Location: Quality and Safety Assessment Research Unit
Project Number: 6040-42440-001-021-A
Project Type: Cooperative Agreement
Start Date: Aug 19, 2025
End Date: Aug 18, 2028
Objective:
1. Examine alternatives to primary poultry processing approaches to reduce pathogen cross-contamination in scalding and picking in order to improve food safety while enhancing process sustainability and product quality.
2. Investigate advanced chilling strategies and develop antimicrobial sensing technologies and control strategies aimed at reducing foodborne pathogens and enhancing food safety efficacy of poultry chilling.
3. Develop novel Artificial Intelligence (AI) based approaches for the detection and separation of foodborne contaminants (pathogens/foreign materials) in poultry processing.
Approach:
This project rethinks poultry processing by incorporating innovative technologies, improved procedures, and data-driven methods to enhance food safety and operational efficiency. It explores alternative primary poultry processing methods, including scalding, de-feathering, chilling, and further processing, to drive improved food safety outcomes. This includes using sensing technologies for managing antimicrobials, such as peracetic acid (PAA), rapid microbial detection in chilling operations, and development of innovative chilling methods. These strategies aim to improve food safety, while also reducing water and energy use through smart sensing and data-driven control of antimicrobial interventions. The project also advances AI-based prediction, detection, and separation of foodborne contaminants and foreign materials during poultry processing.
Because traditional scalding and de feathering can promote pathogen cross contamination, the project investigates novel scald pick methods that use natural rigor processes to enable waterless scalding, eliminating a major contamination pathway. Researchers will compare microbial loads before and after scald/pick between conventional immersion scalding and these new methods.
To improve antimicrobial performance in chillers, the team will deploy rugged sensors capable of measuring PAA, other antimicrobials, and microbial concentrations in real time. These sensors will include filtration, PAA detection, and automated dosing mechanisms to adjust antimicrobial levels based on bacterial load. Once implemented, this system will allow centralized monitoring of antimicrobial levels at multiple points along the processing line. The closed loop control framework will use interferometric sensors and data analytics to automatically manage dosing, optimize antimicrobial efficacy, reduce chemical use, and improve food safety.
To address inconsistent carcass exposure to antimicrobials during chilling, the team will design and evaluate in line immersive chilling systems. These systems use translation and rotation methods to improve antimicrobial contact and thermal consistency. Because carcasses remain on shackles, cross contamination is reduced relative to bulk transport chillers. In line chilling also enhances traceability and lowers energy and water consumption.
The project will also develop automated methods for detecting and removing products with contamination/foreign material using novel manipulation and AI algorithms. Existing USDA hyperspectral datasets will be used to fine tune foundation models capable of generalizing across product streams. The project will evaluate whether lower cost optical or multispectral systems can match hyperspectral detection performance. Further research will explore usage of robotics for human like inspection, enabling full product examination. Once deployed, models will classify contaminants in real time, supporting automated removal and improving traceability and reliability of foreign material detection.