Modelagem fuzzy e efeito das mudanças climáticas nas respostas produtivas de frangos de corte criados em diferentes sistemas comerciais de produção

Abstract

Maintaining the thermal environment within the comfort ranges is essential for highlighting full genetic potential of the broiler lineage. So, the development of algorithms (mathematical models) to control the environment that can be embedded in microcontrollers becomes necessary. However, until now, modeling works on productive performance of broilers were individually performed for each production system or in laboratories where environmental conditions are controlled. Based on the exposed, this research aimed to develop a fuzzy system to predict the productive indexes of broilers, based on data from three commercial production systems collected throughout the year, for actual and future scenarios. The conventional, negative pressure and dark house systems were evaluated, and these systems comprise the diverse technological levels available to the producers. For each system, it was evaluated six flocks during one year, being collected data about the thermal environment and productivity (feed intake, FI; weight gain, WG; feed conversion, FC; and productive efficiency index, PEI). The non-parametric Mann-Kendall test and linear regression analyze were applied to the temperature data of the climatologic local series to analyze the climate change trends, allowing future scenarios simulation. The fuzzy models, independently of the used pertinence curve type (triangular, trapezoidal or Gaussian), allowed estimating FI, WG, FC and PEI, with certain efficiency, of broilers submitted to different commercial production systems. However, the selection of the types of pertinence curves must be studied case-by-case, in order to obtain the smallest possible simulation errors. Studying the expected climate change trends for the studied region, the broiler house that obtained the best result was the Dark House, followed by the Negative Pressure and, finally, the Conventional.