Abstract
A stochastic pig growth model was developed to reproduce the nonlinear relationships between birth, weaning, and nursery-exit weights to later grow-finish BW, body composition, and Lys requirements. Serial grow-finish BW measurements of barrows and gilts were fitted to mixed model generalized Michaelis-Menten (GMM) equations. Two random effects of the GMM equations were predicted as functions of birth weight and 21-d weaning weight. A population of pigs was created to reproduce the variances and covariance of the random effects of the GMM function. The protein and lipid mass of each pig was predicted from serial real-time ultrasonic backfat and loin depth measurements. Predicted protein and lipid mass data of the barrows and gilts were fitted to functions of BW, which included pig-specific random effects. The random effects for protein mass were fitted to regression equations including birth weight, 21-d BW, and parameters of the GMM function. The sorting of pigs into light and heavy groups based on weaning weight, nursery-exit weight, or 84-d BW was evaluated. Sorting based on nursery-exit weight produced groups of pigs with larger differences in days to 125-kg BW than sorting based on weaning weight. Sorting based on BW resulted in more precise feeding of pigs relative to their daily Lys requirements.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call