The Power of Modeling and Data Sharing By Gustavo Reyes, Food Safety Manager
In the evolving landscape of agriculture, ensuring the food safety of produce is a top priority for growers. As the industry experiences technological advancements, the integration of modeling and data- sharing can be a vital and powerful tool to assess and mitigate risk in growing, packing, processing and shipping produce. A collaborative approach must be adopted to create a sense of community and foster community-driven responsibility as food safety challenges are tackled.
Simulations and predictive models are processes that attempt to replicate behaviors of real-world scenarios or use historical data to forecast outcomes. These models have proven to be indispensable tools for nearly every single aspect of our day-to-day lives, from weather to sports and social media to data. In the field of agriculture and food safety, these models can be valuable tools. Simulation models of leafy green supply chains have highlighted the importance of controlling and monitoring wash water chemistry to prevent cross-contamination and controlling temperature during transportation and retail to prevent microbial growth 1, 2 . Agent-based simulation models have highlighted the contamination dynamics, risks and best corrective actions for processing facility equipment surfaces for Listeria monocytogenes 3, 4 . New Center for Produce Safety (CPS)-funded projects, such as one being conducted by the University of Illinois Urbana-Champaign and Cornell University, are looking to build a flexible model that the industry can use to assess best practices and guide food safety investments. 5 Data sharing is the concept of making data available for others to access, use, and in some cases, distribute. Data sharing involves the intentional and controlled use of data for analysis and collaborative research. Like simulations and predictive models, data-sharing has shown to be very beneficial. In healthcare, data-sharing initiatives have embraced data-sharing related to treatment and diagnosis to achieve better medical outcomes. In transportation, 1 Mokhtari, A., et al., Evaluation of Potential Impacts of Free Chlorine during Washing of FreshâCut Leafy Greens on Escherichia coli O157:H7 Cross-Contamination and Risk of Illness. Risk Analysis, 2021. 42 (5): p. 966-988. 2 Pang, H., et al., Quantitative Microbial Risk Assessment for Escherichia coli O157:H7 in Fresh-Cut Lettuce. Journal of Food Protection, 2017. 80 (2): p. 302-311. 3 Barnett-Neefs, C., et al., Using agent-based modeling to compare corrective actions for Listeria contamination in produce packinghouses. Plos one, 2022. 17 (3): p. e0265251. 4 Zoellner, C., et al., EnABLe: An agent-based model to understand Listeria dynamics in food processing facilities. Scientific reports, 2019. 9 (1): p. 495. 5 CPS Center for Produce Safety. Flexible risk process models to quantify residual risks and the impact of interventions . 2023 [cited 2024 Jan 25th ]; Available from: https://www. centerforproducesafety.org/researchproject/491/awards/Flexible_ risk_process_models_to_quantify_residual_risks_and_the_ impact_of_interventions.html.
companies such as Uber are sharing aggregated and anonymized data to better understand traffic patterns for more efficient urban planning. In the produce industry, modeling and data sharing can have a tremendous impact. Data-sharing platforms such as GreenLink® operate as central hubs, enabling growers to contribute their testing and observational data for further analysis. These analyses can help areas such as parametrization, the process of defining variables to describe how something works or behaves, for predictive models as uncertain parameters. This includes in-field contamination prevalence and levels that can be optimized by analyzing aggregated data. Decision-making and guidance documents can take a dynamic approach, where insights from time series analysis guide practices such as increased monitoring during riskier seasons. Association analysis, the task of finding interesting datasets, can be leveraged to prioritize practices based on risk. Finally, data sharing can allow the industry to anticipate food safety issues, to proactively create and provide timely resources to growers, packers, processors and shippers. These collaborative approaches are fundamental to getting us closer to tasking smarter data-based decisions. While data-sharing and modeling offer many benefits, challenges and risks, such as data privacy, security and data quality exist. To address these challenges, data-sharing platforms such as GreenLink® have instituted data-sharing policies and data-sharing agreements to address privacy concerns. Data is aggregated, anonymized and access- controlled to address security risks. Data standards, quality checks and data validation is conducted to control the quality of the data. In conclusion, collaborative simulation models paired with data-sharing platforms are a pivotal advancement to proactively advance food safety. As agriculture and data become more present in day-to-day operations, the path forward should focus on building a connected future where growers collaboratively shape produce safety and increase the amount of usable data for researchers to continue developing these modeling tools.
12
MARCH | APRIL 2024
Western Grower & Shipper | www.wga.com
Made with FlippingBook flipbook maker