Improving The Utilization Of Microbial Pathogen Computer Models For Validating Thermal Processes In The Meat Industry

This research provided a complete secondary model for log-D with temperature, fat content and moisture content as independent variables. It also expanded thermal inactivation parameters for Salmonella lethality models. The results led to improved user interface and functionality for the AMI Process Lethality Spreadsheet.

Download final report

Objectives

The overall goal of this project was to improve the tools used by the industry to ensure that thermal processes are meeting the USDA-FSIS lethality performance standards for ready-to-eat products.  The specific objective was to improve the AMI Process Lethality Spreadsheet (AMI-PLS) by adding a user-friendly “front-end” that accounts for the effects of key product factors (e.g., species, fat content) on the thermal inactivation parameters for Salmonella. 

Conclusions

According to the data analysis and modeling, the D-value for thermal inactivation of Salmonella in meat and poultry products could be successfully modeled in a general way, as a function of temperature, fat content, and moisture content. However, further research is needed to validate the model with independent results.  Additionally, the confidence intervals (CI) for the D-value (and resulting lethality calculations) are still quite wide, because of the relatively large RMSE. Therefore, more data sets and/or advanced statistical simulations are needed to narrow the CI, to improve model accuracy, and to validate the model with independent data before the enhanced version of the AMI-PLS will be ready for distribution.

Deliverable

Models developed were integrated directly into the AMI-PLS, and a user-friendly “input box” was added, so that the user inputs the product characteristics, and the AMI-PLS calculates and utilizes product-specific D-values in the lethality calculations.  The modified AMI-PLS also generates confidence intervals (±>95%) for direct prediction of lethality calculations (i.e., log reductions).