Comparison of Three Parametric Models for Pathogenic Population Reduction Prediction

Abstract

The purpose of this study is to study and to compare the performance of three models:       1) Weibull model 2) Exponential model and 3) Log logistics model. The performance of each model is evaluated and compared to determine the most appropriate model for predicting the 5-log reduction time of pathogen population in fermented fruit juice products. The criteria are used to compare the performance of each model that contains Akaike’s Information Criterion (AIC) and Sum Square Error (SSE). From the results of this study, it indicated that the Weibull model is the best predictor of decline in the pathogen population. As the pathogenic population of the orange juice product dropped to 5 - log reduction in the range of 13 to 14 days, which is in accordance with the Weibull model's 13.567 prediction value. Moreover, with the sum of the square error, the value of the is 0.103 and the AIC of -91.422 are lower than those of the Exponential and log-logistic models.