Dynamic modeling and metaheuristic optimization for decision support in beef cattle production

Author(s)

Luís Gustavo Barioni

• Document type: Doctoral Thesis
• Press: Piracicaba. , gráficos, ilustrações, tabelas.
• Institution: Universidade de São Paulo (USP). Escola Superior de Agricultura Luiz de Queiroz (ESALA/BC)
• Defense date: 2002-04-22
• Advisor: Moacyr Corsi
• Field of knowledge: Agronomical Sciences - Animal Husbandry
• Indexed in: Banco de Dados Bibliográficos da USP-DEDALUS
• Location: Universidade de São Paulo. Biblioteca Central da Escola Superior de Agricultura Luiz de Queiroz; ESALQ-BC/t636.213; B253m 79317

Abstract

Strategic decisions to increase the profitability in feedlot operations such as number of animals, initial body weight, commence and end of the feeding period, and diet composition are required. ln order to better evaluate strategies combining those factors, a dynamic mathematical model of beef cattle growth was developed. The model is based on concepts of hyperplasia and hypertrophy. State variables are protein mass, DNA mass, fat mass and labile metabolizable energy reserve pool. The labile energy pool worked as a buffer between energy intake and energy use and modulated both rates. Dry Matter and Energy intake were calculated interactively and were function of animal energy demand that was calculated base on potential rates of protein and fat accretion. Fat accretion potential was modified by body fatness. Simulation /, results showed that the model enabled simulating decreasing growth rates with advance of animal maturity and increased intake and growth rates after restriction periods. lntensity and duration of compensatory growth were affected by intensity and duration intensity of restriction. Simulations with the growth model were used to determine the optimal slaughter date of the animals. Ration composition was optimized using linear programming and dynamic simulations, linear programming and a genetic algorithm running concurrently was used to determine solutions of an extensive optimization problem subject to constraints of capital and feed availability. Analyses of the simulations and optimizations results’ indicated that both the body weight at slaughter and the extension of the feedlot period were affect by the cost of the ration. Increasing the feed costs lowered the slaughter weights and shortened the feeding periods. Diets with the lowest production cost did not produce the maximum economical return given the conditions of beef prices. The date of purchase and sale were one of the most important aspects in reaching maximum economical return. When forage limitation was included in the problem the solutions tended to have later purchase, heavier animals shorter feeding period, and higher energy density in the diet. Slaughter date was however unchanged. The combination of a genetic algorithm, linear programming to diet optimization and dynamic simulation was able to generate robust solutions to the general feedlot problem. (AU)