At the beginning of the COVID-19 pandemic, several contamination clusters were reported in food-processing plants in France and several countries worldwide. Therefore, a need arose to better understand viral transmission in such occupational environments from multiple perspectives: the protection of workers in hotspots of viral circulation; the prevention of supply disruption due to the closure of plants; and the prevention of cluster expansion due to exports of food products contaminated by the virus to other locations. This paper outlines a simulationbased approach (using agent-based models) to study the effects of measures taken to prevent the contamination of workers, surfaces, and food products. The model includes user-defined parameters to integrate characteristics relating to SARS-CoV-2 (variant of concern to be considered, symptom onset…), food-processing plants (dimensions, ventilation…), and other sociodemographic transmission factors based on laboratory experiments as well as industrial and epidemiological investigations. Simulations were performed for a typical meat-processing plant in different scenarios for llustration purposes. The results suggested that increasing the mask-wearing ratio led to great reductions in the probability of observing clusters of more than 25 infections. In the case of clusters, masks being worn by all workers limited the presence of contamination (defined as levels of at least 5 log10 viral RNA copies) on meat cuts at less than 0.05 % and maintained the production capacity of the plant at optimal levels. Increasing the average distance between two workers from less than 1 m to more than 2 m decreased the cluster-occurrence probability by up to 15 % as well as contamination of food products during cluster situations. The developed approach can open up several perspectives in terms of potential communication-support tools for the agri-food sector and further reuses or adaptations for other hazards and occupational environments.