- Carlos António Bana e Costa, João Carlos Namorado Clímaco, Marcos Pereira Estellita Lins, Aldizio Ferreira Costa, Sérgio Antônio Abunahman, Antônio Braga Coscarelliedit
This work seeks to analyze the centrality of the Brazilian airport commercial system and compare flights data of 2015 and 2019. This paper is an extension and updated study from article presented at VII RIDITA (Pereira et al., 2019). We... more
This work seeks to analyze the centrality of the Brazilian airport commercial system and compare flights data of 2015 and 2019. This paper is an extension and updated study from article presented at VII RIDITA (Pereira et al., 2019). We evaluated the Brazilian airport system by the results of layer centrality and h-centrality. These methodologies of centrality, unlike other traditional methods in graphs theory, try to evaluate the comprehensiveness of each vertex (in the case of the airport) and the influence on the others in the network. The results show the main characteristic of the Brazilian airport system regarding the concentration of the operation in regions and airports. Central airports demand greater attention from the agents in several operational fields (maintenance, traffic control, mobility, etc.) to raise the quality of the system.
Research Interests:
Parametric DEA models presume a functional form for the efficiency frontier and are used for resource redistribution. Parabolic DEA is a particular case of these models used to redistribute a single input in Variable Returns to Scale... more
Parametric DEA models presume a functional form for the efficiency frontier and are used for resource redistribution. Parabolic DEA is a particular case of these models used to redistribute a single input in Variable Returns to Scale scenarios. In this study, we extend this model to redistribute multiple inputs simultaneously, ensuring that all DMUs will become extreme efficient after redistribution maintaining their outputs. The proposed model is a single multi-objective Linear Programming Problem (PPL) which provides a single optimal solution. To solve this model, two approaches are used, the Weighted Sum of the Objective Functions and the Separation of Variables. Two numerical examples considering single and multiple outputs are used and the results obtained are identical for the two approaches.
Research Interests:
Reducing environmental impacts while maintaining high production levels is a challenge for environmentally sustainable agricultural production, which is more complex compared to other production systems. In this sense, it is important to... more
Reducing environmental impacts while maintaining high production levels is a challenge for environmentally sustainable agricultural production, which is more complex compared to other production systems. In this sense, it is important to measure environmental impacts and agricultural production in order to implement improvement plans. Moreover, it is relevant to add flexibility in the decision-making process for carrying out such implementations. The use of the Life Cycle Assessment + Data Envelopment Analysis (LCA + DEA) approach provides a tool to assess the eco-efficiency of units, considering environmental impacts, determining best practices and obtaining a single target for each inefficient unit. In this study, in order to add the necessary flexibility to the decision-making process, a multiobjective DEA model (MORO-D) is used, which provides a set of targets for each inefficient unit. According to their operational/managerial needs or limitations, the decision maker can choose a suitable target. This model is introduced within the four-step method for the Carbon Footprint (CF) + DEA approach, which leads to a new version of this method. We illustrated this new version assessing the eco-efficiency of organic blueberry (Vaccinium corymbosum) orchards. As a way to support the decision-making process, criteria for selecting one target from the set of targets are proposed. The results and also the advantages of the new version of the four-step method over the previous one are discussed. When consulted, the producers highlighted the advantages of having multiple alternatives due to the changing market conditions and they were more predisposed to select the easiest to implement. To the best of our knowledge, this is the first multiobjective model used within the LCA + DEA approach.
Research Interests:
The h-index was proposed in 2005, to objectively measure the impact and relevance of individuals' scientific output, based on their number of publications and citations. Since then, many works have studied the index's advantages and... more
The h-index was proposed in 2005, to objectively measure the impact and relevance of individuals' scientific output, based on their number of publications and citations. Since then, many works have studied the index's advantages and limitations, in various contexts and viewpoints. Still, we contribute to such a vast literature, by adopting a multi-criteria perspective to this matter. More precisely, we study the hindex in light of the fundamental axioms of coherence. We show that the number of publications and the number of total citations alone do not compose a coherent criteria family for the h-index, because it does not follow the axiom of exhaustivity. Thus, in this paper, we study examples from the literature to propose a suitable third criterion for the h-index, specifically, the form of the citation distribution among publications, which is less obvious than the other two. We verify that, with all three criteria, the criteria family that forms the h-index becomes coherent.
Research Interests:
This study aims to evaluate comparatively the situation of the federal units and the Brazilian states in relation to the pandemic of new coronaviruses (COVID19) through the technique of Network Data Envelopment Analysis (Network DEA -... more
This study aims to evaluate comparatively the situation of the federal units and the Brazilian states in relation to the pandemic of new coronaviruses (COVID19) through the technique of Network Data Envelopment Analysis (Network DEA - NDEA). For the development of research, data were collected on the Ministry of Health website, for all regions that register cases of virus cases notified until April 27, 2020. The purpose of the analysis is to assess regional discrepancies in the country. The model consists of the following structure: three inputs (number of doctors, number of respirators and number of clinical beds), an intermediate variable (number of reported cases) and one output (number of deaths). The results indicated that the federative unit with the worst performance overall was Amazonas, while the worst capital was Manaus. With two-dimensional representation, managers can visualize better which locations have the worst performance and assess which locations require more assistance. Depending on the results, managers can develop regional action plans, which can take steps to prevent the collapse of the health system.
