APPLICATION OF ARTIFICIAL NEURAL NETWORKS AS A NONLINEAR MODULAR MODELING TECHNIQUE TO DESCRIBE BACTERIAL-GROWTH IN CHILLED FOOD-PRODUCTS

In many chilled, prepared food products, the effects of temperature, p H and %NaCl on microbial activity interact and this should be taken in to account. A grey box model for prediction of microbial growth is dev eloped. The time dependence is modeled by a Gompertz model-based, non- linear differential equation. The influence of temperature, pH and %Na Cl reflected in the model parameters is described by using low-complex ity, black box artificial neural networks (ANN's). The use of this non -linear modeling technique makes it possible to describe more accurate ly interacting effects of environmental factors when compared with cla ssical predictive microbiology models. When experimental results on th e influence of other environmental factors become available, the ANN m odels can be extended simply by adding more neurons and/or layers. (C) 1998 Elsevier Science B.V. All rights reserved.