Toward in-process technology-aided automation for enhanced microbial food safety and quality assurance in milk and beverages processing.

Ensuring the safety of food products is critical to food production and processing. In food processing and production, several standard guidelines are implemented to achieve acceptable food quality and safety. This notwithstanding, due to human limitations, processed foods are often contaminated either with microorganisms, microbial byproducts, or chemical agents, resulting in the compromise of product quality with far-reaching consequences including foodborne diseases, food intoxication, and food recall. Transitioning from manual food processing to automation-aided food processing (smart food processing) which is guided by artificial intelligence will guarantee the safety and quality of food. However, this will require huge investments in terms of resources, technologies, and expertise. This study reviews the potential of artificial intelligence in food processing. In addition, it presents the technologies and methods with potential applications in implementing automated technology-aided processing. A conceptual design for an automated food processing line comprised of various operational layers and processes targeted at enhancing the microbial safety and quality assurance of liquid foods such as milk and beverages is elaborated.

Modelling of inactivation kinetics of Escherichia coli and Listeria monocytogenes on grass carp treated by combining ultrasound with plasma functionalized buffer.

Linear (first-order) and non-linear (Weibull, biphasic, and log-logistic) models were evaluated for predicting the inactivation kinetics of Escherichia coli and Listeria monocytogenes on grass carp treated by a novel technique (UPFB) combining ultrasound (US) with plasma functionalized buffer (PFB). Results showed that UPFB was more effective for inactivating bacteria when compared with individual applications of US or PFB with reductions of 3.92 and 3.70 log CFU/g for Escherichia coli and Listeria monocytogenes, respectively. Compared with the linear model, the three non-linear models presented comparable performances and were more suitable for describing the inactivation kinetics with superior adj-R2 (0.962-0.999), accuracies (0.970-1.006) and bias factors (0.995-1.031), and by assessing the strengths of evidence, weights of evidence and evidence ratios for the models, the biphasic model was identified as the best fit model. The current study provided new insights into the effective evaluation of decontamination methods.

In situ investigation of cellular water transport and morphological changes during vacuum cooling of steamed breads.

Dynamic investigation of the effects of vacuum cooling on cellular water transport and structural changes of steamed bread was carried out using transverse relaxation times (T2) and proton density-weighted images in a nuclear magnetic resonance (NMR)/magnetic resonance imaging (MRI) analyser, respectively. Initially, both steamed bread at room temperature of 25 °C and freshly steamed bread at 85 °C had three peaks of T21, T22, and T23, respectively representing the tightly bound water, loosely bound water, and free water, while an additional peak T24, was observed in freshly steamed bread at 85 °C. After vacuum cooling, freshly steamed bread at 85 °C had a higher mass loss of 10.29% due to its high initial temperature, and both samples were clearly discriminated with PCA of 88.2%, indicating that the initial food condition affected the vacuum cooling process. Lastly, the NMR/MRI technique and correlations were accurate (R2> 0.98), thus suitable for model validation at microscale and macroscale.

Recent advances in multiscale CFD modelling of cooling processes and systems for the agrifood industry.

Spoilage of agrifood produce is a major issue in the industry. Cooling is an effective technique for extending the shelf life of fresh agrifood produce to minimize spoilage. Due to the practical inability of directly solving the wide spatial and temporal scales in large industrial agrifood cooling systems, the porous medium approach is mostly used. However, improvements of current porous medium models and modeling across much wider scales are needed to better understand the multiscale cooling process and system problems. Recently, as a result of increased computational capacity, multiscale computational fluid dynamics (CFD) modeling approaches have been developed to tackle some of these challenges. The associated problems and applications of CFD in the design and process optimization of cooling processes and systems at different scales are considered. CFD solution and scale bridging techniques relevant for handling multiscale cooling processes and systems problems are discussed. Innovative applications of various CFD modeling techniques at different scales in cooling processes and systems are reviewed. CFD modeling techniques can be used to handle multiscale cooling process and system problems. Lattice Boltzmann method (LBM) is a potentially viable discrete modeling technique for complimentary usages alongside current continuum techniques in future multiscale CFD modeling. The multiscale CFD modeling paradigm can overcome the computational resource limitations associated with the direct modeling approach and enhance model extension across wider spatial and temporal scales. Information from multiscale CFD could be used to improve the accuracy of current porous medium models, and thus the design of more efficient cooling systems.