Filip Poschet

Monte Carlo analysis as a tool to incorporate variation on experimental data in predictive microbiology

Until now, most of the mathematical models used in predictive microbiology are deterministic, i.e. their outcome is a point estimate for the microbial load at a certain time instant. For more advanced exploitation of predictive microbiology in the context of hazard analysis and critical control points and risk analysis studies, stochastic models should be developed. Such models predict a probability mass function for the microbial load at a certain time instant. The objective of this paper is to illustrate methodologically how to generate, starting from the experimental observations and a deterministic growth model, probability density functions for (i) the model parameters and (ii) the predictions as a function of time, by using Monte Carlo analysis. A normal distribution over the experimental data was considered. This probabilistic approach, incorporating experimental variation, is applied to experimental growth data of Escherichia coli K12 and Listeria innocua ATCC 33090.

A Novel Class of Predictive Microbial Grown Models: Implementation in an Individual-Based Framework

Abstract In the field of predictive microbiology, mathematical models are developed to describe and predict the behaviour and possible outgrowth of spoilage and/or pathogenic microorganisms in food products. Research has mostly focused on the development of macroscopic models, which have a number of inherent disadvantages. This paper adopts the methodology of individual-based modelling (IbM) as a complement to macroscopic models to overcome some of these issues. In addition, this paper exploits a new bacterial growth model to circumvent shortcomings of established logistic type models. The new model incorporates substrate limitation and metabolite inhibition factors, providing it with a more solid mechanistic base for modelling the stationary phase. A case study is presented implementing the model in an IbM framework and exploratory results are presented