Bi-objective optimization of nutrient intake and performance of broiler chickens using Gaussian process regression and genetic algorithm

This study investigated whether quantifying the trade-off between maxima of two response traits increases accuracy diet formulation. To achieve this, average daily weight gain (ADG) and gain:feed ratio (G:F) responses 7–21-day-old broiler chickens to dietary supply three nutrients (intake digestible glycine equivalents, threonine, total choline) were modeled using a newly developed hybrid machine learning-based method Gaussian process regression genetic algorithm. The dataset comprised 90 data lines. Model-fit-criteria indicated high model adjustment no prediction bias models. bi-objective optimization scenarios through Pareto front revealed maximized ADG G:F provided information on needed input that interact with each other scenarios. followed nonlinear pattern. choosing target values intermediate after single-objective is less accurate than feed formulation trade-off. In conclusion, knowledge nutrient inputs will help formulators optimize their more holistic approach.