O conhecimento e consequentemente a inteligência passaram a ser umas das principais fontes de vantagem competitiva das empresas, dadas as grandes mudanças nas conjunturas dos mercados. Essa necessidade de inteligência, permeia todos os agentes produtivos que compõem o mercado, inclusive o do agronegócio. Tendo isso em mente, o objetivo central desta pesquisa foi propor um método para o processo de inteligência competitiva que consolide e integre a literatura de inteligência e contemple os aspectos críticos da indústria de insumos agropecuários no Brasil, considerando a hipótese de que que os métodos contemplam de forma superficial às necessidades específicas desse setor, e que um método mais específico poderia ser construído. Para atingir esse objetivo foram empregadas as seguintes estratégias de pesquisa: a) Pesquisa bibliográfica (desk research) e análise documental como um método de pesquisa integrado e complementar; b) Realização de entrevistas com especialistas; e c) Construção de método. Dessa forma, o resultado foi a construção de um método de inteligência competitivo para empresas de insumos do agronegócio brasileiro composto de nove etapas. A hipótese foi aceita e concluiu-se que este método atendeu aos objetivos propostos, pois reuniu a literatura existente, inclusive com conceitos digitais atuais (Big Data e Data Mining), além de atender às necessidades e especificidades do setor ao focar em seus aspectos críticos / The knowledge, and therefore, intelligence became one of the most important sources of competitive advantage for companies, given the markets\' major recent changes. This need for accurate intelligence, for those who really need it and on the time that it is needed permeates all agents, of all markets, in all sectors, including the agribusiness, one of the Brazil\'s most important economic sectors, specially the agribusiness input companies. With that in mind, this research\'s the main objective was to propose a framework for the competitive intelligence process that consolidates the literature regarding competitive intelligence and contemplates the Brazilian agribusiness\' input sector critical aspects. The hypothesis was that the existing frameworks contemplate superficially the sector\'s specific needs, due to the concepts\' generality and that a more specific framework could be built, taking into consideration the characteristics of that sector. To achieve this objective the following research strategies were employed: a) Bibliographic research (desk research) and documental analysis as an integrated research method; b) Interviews with specialists; and c) Framework construction. Therefore, the result was the construction of a competitive intelligence framework form Brazilian agribusiness\' input companies with nine phases. The hypothesis was accepted and in conclusion, this framework met the proposed objectives because it integrated the existing intelligence literature, including contemporary digital concepts such as Big Data and Data Mining, in addition to answering the sector\'s specific needs by focusing on its\' critical aspects

Comme dans les pays du Nord riches, il y a également une augmentation de la production de déchets organiques dans les pays du Sud moins avancés. Connues sous l'appellation de Matières Organiques Exogènes (MOEx), les déchets organiques peuvent être valorisables comme engrais ou amendement pour les sols agricoles ou bien une nouvelle source de biomasse pour la production d'énergie. L'utilisation raisonnée d'une MOEx exige toutefois une connaissance scientifique approfondie de leur nature et de leurs impacts sur l'environnement. Des analyses conventionnelles en laboratoire et des expérimentations doivent être effectués pour maîtriser leur structure complexe liée à leurs propriétés physiques, chimiques et biochimiques. Celles-ci sont généralement longues et onéreuses. Dans notre étude, la spectroscopie proche infrarouge (SPIR) a été choisie comme une alternative aux méthodes classiques pour la caractérisation de ces MOEx en milieu tropical humide. Les objectifs de notre travail sont (1) de montrer que la SPIR est une méthode appropriée pour la prédiction des caractéristiques de MOEx, (2) pour caractériser les MOEx en termes de leur potentiel agronomique et leur potentiel énergétique et (3) pour élaborer une typologie, construite à partir des données prédites par la SPIR, qui prend en compte quelques impacts négatifs liés à l'utilisation des MOEx. L'ensemble des données a inclus plus de 2000 échantillons de MOEx couvrant une large gamme de MOEx fraîches ou transformées