Research Interests:
This work seeks to analyze the centrality of the Brazilian airport commercial system and compare flights data of 2015 and 2019. This paper is an extension and updated study from article presented at VII RIDITA (Pereira et al., 2019). We... more
This work seeks to analyze the centrality of the Brazilian airport commercial system and compare flights data of 2015 and 2019. This paper is an extension and updated study from article presented at VII RIDITA (Pereira et al., 2019). We evaluated the Brazilian airport system by the results of layer centrality and h-centrality. These methodologies of centrality, unlike other traditional methods in graphs theory, try to evaluate the comprehensiveness of each vertex (in the case of the airport) and the influence on the others in the network. The results show the main characteristic of the Brazilian airport system regarding the concentration of the operation in regions and airports. Central airports demand greater attention from the agents in several operational fields (maintenance, traffic control, mobility, etc.) to raise the quality of the system.
Research Interests:
We propose two-dimensional representations for the efficient frontier for two different Network DEA (NDEA) models. Both representations follow a previous study that developed a generalization of the two-dimensional representation method... more
We propose two-dimensional representations for the efficient frontier for two different Network DEA (NDEA) models. Both representations follow a previous study that developed a generalization of the two-dimensional representation method for the classic DEA using a new form of linearization. The graphical representation of the efficient frontier allows managers and decision-makers unfamiliar with linear programming and DEA to understand the results obtained simply and clearly. In this study, we can obtain graphs that provide information about the efficiencies of each DMU. Furthermore, for the NDEA with an efficiency decomposition approach, it is possible to obtain a graph to evaluate the technical efficiency for each stage, as well as for the overall efficiency. The NDEA model based on efficiency composition only produces one graph. However, this unique representation shows the stage-level and the overall process level information within the same graph
Research Interests:
The Olympic Games were created in 1896, where 241 men competed in 10 modalities. Since then, new editions have been held every 4 years, with more athletes, more countries, and more modalities. Although there is no official ranking... more
The Olympic Games were created in 1896, where 241 men competed in 10 modalities. Since then, new editions have been held every 4 years, with more athletes, more countries, and more modalities. Although there is no official ranking released by the International Olympic Committee (IOC), the media always elaborate rankings to evaluate the countries’ participation using the lexicographic method. However, the lexicographic method is often criticized as it overvalues the gold medal and disregards that each modality distributes a different number of medals. We propose a methodology to analyse the performance of the countries, through the application of successive h-indexes and h-cores, where the goal is not to generate a final ranking, but to identify the Olympic Powers in the history of the Games. Olympic Powers are the countries that stand out in a great variety of modalities, so they can be considered the great champions in the general picture. The concept of Olympic Power considers the number of times that a country occupies the first places of the ranking of each modality (or of each edition). As a result, we identified six countries as the Olympic Powers of history: USA, USSR, Germany, UK, China and Russia. Finally, we compared the result of the proposed methodology with the lexicographic ranking.
Keywords:Proposal of ranking based on calculation of successive h-indexes and h-cores; With this methodology we can identify potentials in different fields
Keywords:Proposal of ranking based on calculation of successive h-indexes and h-cores; With this methodology we can identify potentials in different fields
Research Interests: Olympic Games and H Index
The h-index, proposed in 2005, is very popular in the literature and in real-world applications, due to its simplicity and to the importance of measuring scientific impact. Many studies have proposed new indexes with similar... more
The h-index, proposed in 2005, is very popular in the literature and in real-world applications,
due to its simplicity and to the importance of measuring scientific impact. Many studies
have proposed new indexes with similar characteristics, known as h-type indexes, whose
purpose is to overcome certain disadvantages of the original index. This paper studies a
few of these indexes that increase the h-index’s sensitivity to frequently cited publications,
namely the g, A, R and the recently proposed Ψ-index. First, we compare the indexes’
responses to increases in the level of academic production, considering a real case study.
Moreover, we analyze the indexes’ stability, when the citation distribution varies. For a
better understanding of the indexes’ reactions to such changes, we compare them to the
p and z indexes.
Keywords: Decision processes, h-index, Bibliometric indexes, Comparisons between
indexes, h-type indexes, Index Stabil
due to its simplicity and to the importance of measuring scientific impact. Many studies
have proposed new indexes with similar characteristics, known as h-type indexes, whose
purpose is to overcome certain disadvantages of the original index. This paper studies a
few of these indexes that increase the h-index’s sensitivity to frequently cited publications,
namely the g, A, R and the recently proposed Ψ-index. First, we compare the indexes’
responses to increases in the level of academic production, considering a real case study.
Moreover, we analyze the indexes’ stability, when the citation distribution varies. For a
better understanding of the indexes’ reactions to such changes, we compare them to the
p and z indexes.
