Comparison of mathematical and comparative slaughter methodologies for determination of heat production and energy retention in broilers

Review: When worlds collide - poultry modeling in the 'Big Data' era

Within poultry production systems, models have provided vital decision support, opportunity analysis, and performance optimization capabilities to nutritionists and producers for decades. In recent years, due to the advancement of digital and sensor technologies, 'Big Data' streams have emerged, optimally positioned to be analyzed by machine-learning (ML) modeling approaches, with strengths in forecasting and prediction. This review explores the evolution of empirical and mechanistic models in poultry production systems, and how these models may interact with new digital tools and technologies. This review will also examine the emergence of ML and Big Data in the poultry production sector, and the emergence of precision feeding and automation of poultry production systems. There are several promising directions for the field, including: (1) application of Big Data analytics (e.g., sensor-based technologies, precision feeding systems) and ML methodologies (e.g., unsupervised and supervised learning algorithms) to feed more precisely to production targets given a 'known' individual animal, and (2) combination and hybridization of data-driven and mechanistic modeling approaches to bridge decision support with improved forecasting capabilities.

Modeling life-time energy partitioning in broiler breeders with differing body weight and rearing photoperiods

Understanding energy partitioning in broiler breeders is needed to provide efficiency indicators for breeding purposes. This study compared 4 nonlinear models partitioning metabolizable energy (ME) intake to BW, average daily gain (ADG), and egg mass (EM) and described the effect of BW and rearing photoperiod on energy partitioning. Ross 708 broiler breeders (n = 180) were kept in 6 pens, controlling individual BW of free run birds with precision feeding stations. Half of the birds in each chamber were assigned to the breeder-recommended target BW curve (Standard) or to an accelerated target BW curve reaching the 21-week BW at week 18 (High). Pairs of chambers were randomly assigned to 8L:16D, 10L:14D, or 12L:12D rearing photoschedules and photostimulated with 16L:8D at week 21. Model [I] was: MEI_d = a × BW^b + c × ADG × BW^d + e × EM + ε, where MEI_d = daily ME intake (kcal/day); BW in kg; ADG in g/day; EM in g/day. Models [II–IV] were nonlinear mixed versions of model [I] and included individual [II], age-related [III], or both individual and age-related [IV] random terms to explain these sources of variation in maintenance requirement (a). Differences were reported as significant at P ≤ 0.05. The mean square error was 2,111, 1,532, 1,668, and 46 for models [I–IV] respectively, inferring extra random variation was explained by incorporating 1 or 2 random terms. Estimated ME partitioned to maintenance [IV] was 130.6 ± 1.15 kcal/kg^0.58, and the ME requirement for ADG and EM were 0.63 ± 0.03 kcal/g/kg^0.54 and 2.42 ± 0.04 kcal/g, respectively. During the laying period, maintenance estimates were 124.2 and 137.4 kcal/kg^0.58 for standard and high BW treatment, and 130.7, 132.2, and 129.5 kcal/kg^0.58 for the 8L:16D, 10L:14D, or 12L:12D treatments, respectively. Although hens on the standard BW treatment with a 12L:12D rearing photoschedule were most energetically conservative, their reproductive performance was the poorest. Model IV provided a new biologically sound method for estimation of life-time energy partitioning in broiler breeders including an age-related random term.