issues de l'élevage, de la ville et des agroindustries. Le spectre SPIR de chaque EOM a été acquis. En plus des caractérisations classiques de laboratoire (les dosages de carbone, azote, phosphore, potassium, etc.), d'autres expérimentations ont été conduites en vue d’évaluer (1) les potentiels « positifs » des MOEx : (i) suivi de la minéralisation du carbone et de l'azote mesurée dans des conditions contrôlées pour une valorisation agronomique, (ii) détermination du pouvoir calorifique et du potentiel méthanogène pour une valorisation énergétique ; et (2) les potentiels « négatifs » des MOEx : (i) dénitrification potentielle et (ii) phytotoxicité en présence d’éléments trace métallique. Pour élaborer la typologie, la méthode SIRIS (Système d'Ingration des Risques par Interaction des Scores) a été appliquée pour formaliser les étapes d'une procédure logique menant à une classification des MOEx selon l'objectif de l'orientation de recyclage. La diversité des MOEx est illustrée à titre d'exemple par une large gamme des teneurs en (i) C (de 2 à 64 gC.100g-1MS), (ii) N (de 0,2 à 14gN.100g-1MS), (iii) P total (0,01 à 16gP.100g-1MS), K total (0,04 à 30 gK.100g-1MS). En prenant en compte l'ensemble des MOEx, les étalonnages des modèles SPIR sont acceptables pour la prédiction des teneurs en C, en N, en P et en K. Pour les potentiels « positifs » des MOEx (1) suivant les types de MOEx 102 à 955 kgMO.t-1MOEx pourraient être stockées dans le sol pour le maintien de la fertilité, (2) les pouvoirs calorifiques des MOEx varient de 7 à 28 MJ.kg-1MS et les potentiels méthanogènes vont de 69 à 488 NmlCH4.g-1MO. Quant aux potentiels négatifs, que ce soit pour les émissions de N2O ou pour la phytotoxicité des ETM, une forte interaction entre les propriétés du sol (pH, texture, …) avec les MOEx masque souvent certains effets. La méthode SIRIS a permis de classifier les MOEx selon un processus simple basé sur le classement de critères et de scores d'auto-pénalisation. Cette approche est intéressante en l'absence de données précises qui sont souvent difficiles à obtenir. L'aide des prédictions par la SPIR nous a été utile pour cette démarche d'orientation de la valorisation agronomique et/ou énergétique des MOEx. Des approches socio-économiques devraient compléter notre étude pour aboutir à l'orientation finale de valorisation des MOEx. / As observed in northern/rich countries, there is an increase in the production of organic wastes in southern/less advanced countries. So called, Exogenous Organic Matter (EOM), they can be a valuable fertilizer or amendment for agricultural soils or a new supply for energy production. The safe utilisation of EOM requires an in-depth scientific knowledge of their nature and impacts on the environment. Laboratory analysis and experiments have to be carried out in order to know their complex structure related to their physical, chemical and biochemical properties. These techniques represent a relatively high cost and are time consuming. In this study, the near infrared spectroscopy (NIRS) was chosen as an alternative to classical methods for laboratory characterisations of tropical EOM. The objectives of this study are (1) to show that NIRS is a suitable method for predicting EOM characteristics (2) to characterize EOM in terms of their agronomic potential and/or energetic potential, and (3) to elaborate a typology, build with predicted parameters by NIRS, taking into account some environmental impacts of the different utilisations of EOM.The dataset included more than 2000 EOM samples covering a broad range of fresh and transformed organic materials. Each EOM was scanned using a NIR spectrometer from 1100nm to 2500nm. In addition to the classical laboratory characterisations (dosages of total carbon, nitrogen, phosphorus, potassium), other experiments have been carried out: (1) for “positive” potentials: (i) C and N mineralization measured in controlled conditions (for an agronomic use), (ii) determination of High Heating Value and Biochemical Methane Potential (for an energetic use) and (2) for “negative” potential or risk: phytotoxicity assessment due to trace elements, and the determination of potential nitrous oxide emission. To elaborate the typology, the SIRIS method (System of Integration of Risk with Interaction of Scores) was carried out in order to formalize the steps of a logical procedure, leading to a decision according to objective of the recycling orientation.Diversity of EOM is illustrated by the variability of (i) C contents (from 2 to 64 gC.100g-1DM); (ii) N contents (from 0.2 to 14gN.100g-1DM); (iii) total P contents (from 0.01 to 16gP.100g-1DM),(iv) total K contents (from 0.04 to 30 gK.100g-1DM). When taking into account all EOM without regarding their nature, acceptable predictive models were obtained with NIRS calibration for estimating C, N, total P and total K contents for EOM.