Keywords: Decision processes, h-index, Bibliometric indexes, Comparisons between
indexes, h-type indexes, Index Stabil
Research Interests:
Many developing countries have highly unequal health systems across their regions. The pandemic of COVID-19 brought an additional challenge, as hospital structures equipped with doctors, intensive care units and respirators are not... more
Many developing countries have highly unequal health systems across their regions. The
pandemic of COVID-19 brought an additional challenge, as hospital structures equipped
with doctors, intensive care units and respirators are not available to a sufficient extent in
all regions. Using Data Envelopment Analysis, we create a COVID Index to verify whether
the hospital structures in 543 Brazilian microregions are adequate to deal with COVID-19
and to verify whether public policies were implemented in the right direction. The results
indicate that hospital structures in the poorest microregions were the most vulnerable,
although the peak of COVID-19 occurred in the richest microregions (Sao Paulo). The
Southeast states could relocate hospital resources or even patients between their regions.
The relocation was not possible in many states in the Northeast, as the health system poorly
assisted the interior of these states. These findings reveal that the heterogeneity of microregions’
hospital structures follows the patterns of socioeconomic inequalities. We conclude
that it is easier for the wealthier regions to reallocate hospital resources internally than
for the poorest regions. By using the COVID Index, policymakers and hospital managers
have straightforward information to decide which regions must receive new investments
and reallocate underutilized resources.
pandemic of COVID-19 brought an additional challenge, as hospital structures equipped
with doctors, intensive care units and respirators are not available to a sufficient extent in
all regions. Using Data Envelopment Analysis, we create a COVID Index to verify whether
the hospital structures in 543 Brazilian microregions are adequate to deal with COVID-19
and to verify whether public policies were implemented in the right direction. The results
indicate that hospital structures in the poorest microregions were the most vulnerable,
although the peak of COVID-19 occurred in the richest microregions (Sao Paulo). The
Southeast states could relocate hospital resources or even patients between their regions.
The relocation was not possible in many states in the Northeast, as the health system poorly
assisted the interior of these states. These findings reveal that the heterogeneity of microregions’
hospital structures follows the patterns of socioeconomic inequalities. We conclude
that it is easier for the wealthier regions to reallocate hospital resources internally than
for the poorest regions. By using the COVID Index, policymakers and hospital managers
have straightforward information to decide which regions must receive new investments
and reallocate underutilized resources.
Research Interests:
In this paper, we extend the Parabolic DEA frontier to deal with integer values. Thus, we propose a hybrid approach that combines the Parabolic DEA frontier with the Sequential Algorithm to Resource Allocation in DEA. This is a two-phase... more
In this paper, we extend the Parabolic DEA frontier to deal with integer values. Thus, we propose a hybrid approach that
combines the Parabolic DEA frontier with the Sequential Algorithm to Resource Allocation in DEA. This is a two-phase
approach, in which the non-integer reallocation obtained by the parabolic DEA frontier is transformed into an integer
reallocation by the use of the Sequential Algorithm, while maintaining the highest DMUs’ efficiencies. We use this
hybrid approach to redistribute employees among 79 railway stations in Rio de Janeiro. By using this approach, all the
employees were redistributed respecting the integer values, and the system’s average efficiency has increased from 52%
to 96% through input redistribution.
combines the Parabolic DEA frontier with the Sequential Algorithm to Resource Allocation in DEA. This is a two-phase
approach, in which the non-integer reallocation obtained by the parabolic DEA frontier is transformed into an integer
reallocation by the use of the Sequential Algorithm, while maintaining the highest DMUs’ efficiencies. We use this
hybrid approach to redistribute employees among 79 railway stations in Rio de Janeiro. By using this approach, all the
employees were redistributed respecting the integer values, and the system’s average efficiency has increased from 52%
to 96% through input redistribution.
Research Interests:
Research Interests:
When analyzed from the perspective of one input and one output, the classic Data Envelop- ment Analysis (DEA) model (known as BCC after its developers Banker, Charnes, and Cooper) presents an efficient frontier with “downward” concavity... more
When analyzed from the perspective of one input and one output, the classic Data Envelop-
ment Analysis (DEA) model (known as BCC after its developers Banker, Charnes, and Cooper) presents
an efficient frontier with “downward” concavity (convex), therefore delivering variable returns to scale.
However, these returns show a decrease in marginal productivity as the number of inputs increases, that
is, the frontier presents decreasing global returns to scale. Both the convex frontier (DEA BCC) and the
concave frontier (“upward” concavity) present average productivities that vary along the curve; thus, the
local returns to scale are variable. It is claimed that the two formats are complementary, and therefore both
should be verified in the literature. Thus, this article proposes an algorithm capable of modeling an efficient
frontier, for one input or one output, with increasing global returns to scale, whereby an increase in input
causes an increase in marginal productivity.
ment Analysis (DEA) model (known as BCC after its developers Banker, Charnes, and Cooper) presents
an efficient frontier with “downward” concavity (convex), therefore delivering variable returns to scale.
