Application of fuzzy reasoning to prediction of beef sirloin quality using time temperature integrators (TTIs)

Abstract

Time temperature integrators (TTIs) are indicators of food quality changes based on time–temperature history. The key issue in TTI studies is related to the accuracy of estimating food quality changes using mathematical models based on the color of the TTIs attached to food packages. In the original method, the temperature dependency of the reactions is analyzed in terms of Arrhenius relationships. However, when modeling the Arrhenius relationships, there are some problems regarding insufficient data points or non-linearity, thereby leading to significant errors in the predictions. In this study, therefore, a novel method for modeling temperature dependency was developed based on fuzzy reasoning, which is a versatile tool used to account for experimental data with non-linearity. The volatile basic nitrogen (VBN) of beef as well as the color value of the TTIs attached to the beef package were measured during storage. The individual kinetic equations for each temperature under isothermal conditions were incorporated using fuzzy reasoning rather than an Arrhenius equation in order to predict the VBN for a TTI color value under conditions of unknown time–temperature history. It was found that the new method was more accurate in predicting beef quality from TTI color than the original method.