The use of nonlinear programming as a tool for maximum profit feed formulation of broilers

Abstract

The purpose of this study was to compare linear and nonlinear programming models for feed formulation, for maximum profit, considering the real variation in the prices of the corn, soybean meal and broilers during the period from January of 2008 to October of 2009, in the São Paulo State, Brazil. For the nonlinear formulation model, it was considered the following scenarios of prices: a) the minimum broiler price and the maximum prices of the corn and soybean meal during the period, b) the mean prices of the broiler, corn and soybean meal in the period and c) the maximum broiler price and the minimum prices of the corn and soybean meal, in the considered period; while for the linear formulation model, it was considered just the prices of the corn and the soybean. It was used the Practical Program for Feed Formulation 2.0 for the diets establishment. A total of 300 Cobb male chicks were randomly assigned to the 4 dietary treatments with 5 replicate pens of 15 chicks each. The birds were fed with a starter diet until 21 d and a grower diet from 22 to 42 d of age, and they had ad libitum access to feed and water, on floor with wood shavings as litter. The broilers were raised in an environmentally-controlled house. Body weight, body weight gain, feed intake, feed conversion ratio and profitability (related to the prices variation of the broilers and ingredients) were obtained at 42 d of age. It was found that the broilers fed with the diet formulated with the linear model presented the lowest feed intake and feed conversion ratio as compared with the broilers fed with diets from nonlinear formulation models. There were no significant differences in body weight and body weight gain among the treatments. Nevertheless, the profitabilities of the diets from the nonlinear model were significantly higher than that one from the linear formulation model, when the corn and soybean meal prices were near or below their average values for the studied period, for any broiler chicken price.