However, these returns show a decrease in marginal productivity as the number of inputs increases, that
is, the frontier presents decreasing global returns to scale. Both the convex frontier (DEA BCC) and the
concave frontier (“upward” concavity) present average productivities that vary along the curve; thus, the
local returns to scale are variable. It is claimed that the two formats are complementary, and therefore both
should be verified in the literature. Thus, this article proposes an algorithm capable of modeling an efficient
frontier, for one input or one output, with increasing global returns to scale, whereby an increase in input
causes an increase in marginal productivity.
Research Interests:
Goal: This paper presents an analysis of the Brazilian air network based on the measure of h centrality. Design / Methodology / Approach: The purpose of this study is to define and determine the central network using the concepts of... more
Goal: This paper presents an analysis of the Brazilian air network based on the measure of h centrality. Design / Methodology / Approach: The purpose of this study is to define and determine the central network using the concepts of h-centrality. Method adapted from graph theory concepts with bibliometrics, especially the h-index. From the center of this index will be listed the most used airports to define the Brazilian central network and thus determinate the type h-index of Brazil. Results: The main airports in the results represent more than 50% of the homologated flights in Brazil. Limitations of the investigation: The centrality measure presents the situation at the moment when the data were analyzed. In the original h-index this notion reflects a history of the articles published by each author, but in the case of airports the market dynamics changes the behavior of the network. Practical implications: This model of air flight frequencies assists in public policy guidelines and planning both in air traffic control and in the verification of system impacts. Originality / Value: The method seeks to provide the list with main important airports without any prior definition by a decision maker.
Research Interests:
Resumo No período de 1995 a 2002, o governo brasileiro centrou sua política na busca de uma maior eficiência no uso dos recursos por parte das universidades federais. No período de 2003-2009, priorizou a democratização do acesso ao ensino... more
Resumo No período de 1995 a 2002, o governo brasileiro centrou sua política na busca de uma maior eficiência no uso dos recursos por parte das universidades federais. No período de 2003-2009, priorizou a democratização do acesso ao ensino superior, através de políticas afirmativas, oferta de bolsas nas instituições privadas (Prouni) e expansão das universidades federais. Esta pesquisa tem como foco avaliar a eficiência das instituições federais de educação superior (IFES) no conjunto das 52 instituições, no período de 1995 a 2009, face à trajetória do financiamento ao longo de dois governos com estratégias diferentes para atender o aumento de demanda pela educação superior. A avaliação foi feita utilizando-se a Análise Envoltória de Dados (DEA). Os resultados obtidos revelam que o período analisado alternou anos de eficiência e ineficiência, e que essa eficiência foi sendo alcançada paulatinamente. Palavras-chave: Financiamento. Eficiência. Educação Superior. IFES. S u b m e t i d o e m : 1 4 / 0 9 / 2 0 1 8 A p r o v a d o e m : 2 1 / 1 0 / 2 0 1 9 I Universidade Federal Fluminense (UFF), Niterói (RJ), Brasil;
Research Interests:
RESUMO Este estudo propõe um modelo de definição de ranking de empresas aéreas, adaptando os conceitos bibliométricos do índice-h e do índice-R (tipo h). Os destinos mais ofertados por cada aeroporto brasileiro é analisado e definido o... more
RESUMO Este estudo propõe um modelo de definição de ranking de empresas aéreas, adaptando os conceitos bibliométricos do índice-h e do índice-R (tipo h). Os destinos mais ofertados por cada aeroporto brasileiro é analisado e definido o núcleo do índice-h do mesmo. Após essa primeira análise é feita uma segunda analise por companhia aérea para obter o ranking final de cada índice. Verifica-se que o índice-R é o mais adequado para diferenciar aeroportos. Diretrizes para políticas públicas podem auxiliar na verificação de impactos no sistema com a verificação dos pontos centrais de cada companhia aérea e a importância de cada aeroporto. ABSTRACT This paper proposes a model for definition airlines ranking, adapting bibliometric concepts of the h-index and the R-index (type h). The destinations most offered by each Brazilian airport is analyzed and the h-index core of the same is defined. After this first analysis is made a second analysis by airline to obtain the final ranking by index. The R-index is the most suitable for differentiating airports. Guidelines for public policies and network analysis can assist in verifying impacts on the system by verifying each airline's central points and the importance of each airport.
