Agent‐based simulation of cross‐contamination of Escherichia coli O157:H7 on lettuce during processing and temperature fluctuations during storage in a produce facility. Part 2: Model implementation

Abstract

AbstractAn agent‐based simulation was developed to represent the spatial and temporal Escherichia coli O157:H7 cross‐contamination dynamics in a processing facility for fresh‐cut romaine lettuces using NetLogo. An extension was added to the model to predict E. coli O157:H7 growth due to temperature abuses in a cold storage facility. A user‐friendly interface was created to follow variations in model outputs. The number of contaminated bags and lettuce contamination levels were computed and visualized. Experimental data of cross‐contamination from literature was used to describe the facility and validate the model. The key factor affecting cross‐contamination in the process is the free chlorine (FC) concentration dose rate. The number of contaminated bags per batch decreases nonlinearly as FC is added to the flume tank. The contamination level in the shredded lettuces increases linearly with initial microbial load or the initial contamination probability on lettuce heads. Both the level and probability of contamination in the facility's environment alter the number of contaminated bags. Batch size influences the number of contaminated bags when the first incoming lettuce batch is contaminated. The smaller the batch size the smaller number of contaminated bags. Storage room temperature fluctuations (13 and 19°C and 8 and 12°C) show the importance of real‐time monitoring to avoid pathogen growth and number of contaminated bags. This work provides insight on applications of agent‐based modeling approach using NetLogo to visualize cross‐contamination in fresh‐cut leafy green processing operations. It analyzes cross‐contamination and its impact on processing performance by studying the effect of mitigation strategies.Practical applicationsTo support ready‐to‐eat leafy greens postharvest operations, this work investigated the benefits of using visualization to develop strategies to prevent produce cross‐contamination along processing and the effect of temperature abuse during storage. Output from the model can be used in Quantitative Management Risk Assessment (QMRA) applications. By developing a simulation‐based decision support system, the safety of processing and storage steps of fresh‐cut leafy greens are investigated. Pathogen cross‐contamination and growth models are integrated to observe impacts of varying contamination levels in incoming produce and equipment surfaces, concentrations of free chlorine in the wash water, and storage temperatures on the number of contamination bags over time. Computational experiments study a pilot plant of Romaine lettuces contaminated with E. coli O157:H7. Results indicate that the key factor affecting cross‐contamination is the chlorination concentration dose rate in the wash water. Other factors include the level of contamination on the incoming lettuce heads or in the facility environment (equipment). Storage room temperature fluctuations showed the importance of real‐time monitoring to avoid microorganism growth and thus prevent an increase in the number of contaminated bags.