was significantly able to predict oil stability expressed by the peroxide value, the K232 UV index and the oxidative status of fatty acids.