A new analytical and statistical approach to predict the sensory properties of deep frying fats and oils to determine the point of discard during processing

Abstract

The aim of this work was to explore the feasibility and limits of analytical criteria to assess the sensory properties of used frying oils and different fried food. Over a period of 2 years, 148 used frying oils collected in restaurants bakeries, fish caterers, and industrial plants were analyzed for a wide range of parameters, including sensory tests, total polar compounds (TPC), dimerized and polymerized triacyl‐glycerols (DPTG), monomeric oxidized triacyl‐glycerols (MONOX), acid value (AV), para‐anisidine‐value (AnV), and iodine value (IV). All methods were judged for their relevance, concerning the sensory quality of used frying oils categorizing the qualities as good or bad. The binary results were statistically evaluated using logit regression. The statistical evaluation of the tests was used to calculate the probability of occurrence of off‐flavors which allows determining the point of discard and finally the point where the oil is certainly totally abused. All levels of degradation for used frying oils regardless of oil type and kind of food being fried can be statistically described and used to develop a simple equation using the TPC and AV value as variables (y = 117–8 × AV − 3 × TPC) to find the point of disposal and to monitor the degradation process. The new criterion has been proven in the controlling of various industrial frying processes, restaurants and bakeries and kitchen frying.Practical applications: A simple equation of linear regression using AV and TPC as variables (y = 117–8 × AV − 3 × TPC) has been developed to monitor the level of degradation and to determine the point of disposal of used frying fats and oils much more reliable than using only one of the recommended parameters (TPC, AV, or PTG). This criterion is independent of fat composition, food to be fried, or frying process conditions and effectively helps to overcome the limitations of TPC and other major regulatory parameters related to the frying/baking industry. The good correlation of the results of the equation with the sensory quality of the fried food provides a powerful predictive tool for accurate calculation of the oil disposal point. This is useful since the quality of the fried food strongly depends on the quality of the frying oil that is absorbed by the food and becomes a major ingredient in the final product. This tool can easily be used to improve frying oil quality which finally improves the quality of the food to be fried. With continuous improvement this tool can become useful in reducing time and costs required for a human sensory panel.Physical and chemical changes of oils during different frying processes affect the development of color, flavor, and taste of the oil and also the structure of the food being fried. A new analytical approach applicable for all types of frying oil from different sources (bakeries, fisheries, restaurants, and industrial fryers) is proposed to predict the sensory quality of the oil using the content of total polar compounds and acid value resulting in the probability of the point of discard (DEGLEV).