Development and validation of heat and mass transfer models for meat carcass chilling

Abstract

Chilling of meat carcasses immediately after slaughter is important for microbial safety and quality of the meat. Rapid chilling may reduce potential bacterial growth, and may help control post-mortem biochemical changes affecting color and texture of the meat (e.g., pale, soft, exudative meat). However, excessive rapid chilling may lead to other quality issues (e.g., ‘cold shortening’), and reduced yield due to excessive moisture loss. Therefore, optimization of carcass chilling is important for producing high quality and safe meat, without compromising yield. Computer models to simulate heat and mass transfer during chilling of meat carcasses can be used to determine optimal operating conditions to maximize chilling rates and minimize moisture loss and potential growth of foodborne pathogens and spoilage microorganisms. The objective of this research was to develop and validate computer models for simulating air-chilling of meat carcasses; particularly chicken, pork, and beef carcasses. The models considered heat conduction and internal moisture diffusion subjected to convection, surface-to-ambient thermal radiation, and moisture evaporation. Three-dimensional geometries of the carcasses were generated from computer tomography images. Thus, the effect of non-uniform thermal properties corresponding to the meat, bone and fat sections was considered. The models were developed using a combination of computer aided engineering software (e.g., COMSOL Multiphysics^®, Materialise Mimics, and Matlab^®). The models provided accurate predictions using input parameters available for meat processors such as air relative humidity, air velocity, chiller set-point temperature, and carcass weight. RMSE for temperature predictions was 1.3 ± 0.6 °C, 1.48 ± 0.41°C, 1.6 ± 0.3°C for chicken, pork, and beef carcasses, respectively. In addition, moisture loss predictions resulted in RMSE values of 0.11%, 0.31%, and 0.24% for chicken, pork, and beef carcasses, respectively. The developed models were integrated with predictive microbial models of pathogens of interest including Salmonella spp. and Shiga toxin-producing Escherichia coli. The proposed models can be used not only for process optimization, but can help support food safety management systems in developing critical limits for hazard analysis, estimating potential impact of chilling deviations, and simulating multiple processing scenarios for quantitative microbial risk assessment.