Research Interests:
RESUMO Esse artigo propõe uma representação bidimensional para a fronteira de eficiência para um modelo Network DEA (NDEA). A representação é baseada em um estudo anterior que desenvolveu uma generalização do método de representação... more
RESUMO Esse artigo propõe uma representação bidimensional para a fronteira de eficiência para um modelo Network DEA (NDEA). A representação é baseada em um estudo anterior que desenvolveu uma generalização do método de representação bidimensional para o DEA clássico usando uma nova forma de linearização. Os valores dos inputs virtuais e dos outputs virtuais são utilizados para a elaboração do gráfico da fronteira de eficiência. A representação gráfica da fronteira de eficiência permite que gestores e tomadores de decisão não familiarizados com a programação linear e DEA entendam os resultados obtidos de maneira simples e clara. O gráfico obtido fornece informações sobre as eficiências da DMU e as metas que podem ser alcançadas por elas. Além disso, a representação obtida agrega informações tanto dos estágios internos, quanto do estágio global no mesmo gráfico para cada DMU. Palavra-chave: ABSTRACT This paper proposes a two-dimensional representation for the efficient frontier for a Network DEA (NDEA) model. The representation is based on a previous study that developed a generalization of the two-dimensional representation method for classical DEA using a new form of linearization. The values of virtual inputs and virtual outputs are used to draw the efficient frontier graph. The graphical representation of the efficiency frontier allows managers and decision makers unfamiliar with linear programming and DEA to understand the results obtained simply and clearly. The graph developed provides information about DMU efficiencies and the goals that can be achieved by them. In addition, 1
Research Interests:
Resumo Os modelos de Análise Envoltória de Dados (Data Envelopment Analysis – DEA) clássicos supõem que a produção de uma unidade (Decision Making Unit – DMU) não afeta a produção das demais. Isto significa que há total liberdade de... more
Resumo Os modelos de Análise Envoltória de Dados (Data Envelopment Analysis – DEA) clássicos supõem que a produção de uma unidade (Decision Making Unit – DMU) não afeta a produção das demais. Isto significa que há total liberdade de produção; entretanto, em alguns casos esta liberdade não existe. Gomes et al. (2001) propuseram uma alteração no modelo DEA BCC clássico que considera estas limitações. Neste modelo a soma dos outputs é constante, o que é semelhante a um jogo com soma zero, no qual o que é ganho por um jogador é perdido pelos demais. Por este motivo, este modelo é chamado de modelo DEA com Ganhos de Soma Zero (GSZ-DEA). Em alguns casos, podem-se encontrar barreiras práticas na realização em curto prazo dos alvos calculados; uma determinada DMU que busca eficiência pode não ser capaz de alcançar aquele alvo que lhe é atribuído. É, assim, proposta uma busca de alvos por etapas (estratégia de busca seqüencial), com alvos intermediários nas camadas de iso-eficiências. A busca pode ser feita por uma única DMU ou por várias, em regime de cooperação. Podem-se construir resultados análogos para o modelo com soma de inputs constante, que têm aplicação direta à alocação e realocação de recursos. Palavras-chave: DEA; Ganhos de soma zero; Busca seqüencial de alvos. Abstract Classical Data Envelopment Analysis (DEA) models assume that there is complete freedom of production, i.e., that the production of a given Decision Making Unit (DMU) does not affect the production of the others. Nevertheless, in some cases this, freedom does not exist. Gomes et al. (2001) propose changes in the classic DEA models in order to consider this constraint. In this model the sum of outputs is constant. It is similar to a zero sum game in which whatever is won by a player is lost by one or more of the others. For this reason this model is known as Zero SUM Gains DEA model. However, some practical barriers can be found when trying to achieve the computed targets; an inefficient DMU can be not able to reach immediately the computed target. We propose a step-by-step search for achievable targets. These intermediate targets are in the iso-efficiency layers. This search can be done by a solely DMU or by cooperating DMUs. Analogue results can be built for a constant input sum model, with direct application to resources allocation and reallocation.. (2001). Modelos DEA com soma de outputs constante. Anais do XXXIII Simpósio Brasileiro de Pesquisa Operacional, Campos do Jordão, SP, Novembro. Trabalho desenvolvido sob orientação do Prof. Marcos Pereira Estellita Lins (COPPE/UFRJ).
Research Interests:
RESUMO A preocupação com as emissões de Gases do Efeito Estufa (GEE) é crescente e recorrente mundialmente frente a busca pelo aumento de produtividade e eficiência nas indústrias. Este artigo visa contribuir com uma análise da... more
RESUMO A preocupação com as emissões de Gases do Efeito Estufa (GEE) é crescente e recorrente mundialmente frente a busca pelo aumento de produtividade e eficiência nas indústrias. Este artigo visa contribuir com uma análise da ecoeficiência de um grupo de vinhedos localizados na região central do Chile. A proposta apresentada utiliza o MCDEA (Multi Criteria Date Envelopment Analysis) para verificar a ecoeficiência com avaliação da produção com menos recursos e menores impactos ambientais verificados pelo método de Pegada de Carbono (Carbon Footprint) que neste caso é tratado com um output indesejável. O MCDEA de Li e Reeves (1999) proporciona uma melhor discriminação da eficiência da amostra analisada considerando a limitação da quantidade de variáveis para métodos clássicos em DEA. Os resultados apresentados determinam as melhores práticas entre os vinhedos dando maior importância para aqueles que utilizam menos recursos e geram menor impacto ambiental. PALAVRAS CHAVE. Eco-eficiência, LCA, MCDEA, práticas em agricultura. Tópicos DEA, AG&MA-PO ABSTRACT Concern about Greenhouse Gas (GHG) emissions is growing and recurring worldwide in the search for increased productivity and efficiency in industries. This article aims to contribute to an analysis of the eco-efficiency of a group of vineyards located in central Chile. The present proposal uses the MCDEA (Multi Criteria Date Envelopment Analysis) to verify the eco-efficiency with evaluation of the production with less resources and lower environmental impacts verified by the Carbon Footprint method which in this case is treated as undesirable output. The MCDEA of Li and Reeves (1999) provides a better discrimination of the efficiency of the analyzed sample by considering the limitation on variables quantities for classical DEA methods. The results presented show a better discrimination among the vineyards and greater importance for those who use less resources and generate lower environmental impact.
