Decision making in commercial poultry production using mathematical modeling based on intelligent systems and IoT aiming at production improvement

• Grant number: 19/16525-1
• Support Opportunities: Research Grants - Innovative Research in Small Business - PIPE
• Duration: January 01, 2021 - September 30, 2021
• Field of knowledge: Agronomical Sciences - Agricultural Engineering
• Principal Investigator: Thayla Morandi Ridolfi de Carvalho Curi
• Grantee: Thayla Morandi Ridolfi de Carvalho Curi
• Host Company: Thayla Morandi Ridolfi de Carvalho Curi
• CNAE: Atividades de apoio à pecuária ; Serviços de engenharia ; Pesquisa e desenvolvimento experimental em ciências físicas e naturais
• City: Campinas
• Associated researchers: Stanley Robson de Medeiros Oliveira
• Associated scholarship(s): 20/15512-0 - Decision making in commercial poultry production using mathematical modeling based on intelligent systems and IoT aiming at production improvement, BP.PIPE; 20/15612-5 - Decision making in commercial poultry production using mathematical modeling based on intelligent systems and IoT aiming at production improvement, BP.TT

Abstract

The purpose of this project is to develop an intelligent system based on artificial neural network and internet of things to assist decision making in broiler chicken production environmental handling aiming at better productivity, concerning poultry house typology and initial zootechnical conditions, in order to guarantee better efficiency and profitability. Broiler production in Brazil and in the world are very promising due to its ability to adapt, low cost per animal unit, rapid growth and, therefore, a good option of animal protein for human consumption. Although there are several positive aspects concerning broiler chicken production and trading, due to sanitary, nutrition, genetics and, mainly, environmental issues, handling is decisive factor to obtain the maximum potential of animal growth. Considering this scenario, this project's first phase includes the development of the following steps: data acquisition from growers and contractors, database setup for modeling, mathematical modeling development and validation using artificial neural networks. The next project's phase foresees the use of the productivity prediction model associated to the poultry house automation equipment by growers and contractors through a cloud-based platform (Internet of Things). Through the use of the model, the grower will be able to assertively identify the critical points and make better decision concerning the parameters that must be modified in other to achieve higher efficiency in poultry production. As the system will consider different levels of technification of the poultry house, it is expected to reach a large number of growers, offering useful tools that will help them improve their productivity indexes and profitability, so that producers can continuously implement technology in their properties, improve results and, consequently, contribute to Brazilian agribusiness development. (AU)