Probabilistic modeling of Listeria monocytogenes behaviour in diced bacon along the manufacture process chain

To assess the impact of the manufacturing process on the fate of L. monocytogenes, we built a probabilistic model describing successively the different steps of the process. The model was actually designed as a hierarchical Bayesian network leading to the elicitation of human expertise. Contamination evolution was modelled in the adequate units (breasts, dices, then packaging units through the successive process steps). The use of probabilistic modeling allowed taking into account both the process intrinsic variability and parameter variability or uncertainty. Global statistics were deduced, diagrams showing the variability were drawn, and changes on the process were tested to look at the consequences on the final product.