Probabilistic Modeling of the Fate of Listeria Monocytogenes in Diced Bacon During the Manufacturing Process

Abstract

To assess the impact of the manufacturing process on the fate of Listeria monocytogenes, we built a generic probabilistic model intended to simulate the successive steps in the process. Contamination evolution was modeled in the appropriate units (breasts, dice, and then packaging units through the successive steps in the process). To calibrate the model, parameter values were estimated from industrial data, from the literature, and based on expert opinion. By means of simulations, the model was explored using a baseline calibration and alternative scenarios, in order to assess the impact of changes in the process and of accidental events. The results are reported as contamination distributions and as the probability that the product will be acceptable with regards to the European regulatory safety criterion. Our results are consistent with data provided by industrial partners and highlight that tumbling is a key step for the distribution of the contamination at the end of the process. Process chain models could provide an important added value for risk assessment models that basically consider only the outputs of the process in their risk mitigation strategies. Moreover, a model calibrated to correspond to a specific plant could be used to optimize surveillance.

Hierarchical Bayesian Models to Assess Between‐ and Within‐Batch Variability of Pathogen Contamination in Food

“…In addition, these variances characterize dispersion between batches, between units, and between test portions, which makes them more difficult to elicit. In the pork breast study, it was possible to build informative priors for each parameter because a computer model was available ( 34 ) to simulate L. monocytogenes concentration after tumbling. For the CSS case study, it was not possible to find informative priors for the variance parameters.…”

“…It is possible to get some information about within‐ or between‐batch variability by considering the heterogeneity of raw materials, the process, and the environment. For instance, in the pork breast study, ( 34 ) pork breasts from several farms arrive at the plant and are grouped together into one batch so one may think that the within‐batch variability is high. The first step of the process is tumbling: pork breasts exchange the microorganisms on their surfaces during several hours so after tumbling the within‐batch variability is small.…”

“…Another way is to model the origin of the contamination, such as raw material contamination or cross‐contamination. ( 34, 39 ) The proposed approach directly uses colony counts and integrates details about the sampling and analysis protocols in the modeling. This protocol‐based description is then easily transposable to any specific way of collecting data.…”

“…This process-based model produces the empirical distributions for contamination of L. monocytogenes after every step of the process, which is described in the article (Section 2.1). (34) Using this model, simulated data were kept (after the tumbling step) for 1,000 batches of 500 pork breasts so as to calibrate informative priors for the general model. Associated compatible priors were then calculated by minimizing the TD criterion (Equation (6)) for models B-batch, B-unit, and B-test portion.…”

“…Prior distributions of parameters for the B‐reference model (the general one) were determined using simulation data obtained from the model of the process according to Billoir et al ( 34 ) The data used in the study (34) are different from the ones used in this article. This process‐based model produces the empirical distributions for contamination of L. monocytogenes after every step of the process, which is described in the article ().…”