An official website of the United States government.
The RFID wireless sensor technology proposed in this project will enable product to be continuously located, tracked and monitored in terms of temperature and other environmental factors. Traceability of these products throughout the supply chain can be used to identify products at risk for food safety and security. Application of this technology will enable real-time tracking and notification to managers via cell-phones and live networks in the event that products need to be flagged due to temperature abuse or other at-risk activities. Identification of events and product locations most susceptible to temperature abuse during food storage and transport will provide control measures for leafy greens and other fresh-cut produce. Building on previous work by our research team evaluating product placement and environmental factors affecting spoilage in small regional areas using sensor technology, this proposal will integrate sensor technology into food transportation and cold chain distribution on a broader scale, thus providing a strategy to identify and improve key areas during storage and distribution that are found to be at risk. Using the RFID product history data, several mathematical models will then be developed to predict growth and/or survival of E. coli O157:H7 on leafy greens during commercial distribution. The anticipated outcome of the extension program is an increase in the number of individuals who are better educated in food safety, transport and handling procedures, and who are in compliance with regulations regarding human exposure to food-borne pathogens during product distribution. Providing a well-trained workforce will result in a safer and more reliable food supply for the consumer, decrease waste, increase consumer confidence in the food industry, and reduce liability exposure for producers, food transporters, distributors and retailers. These efforts will result in greater economic returns throughout the food supply chain by reducing food pathogen growth and spoilage. The extension component will be subjected to continuous evaluation. The proposed project should yield the following information that will be important in enhancing the safety of fresh-cut leafy greens: Temperature and location historical data for fresh-cut leafy greens during transport in summer, fall, winter and spring. A quantitative model that can be used to predict the growth/survival of E. coli O157:H7 during transport and retail distribution based on time/temperature events. Recommendations to the manufacturing, transportation and retail industries concerning best handling practices during transport, dock transfer and retail display to insure compliance throughout the supply chain for minimizing bacterial growth during distribution of fresh-cut products. And most importantly, a predictive model for E. coli O157:H7 growth/survival in leafy greens based on actual time/temperature histories during transportation and retail distribution that can be built into the current risk assessment to better define the risk associated with the consumption of ready-to-eat bagged salad products.
NON-TECHNICAL SUMMARY: Pre-harvest sources of microbial contamination of are numerous and extremely varied, and there are currently no microbial reduction strategies to completely ensure the safety of fresh-cut leafy greens at the time of consumption. It is therefore imperative that precut greens be maintained at their proper storage temperature from the moment they are packaged through retail purchase, to minimize exposure to Escherichia coli O157:H7 and other food-borne pathogens. In the work being proposed here, active wireless sensors will be used to monitor, in real time, the temperature fluctuations that occur during the commercial transportation of fresh-cut leafy greens. Various distribution systems will be employed in the study. The data will be used to develop and validate a mathematical model which will be capable of predicting changes in E. coli O157:H7 growth and/or survival during commercial distribution of these products. The proposed research will therefore identify and assess problems that occur during the transportation of fresh produce between processing facilities and retailers. This information is critical for establishing the best handling practices for precut produce and for determining scientifically-based "best-used-by" dates to reduce the risk of food-borne illness. It will also be used as a basis to provide training for packaging and distribution professionals who wish to prevent food-borne illness through the monitoring of "at risk" product during transport and distribution. Ultimately, this research will result in precut vegetables that not only have extended shelf-life but which are safer for consumption.
<P>
APPROACH: Temperature fluctuations will be determined using RFID sensors. These sensors will be placed in three key zones within each refrigerated truck with the trailer cargo zone being used as a blocking variable with respect to the experiment and the predictive model. Sensor resources will be used with typical refrigerated trailers so as to better pinpoint truck-to-truck variability with respect to the treatments. RFID Sensor Wireless data will be gathered during transport lasting an anticipated 1 to 5 days. Temperature/relative humidity/time data obtained from the five most different transport scenarios will then be uploaded to a Blue M Environmental Chamber. Inoculated and uninoculated bagged samples will then be placed in this environmental chamber and evaluated at Michigan State University for growth of E. coli O157:H7 after cycling through simulated time and temperature events. A Baranyi primary model, with a square-root secondary model for maximum growth rate, and a linear model for MPD will be applied to the varying temperature conditions, and the dynamic solution will be based on the coupled differential equations. The model will be used to predict E. coli O157:H7 growth/survival for every temperature profile collected from the trucks in the field. Those initial predictions will be used to select the temperature profiles to be used in the previously described laboratory experiments. The validity of the existing growth model will be quantified in terms of the root mean squared error (RMSE) between predicted and observed experimental growth data. The RMSE will be evaluated for each individual temperature treatment, the aggregate data for each product, and for the entire aggregated experimental data set. These results will be used to compute the prediction interval for the prediction of future outcomes by the model. Robustness will be evaluated in terms of the ratio of the RMSEprediction to the RMSE reported from the original data set used to estimate the model parameters (i.e., RMSE). This index will be used to quantify the relative validity of the model for each of the products. If differences in model performance among the various products are not statistically significant (á = 0.05), then a single model will be applied. If significant differences do exist, then 2/3 of the experimental data (randomly selected) will be used to estimate product-specific model parameters, based on optimization routines to minimize the sum of squared errors between the observed and predicted counts (across all data for a specific product). The remaining 1/3 of the data will be used to quantify the RMSE. Once the best model is determined and validated, E. coli O157:H7 growth/survival will be predicted for all 405 temperature curves collected from the truck study. Each growth prediction will also include a quantitative estimate of uncertainty (e.g., the 95% prediction interval), which will account for the experimental variability and uncertainty resulting from the model form and regression results (Martino, 2006). The resulting set of 405 growth predictions will be quantified in terms of their statistical distribution and probabilities of measurable growth.