O ordenamento territorial em assentamentos rurais: uma análise utilizando algoritmos genéticos

• Ano de defesa: 2007
• Autor(a) principal: Santos Junior, Edgard Carneiro dos
• Orientador(a): Ferreira Neto, José Ambrosio
• Banca de defesa: Alves Filho, Eloy , Fiúza, Ana Louise de Carvalho , Ker, João Carlos
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Viçosa
• Programa de Pós-Graduação: Mestrado em Extensão Rural
• Departamento: Instituições sociais e desenvolvimento; Cultura, processos sociais e conhecimento
• País: BR
• Palavras-chave em Português: Ordenamento territorial; Assentamentos rurais; Algoritmos genéticos
• Palavras-chave em Inglês: Territorial order; Agricultural nestings; Genetic algorithms
• Área do conhecimento CNPq: CNPQ::CIENCIAS AGRARIAS::AGRONOMIA::EXTENSAO RURAL
• Link de acesso: http://locus.ufv.br/handle/123456789/4083

In this work it was tried to analyze the territorial organization process from six human nesting projects of three distinct physiographic regions from the Minas Gerais State, taking as reference the pedological aspects, and the vegetal covering.Using analysis, it was tried to identify the criteria used for the INCRA (Instituto Nacional de Colonização e Reforma Agrária) (National Institute of Colonization and Agrarian Reformation) in the effectuation of this action, as well as comparing the support capacity of each environment, aiming the human occupation. This research evidenced that the PAs (Projetos de Assentamento) (Nesting s Projects) of the Noroeste Mineiro (Northwestern region from the Minas Gerais State) offer better conditions for human occupation when compared with the PAs of the Triângulo Mineiro (Triangle region from the Minas Gerais State) and these ones, are better that the PAs from the Vale do Jequitinhonha (Jequitinhonha Valley). A new procedure to carry through the territorial order in human nesting s projects through a Genetic Algorithm s (AG) utilization was also developed. (GAC). The genetic algorithm was tested at Veredas Agricultural Nestings, located in Padre Paraíso City, at the Jequitinhonha Valley s region, in Minas Gerais State, Brazil, and its implementation was based on the system of lands agricultural aptitude and on their productivity index attribution. The tests sequence was carried through on a 391,88 ha area, containing eight different agricultural aptitude classes, parceled out in 12 lots. The effectiveness measurement used was the shunting line standard s value of a parceled area lots productivity index. To each evaluated parameter, a battery of 15 repetitions was carried through, recording the best individual s fitness average (MMI) found out for each parameter s values variation. The best parameters combination found out in the tests and used to generate the new parceling proposal by the GAC were the following ones: generation number being 320, population s size of 40 individuals, 0,8 mutation s tax and, a 0,3 renewal s tax. The new proposal generated rather homogeneous lots in productive capacity terms.