A hybrid digital shadow to assess biological variability in carrot slices during drying

Digital twins allow non-invasive access to the hygrothermal processes of fruits and vegetables during drying. However, existing methods do not consider the heterogeneity of product particles throughout the process. This study uses physics-based and Monte Carlo simulations to predict natural variability in carrot slices from raw material to the final product. Data from literature to model sorption isotherm and effective diffusivity, to validate the hygrothermal model, and to provide input ranges for Monte Carlo simulations was used. Nusselt correlations define heat and mass transfer coefficients, while conductive heat transfer accounts for thermal contact between the mesh tray and carrot slice. Validation shows good agreement between model and experimental data. A first-order reaction kinetic describes the thermal decay of β-carotene satisfactorily. Sensitivity analysis identifies the parameters that impact drying time and carotene retention most: slice thickness and initial moisture content. The study also introduces a water activity assessment method for the carrot slice population in the dryer, applying it to assess convective drying efficiency under different conditions. Raising the drying air temperature from 50 °C to 70 °C, along with a shift to product temperature-controlled drying, achieves a 45 % reduction in drying time, a 27 % decrease in required heat energy, and an 8 % improvement in β-carotene retention. The described approach holds promising insights for optimizing drying processes without additional equipment. In addition, the combination of physical and Monte Carlo simulations will enable progress in areas where variability plays a decisive role.