Validation of Prediction Equations of the Egg Characteristics in Laying Hens

Poultry researchers have used mathematical models to explain some laying bird responses. There are many equations of the egg characteristics, but their validity has not yet been confirmed. Therefore, the aim of the current study was to assess the accuracy of the equations proposed to predict egg characteristics in laying hens. A total of one hundred forty-eight prediction equations of egg characteristics were collected from different studies. A total of 781 eggs from 75-week-old Hy-Line laying hens were gathered to measure egg quality characteristics, and other egg characteristics were calculated using these measurements. The residuals of the difference of observed and predicted values were used to calculate accuracy measurements like mean absolute deviation (MAD), mean squared error of prediction (MSEP), mean absolute percentage error (MAPE) and root mean squared error of prediction (RMSEP). RMSEP was used to estimate the error of the model (EM) with a 12% as the maximum level of validation of the egg characteristics prediction equation and 1.2% only for specific gravity. Nine egg characteristic prediction equations were validated with great accuracy because validated equations showed a value of MAD, MSEP, MAPE and RMSEP very low and EM less than 12%. Equations validated for external egg characteristic used easy-to-measure traits (i.e., egg weight, egg length, and egg width) as predictor variables. Fifteen egg characteristic prediction equations were validated with considerable accuracy. These equations might shorten the process of egg quality determination, reduce the waste of eggs, and thereby saving time and money.