“Positive” potential assessment provided information about (1) remaining organic carbon might be stocked into soil after applying EOM (from 102 to 955 kg.t-1TOM) (2) High Heating Value of EOM when recycling as a supply of small scale combustion or thermo chemical conversion (from 7 to 28 MJ.kg-1DM), (3) Biochemical Methane Potential after EOM anaerobic digestion (from 69 to 488 NmlCH4.g-1TOM). Both “negative” potential which took into account phytotoxicity due to trace elements and nitrous oxide emissions were highly interacted on the soil properties (pH, texture). The SIRIS method appeared to be an efficient decision-making tool in this study. It made it possible to classify the EOM according to a simple process based on the ranking of criteria and a self-penalization scoring system. This approach can be used in the absence of accurate, reliable data, which are often difficult to obtain. The help of NIR prediction was useful for better orientate the EOM valorization between soil fertilization and energy production. Further concern will be focused on socio-economical approach in order to complete the final destination of EOM re-use.

Made available in DSpace on 2014-06-11T19:31:02Z (GMT). No. of bitstreams: 0 Previous issue date: 2011-11-04Bitstream added on 2014-06-13T20:21:41Z : No. of bitstreams: 1 andrade_mg_dr_jabo.pdf: 575183 bytes, checksum: c078d7148dcec00acf084e11bfb365c2 (MD5) / O uso do lodo de esgoto (LE) com finalidade agronômica e/ou condicionador é uma das alternativas ambientalmente e economicamente atraente, quando comparadas a outros destinos (incineração, aterro sanitário, etc.). As substâncias inorgânicas potencialmente tóxicas contidas no LE são uma das maiores preocupações ao meio ambiente e à cadeia trófica. O objetivo deste trabalho foi avaliar os teores e o efeito cumulativo dos elementos As, Ba, Cd, Cr, Cu, Hg, Mo, Ni, Pb, Se e Zn, em dois latossolos vermelhos (um eutroférrico – LVef, e um distrófico – LVd) e plantas de milho (Zea mays L.) cultivadas nos mesmos após a aplicação de LE por treze anos ininterruptos. Em condições de campo e delineamento experimental de blocos ao acaso, cada solo recebeu quatro tratamentos (0; 5; 10 e 20 t de LE ha1) com cinco repetições. Amostras dos insumos agrícolas aplicados na cultura do milho, LE, solos, materiais vegetais e materiais de referência certificados foram digeridas em forno de micro-ondas (método 3051A), e os analitos foram quantificados por ICP-OES. Após 13 aplicações anuais sucessivas de LE no LVef e LVd, os maiores teores dos ETs foram encontrados no LVef e ainda não atingiram o valor de investigação (VI) agrícola (mais restritivo) da resolução 420 do CONAMA. Persistindo a atual taxa de aplicação de LE e teores médios de ET incorporados aos solos, é de se esperar que o Ba, Cd, Cr, Cu, Ni e Zn serão os primeiros elementos a atingir o VI no LVef. Para o LVd, o primeiro elemento a atingir o VI é o Cr somente. Os teores de ET encontrados nos insumos agrícolas não devem causar impacto ambiental imediato. Os ETs não proporcionaram fitotoxidade ou contaminação nas partes aéreas, folhas para diagnose e grãos das plantas de milho / The use of sewage sludge (SS) as agronomic and/or soil conditioner purposes presents a correct environmental and economically attractive alternative when compared to other disposal methods as incineration and landfills. The potentially toxic inorganic substances contained in SS is a major concern for the environment and food chain. This work aimed evaluate the content and the cumulative effect of the Trace Elements (TE) As, Ba, Cd, Cr, Cu, Hg, Mo, Ni, Pb, Se, and Zn in two different soils: an Eutroferric Clayed Red Latosol (Typic Eutrorthox – “E”) and a Dystrophic Red Latosol (Typic Haplorthox – “H”) and with maize plants (Zea mays L.) growth after fertilize the soil with SS for thirteen years continuously. Under field conditions and under a randomized blocks experimental design each soil received four treatments (0, 5, 10, and 20 t of SS ha-1) with five replications. Fertilizers, SS, soils, vegetal materials, and the reference material were digested in microwave oven (method 3051A) and the analytes were quantified by ICP-OES. After thirteen years fertilizing the soil with SS in the soils “E” and “H”, the higher amounts were found in the “E” and these amounts did not reach the agricultural investigation value (more restrictive) yet. Persisting the actual fertilization rate of SS and an average amount of TE incorporated in the soil, Ba, Cd, Cr, Cu, Ni, and Zn will be probably the firsts elements to reach the agricultural investigation value in the “E”. To “H”, the first element to reach this value will be probably Cr. The TE found in the fertilizers should not cause immediate environmental impact. The TE did not provide neither phytotoxicity nor contamination in the maize aerial part, diagnose leaf, and grains

