A multi-agent quantitative microbial risk assessment of raw milk cheeses: from farm-level practices to consumer risk
Abstract
Ensuring the microbial safety of raw milk cheeses demands a robust, quantitative understanding of contamination pathways and risk mitigation strategies. In this study, we developed an innovative agent-based Quantitative Microbial Risk Assessment (QMRA) model that explicitly represents farms, animals, cheese batches, and consumers as individual agents. This approach enables a realistic simulation of microbial transmission throughout the entire production chain and facilitates the assessment of the effectiveness of control measures. Grounded in extensive knowledge of the French raw milk cheese sector, the model simultaneously addresses three major pathogens of concern: Shiga toxin-producing Escherichia coli (STEC), Salmonella, and Listeria monocytogenes. By incorporating stochastic Monte Carlo simulations, the model captures variability in microbial shedding at the farm level, contamination dynamics during milk collection and cheese processing, as well as exposure scenarios at the consumer level. This multiagent framework provides a structured approach that enhances the interpretation of risk scenarios and facilitates risk communication, particularly for risk managers aiming to implement evidence-based food safety policies or consumer guidelines. By explicitly modeling each component of the system, this methodology bridges the gap between theoretical risk assessment and practical decision-making. The approach presented in this article serves as a flexible and transparent tool for assessing microbial risks associated with STEC, Salmonella, and L. monocytogenes in bovine raw milk cheeses, and can be adapted to other foodborne hazards and production systems.
Related
- Distinct mechanical signatures of dairy and plant-based cheeses
- Effects of Processing Practices (Boiling and Drying) on the Nutritional, Microbial, and Physical Quality Attributes of Small Fish (Dagaa) in Small-Scale Fishery Systems.
- Microbial community evolution and flavor formation during sour meat fermentation: Integrated quantitative microbiome profiling, HS-SPME-GC–MS, and UPLC-Q-TOF-MS/MS
- Assessment of Microbial Consortia for Biogenic Mineralization Potential of Lead (Pb) using Microbiologically Induced Calcite Precipitation and Nitrate Reduction
- Quantitative MRI-based assessment of the ultrastructural compactness of myelin lamellae
BibTeX
@article{guillier2026multiagent,
title = {{A multi-agent quantitative microbial risk assessment of raw milk cheeses: from farm-level practices to consumer risk}},
author = {Laurent Guillier and Estelle Chaix and Frédéric Auvray and Lucie Collineau and Noémie Desriac and Narjes Mtimet and Eric Oswald and Florence Dubois-Brissonnet and Jeanne-Marie Membré},
journal = {Journal of Food Protection},
year = {2026},
month = oct,
doi = {10.1016/j.jfp.2026.100913},
url = {https://anses.hal.science/anses-05750172v1},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - JOUR
TI - A multi-agent quantitative microbial risk assessment of raw milk cheeses: from farm-level practices to consumer risk
AU - Laurent Guillier
AU - Estelle Chaix
AU - Frédéric Auvray
AU - Lucie Collineau
AU - Noémie Desriac
AU - Narjes Mtimet
AU - Eric Oswald
AU - Florence Dubois-Brissonnet
AU - Jeanne-Marie Membré
PY - 2026
DA - 2026/10/02
JO - Journal of Food Protection
DO - 10.1016/j.jfp.2026.100913
UR - https://anses.hal.science/anses-05750172v1
AB - Ensuring the microbial safety of raw milk cheeses demands a robust, quantitative understanding of contamination pathways and risk mitigation strategies. In this study, we developed an innovative agent-based Quantitative Microbial Risk Assessment (QMRA) model that explicitly represents farms, animals, cheese batches, and consumers as individual agents. This approach enables a realistic simulation of microbial transmission throughout the entire production chain and facilitates the assessment of the effectiveness of control measures. Grounded in extensive knowledge of the French raw milk cheese sector, the model simultaneously addresses three major pathogens of concern: Shiga toxin-producing Escherichia coli (STEC), Salmonella, and Listeria monocytogenes. By incorporating stochastic Monte Carlo simulations, the model captures variability in microbial shedding at the farm level, contamination dynamics during milk collection and cheese processing, as well as exposure scenarios at the consumer level. This multiagent framework provides a structured approach that enhances the interpretation of risk scenarios and facilitates risk communication, particularly for risk managers aiming to implement evidence-based food safety policies or consumer guidelines. By explicitly modeling each component of the system, this methodology bridges the gap between theoretical risk assessment and practical decision-making. The approach presented in this article serves as a flexible and transparent tool for assessing microbial risks associated with STEC, Salmonella, and L. monocytogenes in bovine raw milk cheeses, and can be adapted to other foodborne hazards and production systems.
SN - 0362-028X
ER -