Predictive modelling of internal egg quality: Correlation structure, validation of existing models, and development...

The objective of this study was to evaluate the effects of storage time on internal and external egg quality traits, determine the relationships among these variables, validate previously published prediction equations, and develop new regression models for key quality parameters. A total of 670 eggs were evaluated at three storage times (0, 7, and 14 d). External and internal characteristics were measured individually, and correlation analyses, equation validation, and model development were performed using linear regression. Correlations revealed strong associations among egg weight, egg dimensions, albumen weight, yolk weight, and eggshell surface area, while albumen height and Haugh Units, and yolk height and Yolk Index also showed strong positive correlations. Storage-induced changes were reflected in negative correlations between albumen diameter and albumen height, and between yolk diameter and Yolk Index. Four published prediction equations (yolk height, yolk weight, eggshell surface area, and egg shape index) were validated with acceptable error values (<12 %). Eight equations were evaluated, of which four met the acceptance criteria. Six new prediction equations were developed for albumen diameter (two models), albumen height, albumen weight, shell weight, and shell thickness. The models showed a wide range of predictive performance, with coefficients of determination (R²) from 1.13 % to 83.34 %. Albumen weight exhibited the strongest predictive accuracy, whereas albumen diameter showed low predictability, reflecting its high biological variability and sensitivity to storage effects. Lack-of-fit tests confirmed adequate model structure for albumen diameter (equation 1), albumen weight, and shell weight. Overall, the results demonstrate that storage time markedly influences egg quality traits and that multiple traits are strongly interrelated. The validated and newly developed equations provide practical tools for estimating internal quality characteristics through non-destructive external measurements, reducing the need for destructive sampling and contributing to improved decision-making in egg quality control programs.