Thesis (MBA.) -- University of Limpopo, 2016 / The objective of this study was to determine the environmental impact of agricultural inputs, to know if the environmental impacts of agricultural inputs have environmental costs, and to know who bears the environmental costs of agricultural inputs. Using a purposive sampling method, the researcher studied six farms from Chief Albert Luthuli Municipal area, Mpumalanga. The research design for the study was a mix of qualitative and quantitative research approaches. Data collection was from primary and secondary sources. Data was collected from the six farmers in Albert Lithulu, in addition secondary data was collected from the archives of Index Mundi, the University of Pretoria, the US EPA and the World Bank. Using a mix of correlation and regression analysis, findings from the study provided an answer to the three research objectives. Findings from the analysis of correlation and regression indicated that agricultural input (fertilizer, used in this study) does affect the environment; it causes an agricultural induced emission of greenhouse gases (nitrous oxide and methane). Furthermore, findings from analysis of potential environmental costs of environmental impacts (methane and nitrous oxide) showed that agricultural inputs have social costs for South Africa. Furthermore, the final findings in this study showed that the environmental costs from agricultural inputs are born by the victims. Consequently the study recommends additional environmental regulation to enable farmers internalise some of the environmental costs of agricultural inputs that are born by victims. Further research is suggested to determine the model that may be used to internalise environmental costs of agricultural inputs back to the farms. Key Words: farms, environmental costs, agricultural inputs, externalities

Orientador: Wanderley José de Melo / Banca: Ademir Sérgio Ferreira de Araújo / Banca: Beatriz Monte Serrat / Banca: Marcos Omir Marques / Banca: Otávio Antônio de Camargo / Resumo: O uso do lodo de esgoto (LE) com finalidade agronômica e/ou condicionador é uma das alternativas ambientalmente e economicamente atraente, quando comparadas a outros destinos (incineração, aterro sanitário, etc.). As substâncias inorgânicas potencialmente tóxicas contidas no LE são uma das maiores preocupações ao meio ambiente e à cadeia trófica. O objetivo deste trabalho foi avaliar os teores e o efeito cumulativo dos elementos As, Ba, Cd, Cr, Cu, Hg, Mo, Ni, Pb, Se e Zn, em dois latossolos vermelhos (um eutroférrico - LVef, e um distrófico - LVd) e plantas de milho (Zea mays L.) cultivadas nos mesmos após a aplicação de LE por treze anos ininterruptos. Em condições de campo e delineamento experimental de blocos ao acaso, cada solo recebeu quatro tratamentos (0; 5; 10 e 20 t de LE ha1) com cinco repetições. Amostras dos insumos agrícolas aplicados na cultura do milho, LE, solos, materiais vegetais e materiais de referência certificados foram digeridas em forno de micro-ondas (método 3051A), e os analitos foram quantificados por ICP-OES. Após 13 aplicações anuais sucessivas de LE no LVef e LVd, os maiores teores dos ETs foram encontrados no LVef e ainda não atingiram o valor de investigação (VI) agrícola (mais restritivo) da resolução 420 do CONAMA. Persistindo a atual taxa de aplicação de LE e teores médios de ET incorporados aos solos, é de se esperar que o Ba, Cd, Cr, Cu, Ni e Zn serão os primeiros elementos a atingir o VI no LVef. Para o LVd, o primeiro elemento a atingir o VI é o Cr somente. Os teores de ET encontrados nos insumos agrícolas não devem causar impacto ambiental imediato. Os ETs não proporcionaram fitotoxidade ou contaminação nas partes aéreas, folhas para diagnose e grãos das plantas de milho / Abstract: The use of sewage sludge (SS) as agronomic and/or soil conditioner purposes presents a correct environmental and economically attractive alternative when compared to other disposal methods as incineration and