Research Interests:
RESUMO Este artigo tem como objetivo avaliar a eficiência de empresas distribuidoras de gás natural no Brasil. Utilizou-se do modelo de Análise Envoltória de Dados com redes (NDEA) para compreender as ineficiências dos processos... more
RESUMO Este artigo tem como objetivo avaliar a eficiência de empresas distribuidoras de gás natural no Brasil. Utilizou-se do modelo de Análise Envoltória de Dados com redes (NDEA) para compreender as ineficiências dos processos produtivos desse setor. Além disso, foi exigido um compartilhamento dos inputs, tendo em vista que as empresas apresentam um processo subdividido em dois estágios. O modelo apresenta a variável "custos operacionais" como único input do sistema, mas este é compartilhado por duas etapas; (1) atividades gerais e (2) manutenção da rede. A variável "extensão da rede" foi considerada como uma variável intermediária. Como outputs do sistema, considerou-se variáveis "volume comercializado" e "número de consumidores". A partir dessa modelagem, reuniu-se 12 concessionárias de gás canalizado, das quais foram coletados, em seus respectivos relatórios administrativos anuais, os dados necessários sobre as variáveis referentes ao ano de 2017. Finalmente, nenhuma das empresas apresentaram 100% de eficiência o que denota a ineficiência desse setor, pois necessita de estudos mais aprofundados. PALAVRAS CHAVE. Distribuidoras de Gás. NDEA. Inputs Compartilhados. Tópico do artigo (DEA-Análise Envoltória de Dados) ABSTRACT This article aims to evaluate the natural gas distribution companies in Brazil. To obtain this assessment, use the Data Involvement Analysis model to understand how the inefficiencies of the production process in the industry. In addition, as companies of a process subdivided into two stages that require a sharing of inputs. The model presents a variable "operating costs" as an individual input of the system, but it shares by two steps; (1) general activities and (2) maintenance of the network. A variable "network extension" was considered as an intermediate variable. As outputs from the system, variables "marketed volume" and "number of consumers" were considered. From this modeling, bringing together 12 piped gas utilities, the companies were updated, the companies were edited, and the data demanded information for the year 2017. Finally, none of the companies did 100% efficiency.
Research Interests:
RESUMO O presente trabalho tem como objetivo estudar a aplicabilidade do modelo DEA parabólico (Silveira et al., 2018) em cenários de um input e múltiplos outputs. Assim, realizou-se o dimensionamento de equipes de 80 estações... more
RESUMO O presente trabalho tem como objetivo estudar a aplicabilidade do modelo DEA parabólico (Silveira et al., 2018) em cenários de um input e múltiplos outputs. Assim, realizou-se o dimensionamento de equipes de 80 estações ferroviárias da malha eletrificada do Rio de Janeiro. Para isso, foi utilizado como único input a ser redistribuído o número de funcionários de cada estação além de quatro outputs relacionados ao número de linhas atendidas, movimentação de trens, média de passageiros e média máxima de passageiros em horários de pico por estação sendo as três últimas variáveis citadas limitadas a dias úteis. Como a variável funcionários é discreta, foi utilizado complementarmente um algoritmo de redistribuição sequencial de recursos. PALAVRAS CHAVE. DEA; DEA parabólico; DEA paramétrico; Redistribuição de Recursos; Ferrovia. Tópicos: DEA-Análise Envoltória de Dados ABSTRACT This paper aims to study the applicability of parabolic DEA model (Silveira et al., 2018) in one input and multiple outputs scenarios. Thus, teams of 80 railway stations of the Rio de Janeiròs electrified railroad were resized. The number of employees of each station was considered as a single input to be redistributed. Four other variables related to the number of lines served, train movement, average passengers and maximum average passengers at peak times per station were included as outputs. The last three ones limited to working days. As employee is a discrete variable, a sequential resource redistribution algorithm was applied.
