Building models to evaluate internal comprehensive quality of apples and predict storage time

Apple quality and customer satisfaction are significantly impacted by variations in apple quality throughout storage. In this study, the evaluation of apples' internal overall quality and the estimation of storage time were investigated. By using Pearson correlation analysis and an analytical hierarchy approach, an internal comprehensive quality index was created. A calibration model and a high-order kinetic model were created for the internal comprehensive quality index using the competitive adaptive reweighted sampling (CARS) algorithm in conjunction with partial least squares regression (PLSR) and in accordance with the results of fitting chemical kinetic reactions to variations in internal comprehensive quality with storage time. The calibration model and the high-order kinetic model were combined to create a prediction model for the storage time of apples. Results revealed that the determination coefficient of the prediction (R2p) and root mean square error (RMSE) of the calibration model were 0.9419 and 0.0023 respectively, and a residual predictive deviation (RPD) of 5.77; the correlation coefficient (R) and RMSE of the higher-order kinetic model were 0.9620 and 0.0038 respectively; the R2p of the prediction model was determined as 0.8957, with a root mean square error of 4.63 d. Results show that the proposed calibration model and higher-order kinetic model are capable of evaluating the internal comprehensive quality of apples, and that the determined prediction model is capable of projecting the storage time of apples with an acceptable margin of error while still meeting the real requirements.