Efficiency analysis of a learning system based on Artificial Intelligence to estimate welfare of broiler breeders

• Grant number: 08/05543-4
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: November 01, 2008
• End date: October 31, 2010
• Field of knowledge: Agronomical Sciences - Agricultural Engineering - Rural Buildings and Ambience
• Principal Investigator: Danilo Florentino Pereira
• Grantee: Luiza Souza da Costa
• Host Institution: Universidade Estadual Paulista (UNESP). Campus de Tupã. Tupã , SP, Brazil

Abstract

The animal welfare is such a relevant point to Europeans consumers that, to stop the social concern about the welfare of animals destined for human consumption, the European Union arranged, in its legislation, several directress and regulations about animals breeding and slaughtering. To attempt these regulations, protein animals' productions systems must adequate themselves and, in some cases, reorganize all production system in order to continue doing business at that market. On the search to attempt that demand, lots of papers have studied efficient ways to monitor shelteredanimals' welfare at intensive productions' system. This paper has the proposal of apply and validate in field conditions a fuzzy controller already developed to esteem welfare of broiler breeders in function of behaviors frequencies and durations expressed by the animals. It is also a main point of this present paper to validity a learning algorithm developed for that controller, analyzing the of the right rate evolution. This proposal is a integrant part of a bigger proposal that intend to develop a specialist system to esteem welfare of broiler breeders based on machine and artificial intelligence. The learningalgorithm validity is an important step to the elaboration of a based on artificial intelligence (AI) software that may become an important tool used for taking decisions about welfare of broiler breeders in intensive production.