Neural network models for predicting early egg weight in broiler breeder hens

作者：Faridi A^*; France J; Golian A

来源：Journal of Applied Poultry Research, 2013, 22(1): 1-8.

DOI：10.3382/japr.2011-00446

• broiler breeder
• egg weight
• neural network
• optimization

In this study, neural network (NN) models were developed to predict egg weight in broiler breeder hens. The input variables for developing the NN models were ME (kcal/bird per day) and CP, TSAA, Lys, Ca, available P, and linoleic acid (all as g/bird per day). By grouping the data collected from 98 breeder houses into weekly intervals, 4 NN-based models were developed for 25 to 28 wk of age. From the available data set (98 data lines for each week), a training set (n = 69) and a testing set (n = 34) were extracted. The models developed were subjected to an optimization algorithm to find the optimal values of input variables that might maximize early egg weight in broiler breeder hens. According to goodness-of-fit statistical criteria, the NN-based models could effectively estimate egg weight in broiler breeder hens. Maximum egg weight, using optimization results, may be obtained with 406, 454, 466, and 487 kcal/bird per day of ME; 21.3, 24.9, 25.6, and 26 g/bird per day of CP; 0.88, 0.97, 1.09, and 1.1 g/bird per day of TSAA; 1.02, 1.1, 1.22, and 1.23 g/bird per day of Lys; 4.13, 4.8, 5.2, and 5.27 g/bird per day of Ca; 0.52, 0.57, 0.6, and 0.62 g/bird per day of available P; and 1.97, 2.01, 2.28, and 2.3 g/bird per day of linoleic acid for 25, 26, 27, and 28 wk of age, respectively. Therefore, the energy and other nutrient requirements of broiler breeder hens for maximum egg weight do not change in parallel with age. Moreover, the Ross guideline recommendation seemed to underestimate the nutrient requirements of hens during these weeks.

• 出版日期2013-3