Investigation of drying behavior and color development of beef slices for development of non-invasive monitoring approaches

Meat drying has a long tradition worldwide and technically dried meat is increasingly valued as a low-fat snack or high-protein food product. Convection drying is widely used in industrial food preservation and has to provide high process efficiency and high product qualities, which can be contradictory. Therefore, it is important to understand the changes that occur in the product in order to develop drying strategies that consider process and product efficiency each at an acceptable level. The objective of this work was to investigate the cumulative influence of maturation, cold storage, marinating, and drying temperature on the drying behavior and color change of beef with respect to the development of optimized drying strategies. Furthermore, prediction algorithms combining hyperspectral and laboratory data were developed to replace conventional invasive measurements with non-invasive measurement systems in the future in order to gain a deeper understanding of changes in the product via continuous real-time measurements. In a first trial, invasive laboratory measurements were used to investigate the drying behavior and color changes of fresh and matured meat, as well as the respective frozen and thawed variants. The investigations also included marinating with salt or a salt/vinegar solution. In a second experiment, the influence of the marinade and different drying air temperatures (50, 60 and 70 °C) were investigated. Simultaneous to the laboratory measurements, hyperspectral data were acquired from the samples, which were used together with the laboratory data to develop predictive models. Regarding the easily and cost-effective application of non-invasive measurement systems, the models were simplified by reducing wavelengths and the predicted data were checked for their agreement with invasively measured data. The results showed a clear influence of meat maturation and freezing/thawing on drying behavior, as well as marinating and air temperature. The influence of the various factors slightly affected the color change during drying, but did not result in any visible differences between the samples after drying. Predictive models for spectral measurement of moisture and color were successfully developed. By selecting wavelengths, the models could be simplified in each case. The robustness could be significantly increased over the course of the experiment by including the influencing factors used, and finally even an overall model could be developed that can predict moisture and color parameters simultaneously and was simplified by the selection of only 10 wavelengths. The results show that product-specific drying strategies can lead to more efficient processes and that spectral product monitoring is a promising method, especially against the background of the development of product-controlled, "intelligent drying systems".