landfills. The potentially toxic inorganic substances contained in SS is a major concern for the environment and food chain. This work aimed evaluate the content and the cumulative effect of the Trace Elements (TE) As, Ba, Cd, Cr, Cu, Hg, Mo, Ni, Pb, Se, and Zn in two different soils: an Eutroferric Clayed Red Latosol (Typic Eutrorthox - "E") and a Dystrophic Red Latosol (Typic Haplorthox - "H") and with maize plants (Zea mays L.) growth after fertilize the soil with SS for thirteen years continuously. Under field conditions and under a randomized blocks experimental design each soil received four treatments (0, 5, 10, and 20 t of SS ha-1) with five replications. Fertilizers, SS, soils, vegetal materials, and the reference material were digested in microwave oven (method 3051A) and the analytes were quantified by ICP-OES. After thirteen years fertilizing the soil with SS in the soils "E" and "H", the higher amounts were found in the "E" and these amounts did not reach the agricultural investigation value (more restrictive) yet. Persisting the actual fertilization rate of SS and an average amount of TE incorporated in the soil, Ba, Cd, Cr, Cu, Ni, and Zn will be probably the firsts elements to reach the agricultural investigation value in the "E". To "H", the first element to reach this value will be probably Cr. The TE found in the fertilizers should not cause immediate environmental impact. The TE did not provide neither phytotoxicity nor contamination in the maize aerial part, diagnose leaf, and grains / Doutor

In this thesis I inspect learning about adoption of technologies among cocoa farmers in Ghana, which are represented by non-labor inputs, particularly by fertilizer and hybrid seeds. Earlier research focused mainly on learning about returns associated with adoption of such innovative inputs. However, it is not clear whether the adopters learn about these returns or rather about what are the optimal amounts of these inputs. Therefore the focus of this thesis is to examine how do the farmers choose and learn about optimal amounts of inputs. Cocoa farming is very labor intensive, and thus this thesis concentrates on learning about both non-labor and labor inputs, which are closely connected. Similar research carried out in India suggests that heterogeneous returns among farmers might cause that the farmers rely rather on their own considerations than on observation of behavior of their village neighbors, i.e. social learning. The heterogeneous returns are also present among the Ghanaian cocoa farmers, which suggest that these farmers should similarly prefer individual learning over the social one. Using a model developed for estimation of the prevailing type of learning about the optimal amount of inputs, I show that the farmers do tend to prefer individual learning in case of the non-labor inputs but rather...

Despite 25 years of concerted efforts by African governments to adopt consistent policies for increasing food production, hunger and poverty are still prevalent in the continent. Using Bernanke's conceptualization of the credit channel theory of monetary policy, the purpose of this correlational study was to investigate whether a subsidy program, the National Accelerated Agricultural Inputs Access Program (NAAIAP), affected the rates of fertilizer usage and food production in Kakamega County, Western Kenya. Purposive stratified sampling was used to select 114 participants consisting of 72 farmers in each of the 2 groups: NAAIAP beneficiaries and nonbeneficiaries. Participants completed a survey on fertilizer usage rates, income earned, and surplus maize yield. Data were analyzed using multiple regression to test whether there was a difference between the beneficiary and nonbeneficiary groups regarding income, surplus product, and the dependent variable of fertilizer usage. Results indicated that beneficiaries of NAAIAP credit program bought and prepared to use fertilizers significantly earlier than did their counterparts. Further, the results of multiple regression indicated significant positive correlation (p <.05) between income earned from sale of surplus maize yield and quantity of fertilizer used by farmers in Kakamega County. These findings suggest that NAAIAP improved food security and farmers' income in Kakamega Count. This study contributes to social change by recommending to subsidy program administrators in Kakamega County to consider policy changes. Such policy changes may improve program outreach to resource-poor farmers and improve income and product yield in the agricultural sector of Kenya.

