Comparative analysis of classical growth models and artificial neural networks in predicting egg production parameters in three commercial broiler parent stocks.

ABSTRACT: In the rapidly evolving poultry industry, effective management is dependent on precise forecasting of production metrics. As producers seek to maximize yield while minimizing costs, the ability to accurately predict egg production parameters becomes essential. Traditional growth models have long been employed for this purpose, but with advancements in computational techniques, artificial neural networks (ANNs) present a promising alternative. This paper examines the efficacy of various classical growth models, Gamma, compartmental, logistic, curvilinear Gompertz, Richards, and Morgan models, alongside a multi-layer feed-forward ANN model in predicting weekly and cumulative production metrics of three distinct commercial broiler parent stock (Ross, Arbor Acres, and Cobb). The analysis revea

SUBMITTER: Moradi Gharajeh Z 

PROVIDER: S-EPMC12828518 | biostudies-literature | 2025 Dec

REPOSITORIES: biostudies-literature