Research Interests:
Em 2016, Chaves e outros apresentaram uma formulação dual para o modelo MCDEA de Li e Reeves (1999). No entanto, eles consideraram formulações parciais em duas funções bi-objetivo de cada vez usando o método lexicográfico para resolver o... more
Em 2016, Chaves e outros apresentaram uma formulação dual para o modelo MCDEA de Li e Reeves (1999). No entanto, eles consideraram formulações parciais em duas funções bi-objetivo de cada vez usando o método lexicográfico para resolver o modelo tri-objetivo original de Li e Reeves (1999). O modelo de Chaves el al (2016) apresenta pouca liberdade para os decisores e não consideram todos as características do método de Li e Reeves (1999). Neste artigo tem como objetivo apresentar melhoria para formulação dual com uma abordagem que considera todas as características e aspectos do modelo tri-objetivo de Li e Reeves (1999). Combina-se, pelo método de soma ponderada, as três funções objetivo em uma função única. O modelo dual proposto oferece a busca de benchmarks. Com a função única pela método da soma ponderada pode-se ainda analisar a eficiência para a DMU no método do Li e Reeves. Além disso, o modelo apresentado apresenta maior liberdade ao decisor nas análises. PALAVRAS CHAVE. MCDEA, dual, multiobjetivo. ABSTRACT In 2016, Chaves et al presented a dual formulation for MCDEA from Li and Reeves (1999) model. Nevertheless, they considered just partials formulations for two objective functions each time, using a lexicography method to solve the tri-objective model from Li and Reeves (1999). Chaves et al (2016) model present lower freedom for decision-makers and did not consider whole aspects from Li and Reeves (1999) method. In this paper, we aim to present an improvement in the dual approach considering all aspects and characteristics from tri-objective Li and Reeves (1999) method. We combine, by weighted sum, the three objective functions into single one. We seek to provide a dual model, which is possible to check benchmarks. With the weighted sum method, we can also analyze the efficiency value for the DMU in Li and Reeves (1999) method. Moreover, our approach for formulation brings more freedom for the decision-maker analysis. 1. Introduction Li and Reeves (1999) developed a model using of Data Envelopment Analysis (DEA) associated with Multiobjective Linear Programming to solve recurrent problems in classic DEA models. This intervention aims to increase the discriminatory power of the evaluated units (DMU-decision making unit) and also obtain a more homogeneous distribution in the weights. To solve these problems, Li and Reeves (1999) model appears with the use of a resolution by the multiobjective function that considers additional the criteria of minisum and minimax. In the end, after the resolution procedure, the problem gets few efficient DMUs which becomes valid when the sample of the problem is small (Cooper et al., 2001). In this line, Chaves et al (2016) proposed a dual model for Li and Reeves (1999) formulation to provide benchmarks and targets for the DMU´s. However, this approach considers in pairs the objective functions from Li and Reeves (1999) model by lexicographic method. With this process, Chaves et al (2016) do not aggregate all aspects from the tri-objective Li and Reeves (1999) formulation According to Li & Reeves (1999), all three objective functions are independent and equally important. Herein, we bring the characteristics from Li and Reeves (1999) formulation of efficiency analysis, equality of the sample and overall optimization for a dual approach (Soares de Mello et al, 2009). We use a weighted sum methodology to linearize the tri-objective functions into a single one and then we formulate the dual model. We show in the first part of the article theoretical bases regarding Multiple Criteria Data Envelopment Analysis, then a review for TRIMAP software and MCDEA, following with our dual formulation Li and Reeves (1999) model. Then we use this dual method in a numerical example, finalizing with some conclusions.
Research Interests:
Este trabalho tem como objetivo avaliar as empresas distribuidoras de energia elétrica brasileira com o uso do modelo Networking Data Envelopement Analysis. O gastos das distribuidoras são compartilhados entre os processos administrativos... more
Este trabalho tem como objetivo avaliar as empresas distribuidoras de energia elétrica brasileira com o uso do modelo Networking Data Envelopement Analysis. O gastos das distribuidoras são compartilhados entre os processos administrativos e operacionais, porém esse compartilhamento é ignorado pela reguladora nacional. A partir disso, a modelagem matemática requereu utilizar a configuração de inputs compartilhados para que realidade fosse melhor representada. Com essa nova proposta, o modelo atenuará os problemas vigentes no mercado brasileiro, pois será possível identificar as ineficiências entres os processos restritivos das distribuidoras. Ademais, esse modelo enfatizou sua orientação para o input com a finalidade de redução de custos. O processo de avaliação das distribuidoras de energia elétrica contemplará duas variáveis de insumo (CAPEX; OPEX), duas variáveis intermediárias (perdas não técnicas; extensão da rede em km) e 3 variáveis de produto 1/(DEC*FEC); número de consumidores totais; volume de energia consumido) e. Por fim, o modelo obtiveram 3 empresas eficientes na modelagem devido sua excelência na distribuição o que torna essencial para auxiliar nas tomadas de decisões gerenciais.
Research Interests:
Long-term planning to meet the growth of electricity demand in a country or region is a decision of strategic importance that should aim at maximizing the benefits provided and at minimizing negative impacts on the environment and... more
Long-term planning to meet the growth of electricity demand in a country or region is a decision of strategic importance that should aim at maximizing the benefits provided and at minimizing negative impacts on the environment and society. In this sense, the objective of this study is to develop a methodology to determine the most suitable electric energy matrix of a country, that is, the best combination for the use of electricity generation sources. Thus, a multiobjective linear programming model is proposed to calculate the amount of energy that must be generated by each source available in the country, considering its internal demand and capacity constraints. In addition, this article aims to present the results of a case study in Brazil in the determination of its electric energy matrix, considering the use of several sources of electricity generation. The methodology developed was applied for the years 2015, 2020 and 2030 and generated some important reflections when compared with the actual values and current practices. The proposed model proved to be a useful tool to assist in the analysis and planning of the use, and possible extension, of the generation capacity of each electric power source of a given country.