O objetivo principal do estudo é ofercer uma contribuição metodológica testando os métodos mais usados na análise da eficiência relativa, vis a vis comparando um produtor com um grupo de produtores assemelhados, ou então, um produtor com toda a amostra, como é o caso da fronteira estocástica. Os métodos testados foram o DEA (análise envoltória de dados), a fronteira estocástica e o procedimento de Varian, tendo como objetivo da investigação produtores de leite. Para se testar a hipótese que os produtores de leite são eficientes, quanto à escolha da isoquanta e do ponto que minimiza custo, dada a pressuposição de que os agricultores realizam escolhas corretas, utilizam-se dois grupos de métodos. Os métodos não paramétricos aproximam-se mais do ideal de comparar um produtor com o grupo em que se insere. A base de tecnologia é a do grupo, sem apelo à função de produção. Os paramétricos são mais exigentes, pois pressupõem uma função de produção que tem que ser estimada; porém, são mais ricos e consistentes, no que tange ao teste de hipótese. O DEA é classificado de não paramétrico, porque não propõe uma função com os parâmetros que são estimados. Mas, implicitamente, gera uma função de produção, via programação linear. São menos estruturados que a fronteira estocástica, e mais exigentes que o procedimento de Varian, no sentido de que este não pressupõe qualquer fronteira explícita, como o método da fronteira estocástica, ou fronteira implícita, como o DEA. O DEA e o procedimento de Varian são deterministas, visto não associar aos modelos qualquer estrutura de probabilidade. A fronteira estocástica explicitamente associa ao modelo uma estrutura de probabilidade, pela forma que se define o termo do erro da regressão; desta forma é mais rica em testes de hipótese. Duas amostras foram utilizadas para se testar as hipóteses propostas. A primeira refere-se a um grupo de 143 produtores comerciais de leite, com as propriedades distribuídas nos seis maiores estados produtores do Brasil. A segunda amostra de produtores de leite é composta de 114 observações localizadas no estado de Minas Gerais. O método de Varian foi o que produziu menos distúrios em relação aos insumos ou produto. Estabelece maiores incrementos à renda líquida que a fronteira estocástica e menos do que DEA. Sua solução fica mais próxima daquilo que é factível para cada produtor fazer. O método prescinde do conceito de uma fronteira, reordenando os custos em relação aos produtores que produziram mais e gastaram menos, embora o procedimento de reordenamento requeira programação quadrática. Portanto, é um método aderente à gestão. A fronteira estocástica não objetiva nem aumentar a renda líquida e nem reduzir custos. Seu efeito sobre a renda líquida foi até negativo e produziu um maior distúrbio nos insumos e muito pequeno no produto, o que sempre ocorrerá, quando a função de produção se ajustar adequadamente aos dados. O método Varian foi mais apropriado para testar a hipótese de que os produtores, por tentativa e erro, acabam se localizando na fronteira eficiente, no ponto que minimiza o custo. Uma ves que exige somente que os agricultores racionalizem os custos, obedecendo à regra de racionalização, e passando ao largo da dinâmica de mercado, pela qual os produtores convergem para o custo mínimo, o que pode demandar muito tempo, em função de restrições, inclusive de emprego em outros setores da economia. Com os dados de um ano ou de poucos anos, quando não se pode captar o movimento de convergência, o DEA e a maximização da renda líquida são procedimentos inadequados para testar a hipótese de Schultz. / The objective of the study is test the most applied methods in the analysis of the relative efficiency, purposing the comparation between a producer and a group of resembled producers, or a producer with the entire sample, as the stochastic frontier’s (random border) case. The tested methods were the DEA (Data Envelopment Analysis), the stochastic frontier and the Varian procedure, analyzing mainly milk producers. Looking for testing the hypothesis that the milk producers are efficient, relating choice of isoquant and the cost minimization point, giving the assumption of the producers achieve correct choices, and it was selected two group methods. The non parametric methods (distribution free) is the best approach to compare a producer with its group. The technology basis is supported by the group, without going through the production function. The parametric procedures are more exigent because they assume a production function that has been estimated, however is more consistent to the hypothesis test. The DEA is classified of non parametric approach, because it does not consider a function with