Predictive Mathematical Modeling of Microbial Growths in Meat

Abstract

This thesis investigates mathematical modeling of the bacterial growth and decay on meat to predict quality, shelf life and safety of the food. Different literature related to mathematical modeling and microbial growth in food generally and on meat specifically were assessed. Regarding to modeling and its types, microorganism, bacterial growth in meat, its growth phases and counting methods and the most significant influence of meat spoilage factors in pathogenic growth were also set. Accordingly, this research concludes to use the primary, secondary and tertiary modeling methods in modeling the total bacterial growth of the four most dominantly meat spoilage bacteria, which are, L. monocytogenes, S. aureus, E. coli O157: H7 and Salmonella spp. Thus, the Modified Gompertz model (the most generic model) for the primary model, square root model for secondary, and FSSP and ComBase soft wares for tertiary models were used to model the growth. In the context of setting the factors that have significant influence in microbial growth with no preservatives in a food the study is bounded with four factors. These are temperature, PH, Time and moisture content. Finally, the task goes to result validation which is done by tertiary model and by using predictive microbiology in a risk-based approach (Af and Bf). Modeling errors were compared. Modeling error was decreased in all data-sets through using GA (7.22%, 62.81%, 1.65% and 3.86% for first, second, third and fourth data-sets respectively). Based on results, GA is an appropriate method for optimization of parameters in logistic with delay method.