The prediction of shelf-life is a useful tool for the development of new products and packaging materials. The prediction of product shelf-life can be obtained by mathematical modelling of the changes in selected quality characteristics of packaged food during storage. In order to investigate the quality changes during storage, orange juice with five different initial oxygen concentrations (1 to 10 ppm) was filled in 1 L Tetra Brik Aseptic cartons, and stored at seven different temperatures (50, 40, 30, 20, 8, 4 and -18°C). Samples of orange juice were collected during time course and were analysed for ascorbic acid content, browning index, off-flavour production (p-vinyl-guaiacol). Mathematical models including the effect of the storage temperature and the initial concentration of dissolved oxygen in the juice were developed for the quality characteristics analysed. The models showed to be adequate to describe the changes in orange juice quality during storage.