Application of neural network models in predicting the response of broilers and layers to dietary amino acids

• Grant number: 13/22074-6
• Support Opportunities: Scholarships in Brazil - Doctorate
• Start date: March 01, 2014
• End date: September 30, 2014
• Field of knowledge: Agronomical Sciences - Animal Husbandry - Animal Nutrition and Feeding
• Principal Investigator: Nilva Kazue Sakomura
• Grantee: Ako Faridi
• Host Institution: Faculdade de Ciências Agrárias e Veterinárias (FCAV). Universidade Estadual Paulista (UNESP). Campus de Jaboticabal. Jaboticabal , SP, Brazil

Abstract

Proteins are one of the costly major items in animal diets. Therefore, maximizing the efficiency of protein and amino acids (AAs) utilization is very important for the reduction of feed costs and maximization of lean meat production with an absolute minimum intake of AAs. Even though the essential AA requirements for broiler chickens have been well documented, the prediction of performance in practical and useful terms to be used in deciding the most advantageous dietary AA patterns is still difficult. Growth modeling techniques allow nutritionists and poultry researchers to predict dynamic or daily AA requirements, by a process which seems to be more adequate than the use of fixed requirements. Among different mathematical based models, neural networks (NNs) are a relatively new option to model growth in animal production systems. NNs are computational modeling tools that have recently emerged and found extensive acceptance in many disciplines for modeling complex real-world problems. Therefore, it seems to be useful to apply the NN models for predicting the performance of broiler chickens and layers in response to AAs.