Suporte à decisão para o planejamento agrícola utilizando algoritmos genéticos

Abstract

The North of Minas Gerais has in the irrigated agriculture an option for its development. However, irrigation projects contain many components that are interrelated, forming a complex system. The inadequate use of the irrigation systems can limit the agricultural production, being fundamental the implementation of models that provide an efficient use ofthe available resources. As the water resources are limited and the economical and financial resources are difficult to obtain, all effort for a rational agricultural planning has a real social interest. In view of that, this dissertation presents the use of a model for the optimization of the implantation of irrigation projects. The objective is to demonstrate the viability of the use of an optimization model based on the method of the Genetic Algorithms (AG) applied in the maximization of the expected income for perennial cultures. For so much, a support model was used for the irrigated agriculture, program IRRIGA, to estimate the consumption of water of the different cultures considered in this study. For such, the needs of irrigated water of the cultures were simulated for the overhead irrigations by aspersion and located with the type of supplemental irrigation. The results of those simulations are one of the entrance variables the daily consumption of water of the cultures for the optimization model through AG. The method of AG was used to maximize the income of the agricultural property in function of the areas used for irrigation of the different cultures. The proposed model was applied in amanagerial typical lot, with an area of 50ha of the Gleba C2 of the first Stage of the Jaíba Irrigation Project. In this sense, the results indicate the efficiency and effectiveness of the application of the method of AG to solve problems of allocation of resources, coming, therefore, as an important tool to supply alternatives of agricultural planning.