estimated parameters, but, implicitly, it generates a production function, using linear programming. Its less structuralized than the stochastic frontier, more exigent than the Varian procedure, considering the sense that this method does not estimate any frontier, explicitly, as the stochastic frontier method (implicit border), or the DEA. The DEA and the Varian procedure are deterministic, not associated with models that hold back some structure of probability. The stochastic frontier is an associate to the probability structure model, since it defines the term of the regression error making richer the hypothesis tests. Two samples had been used to test the hypotheses. The first one takes a group of 143 commercial milk producers, whose farm enterprises were distributed in the six biggest producer states of Brazil. The second is established by a group of 114 milk producers located in the state of Minas Gerais. The Varian method produced little disturb related to the inputs or outputs. The net income obtained was bigger than the stochastic frontier and smaller than the DEA. This solution is near to the decision of producer based on feasible to make. The method requires the concept of a frontier, rearranging the costs in related to the producers that had produced more and spent less, even so, the reordering procedure requires quadratic programming. Therefore, it is a very convenient method of management. The stochastic frontier doesn’t pursue increase in the net income reduce the costs. In this study the effect on the net income was negative and produced larger disturbs in the inputs and smaller in the products. Facts that will always occur when the production function is adjust to the data. The Varian method was more appropriate in testing hypothesis of the producers, using the experiment and error test, locating them in the efficient frontier, where the point minimizes the costs is located. Therefore it demands that the producer rationalize the costs following the rationalization rule, where the producers drive themselves minimum cost. This procedure can demand much time, in function of constraints of the other sectors of the economy. Considering data of one or more years the DEA and the maximization of the net income have been inadequate procedures to test the hypothesis of Schultz, when it cannot collect convergence movement.

O objetivo principal do estudo é ofercer uma contribuição metodológica testando os métodos mais usados na análise da eficiência relativa, vis a vis comparando um produtor com um grupo de produtores assemelhados, ou então, um produtor com toda a amostra, como é o caso da fronteira estocástica. Os métodos testados foram o DEA (análise envoltória de dados), a fronteira estocástica e o procedimento de Varian, tendo como objetivo da investigação produtores de leite. Para se testar a hipótese que os produtores de leite são eficientes, quanto à escolha da isoquanta e do ponto que minimiza custo, dada a pressuposição de que os agricultores realizam escolhas corretas, utilizam-se dois grupos de métodos. Os métodos não paramétricos aproximam-se mais do ideal de comparar um produtor com o grupo em que se insere. A base de tecnologia é a do grupo, sem apelo à função de produção. Os paramétricos são mais exigentes, pois pressupõem uma função de produção que tem que ser estimada; porém, são mais ricos e consistentes, no que tange ao teste de hipótese. O DEA é classificado de não paramétrico, porque não propõe uma função com os parâmetros que são estimados. Mas, implicitamente, gera uma função de produção, via programação linear. São menos estruturados que a fronteira estocástica, e mais exigentes que o procedimento de Varian, no sentido de que este não pressupõe qualquer fronteira explícita, como o método da fronteira estocástica, ou fronteira implícita, como o DEA. O DEA e o procedimento de Varian são deterministas, visto não associar aos modelos qualquer estrutura de probabilidade. A fronteira estocástica explicitamente associa ao modelo uma estrutura de probabilidade, pela forma que se define o termo do erro da regressão; desta forma é mais rica em testes de hipótese. Duas amostras foram utilizadas para se testar as hipóteses propostas. A primeira refere-se a um grupo de 143 produtores comerciais de leite, com as propriedades distribuídas nos seis maiores estados produtores do Brasil. A segunda amostra de produtores de leite é composta de 114 observações localizadas no estado de Minas Gerais. O método de Varian foi o que produziu menos distúrios em relação aos insumos ou produto. Estabelece maiores incrementos à renda líquida que a fronteira estocástica e menos do que DEA. Sua solução fica mais próxima daquilo que é factível para cada