Agent‐based simulation of cross‐contamination of Escherichia coli O157:H7 On lettuce during processing with temperature fluctuations during storage in a produce facility. Part 1: Model development

Abstract

AbstractA food safety agent‐based model (FS‐ABS) with discrete time event simulation using NetLogo© was developed to predict the levels of cross‐contamination in a lettuce processing facility. Visualization of the whole processing line shows different lettuce pieces around each equipment's surface. The focus was set on Romaine lettuce processing and on the impact of cross‐contamination, free‐chlorine concentration on wash‐water, and cold storage room temperature fluctuation strategies on E. coli O157:H7 growth. Microorganism decay, growth, and transfer models in the wash water, equipment, and lettuces were embedded in a discrete event simulation model to deal with the growing complexity and uncertainties occurring in processing plants. The model was developed based on literature data. Lettuces were considered agents that move through the processing line moving from patch‐to‐patch agents that have their own attributes, including levels of contamination and probability of microbial transfers. Space was represented by a 2‐D grid‐based equipment with multiple attributes. Space provides the environmental basis for movements and interactions of lettuces. The storage facility was simulated as a continuous agent‐based model, which runs parallel with the main program (processing facility). This article describes the development of the model along with assumptions and semi‐empirical models used to describe cross‐contamination during processing and pathogen growth during storage. Part 2 discusses the model verification, calibration, validation, sensitivity analysis, and results from different scenarios.Practical ApplicationsWe simulated the cross‐contamination of a common pathogen, E. coli O157:H7, in a fresh‐cut lettuce processing facility using the agent‐based model (ABM) approach. This approach captures the complex interactions between factors and emergent results based on agents' (lettuce and equipment) interactions within the model that other types of models cannot. These interactions are essential in understanding the dynamics of cross‐contamination in the post‐harvest processing chain of fresh‐cut leafy‐greens. Most of the factors that affect cross‐contamination in fresh‐cut lettuce (equipment surfaces, lettuces, wash water) were considered in this study. Wash water quality was monitored in terms of the residual free chlorine (FC) levels and the organic matter accumulation measured as chemical oxygen demand (COD). Pathogen growth due to temperature abuses during storage (post processing) were integrated into the model. Available literature data from a pilot size fresh‐cut processing facility were used to create the model.