produtor fazer. O método prescinde do conceito de uma fronteira, reordenando os custos em relação aos produtores que produziram mais e gastaram menos, embora o procedimento de reordenamento requeira programação quadrática. Portanto, é um método aderente à gestão. A fronteira estocástica não objetiva nem aumentar a renda líquida e nem reduzir custos. Seu efeito sobre a renda líquida foi até negativo e produziu um maior distúrbio nos insumos e muito pequeno no produto, o que sempre ocorrerá, quando a função de produção se ajustar adequadamente aos dados. O método Varian foi mais apropriado para testar a hipótese de que os produtores, por tentativa e erro, acabam se localizando na fronteira eficiente, no ponto que minimiza o custo. Uma ves que exige somente que os agricultores racionalizem os custos, obedecendo à regra de racionalização, e passando ao largo da dinâmica de mercado, pela qual os produtores convergem para o custo mínimo, o que pode demandar muito tempo, em função de restrições, inclusive de emprego em outros setores da economia. Com os dados de um ano ou de poucos anos, quando não se pode captar o movimento de convergência, o DEA e a maximização da renda líquida são procedimentos inadequados para testar a hipótese de Schultz. / The objective of the study is test the most applied methods in the analysis of the relative efficiency, purposing the comparation between a producer and a group of resembled producers, or a producer with the entire sample, as the stochastic frontier’s (random border) case. The tested methods were the DEA (Data Envelopment Analysis), the stochastic frontier and the Varian procedure, analyzing mainly milk producers. Looking for testing the hypothesis that the milk producers are efficient, relating choice of isoquant and the cost minimization point, giving the assumption of the producers achieve correct choices, and it was selected two group methods. The non parametric methods (distribution free) is the best approach to compare a producer with its group. The technology basis is supported by the group, without going through the production function. The parametric procedures are more exigent because they assume a production function that has been estimated, however is more consistent to the hypothesis test. The DEA is classified of non parametric approach, because it does not consider a function with estimated parameters, but, implicitly, it generates a production function, using linear programming. Its less structuralized than the stochastic frontier, more exigent than the Varian procedure, considering the sense that this method does not estimate any frontier, explicitly, as the stochastic frontier method (implicit border), or the DEA. The DEA and the Varian procedure are deterministic, not associated with models that hold back some structure of probability. The stochastic frontier is an associate to the probability structure model, since it defines the term of the regression error making richer the hypothesis tests. Two samples had been used to test the hypotheses. The first one takes a group of 143 commercial milk producers, whose farm enterprises were distributed in the six biggest producer states of Brazil. The second is established by a group of 114 milk producers located in the state of Minas Gerais. The Varian method produced little disturb related to the inputs or outputs. The net income obtained was bigger than the stochastic frontier and smaller than the DEA. This solution is near to the decision of producer based on feasible to make. The method requires the concept of a frontier, rearranging the costs in related to the producers that had produced more and spent less, even so, the reordering procedure requires quadratic programming. Therefore, it is a very convenient method of management. The stochastic frontier doesn’t pursue increase in the net income reduce the costs. In this study the effect on the net income was negative and produced larger disturbs in the inputs and smaller in the products. Facts that will always occur when the production function is adjust to the data. The Varian method was more appropriate in testing hypothesis of the producers, using the experiment and error test, locating them in the efficient frontier, where the point minimizes the costs is located. Therefore it demands that the producer rationalize the costs following the rationalization rule, where the producers drive themselves minimum cost. This procedure can demand much time, in function of constraints of the other sectors of the economy. Considering data of one or more years the DEA and the maximization of the net income have been inadequate procedures to test the hypothesis of Schultz, when it cannot collect convergence movement.