Research Interests:
The Brazilian air transport system includes more than two thousand airports and landing areas, however only 128 of them are commercially used. This work seeks to analyze the centrality of the Brazilian airport commercial system and... more
The Brazilian air transport system includes more than two thousand airports and landing areas, however only 128 of them
are commercially used. This work seeks to analyze the centrality of the Brazilian airport commercial system and including
operational consequences.
The work analyzes the Brazilian airport system comparing the results of layer centrality (Brand?o et al, 2015) and h-type
centrality (Pereira et al, 2015). These methodologies of centrality, unlike other traditional methods in graphs, try to evaluate
the comprehensiveness of each vertex (in the case of the airport) and the influence on the others in the network.
Considering that only 3 Brazilian airports are on the list of the top 100 airports in the world, the centrality assessment
provides a deep verification of the agents for targeting the improvement resources such as infrastructure. These central
airports require accurate planning in several operational fields (maintenance, traffic control, mobility, etc.) to raise the
quality of the system, especially if we consider current Brazilian economic challenges.
are commercially used. This work seeks to analyze the centrality of the Brazilian airport commercial system and including
operational consequences.
The work analyzes the Brazilian airport system comparing the results of layer centrality (Brand?o et al, 2015) and h-type
centrality (Pereira et al, 2015). These methodologies of centrality, unlike other traditional methods in graphs, try to evaluate
the comprehensiveness of each vertex (in the case of the airport) and the influence on the others in the network.
Considering that only 3 Brazilian airports are on the list of the top 100 airports in the world, the centrality assessment
provides a deep verification of the agents for targeting the improvement resources such as infrastructure. These central
airports require accurate planning in several operational fields (maintenance, traffic control, mobility, etc.) to raise the
quality of the system, especially if we consider current Brazilian economic challenges.
Research Interests:
Macbeth model is a multicriteria tool for the multicriteria decision making that has been used in different projects and studies of renewable energies. The purpose of this paper was apply the software Macbeth for the evaluation of canola,... more
Macbeth model is a multicriteria tool for the multicriteria decision making that has been used in different projects and studies of renewable energies.
The purpose of this paper was apply the software Macbeth for the evaluation of canola, corn, palm and soy crops taking in to account technician,
environmental and economic criterias and their weights. Canola crop was the best option for the biodiesel production, with a global score of 74
points because of their good score in environmental criterias, wich had a higher weight than the other criterias. Corn crop was the second positioned,
presenting good results in all criterias, followed by palm crop. The final one was soy crop with 36,67 points in this analysis
The purpose of this paper was apply the software Macbeth for the evaluation of canola, corn, palm and soy crops taking in to account technician,
environmental and economic criterias and their weights. Canola crop was the best option for the biodiesel production, with a global score of 74
points because of their good score in environmental criterias, wich had a higher weight than the other criterias. Corn crop was the second positioned,
presenting good results in all criterias, followed by palm crop. The final one was soy crop with 36,67 points in this analysis
Research Interests:
Resumo: Este estudo propõe uma metodologia que permite: i) avaliar a distribuição de recursos do Governo Federal aos estados brasileiros; ii) avaliar como estes recursos são utilizados; e iii) propor uma nova distribuição dos recursos.... more
Resumo: Este estudo propõe uma metodologia que permite: i) avaliar a distribuição de recursos do Governo Federal aos estados brasileiros; ii) avaliar como estes recursos são utilizados; e iii) propor uma nova distribuição dos recursos. Para isso, usamos uma metodologia dividida em duas etapas. A primeira consiste no uso de um modelo não-radial DEA BCC com orientação a inputs para avaliação da eficiência, e a segunda propõe uma redistribuição de recursos usando um modelo DEA com ganhos de soma zero. Verificamos que existem discrepâncias entre os valores destinados a cada estado e sua transformação efetiva em ações de desenvolvimento para sua população, de modo que apenas 11 estados foram considerados eficientes inicialmente. No entanto, após a nova distribuição, todos os estados alcançaram 100% de eficiência. Assim, foi possível identificar que o governo não está distribuindo o orçamento de forma eficiente. Palavras-chave: Distribuição de recursos públicos. Análise Envoltória de Dados. Ganhos de Soma Zero.
Research Interests:
In all editions of the Olympic Games, the media creates rankings to evaluate the participation of countries, although there is no official ranking released by the International Olympic Committee. This study proposes an alternative way to... more
In all editions of the Olympic Games, the media creates rankings to evaluate the
participation of countries, although there is no official ranking released by the International
Olympic Committee. This study proposes an alternative way to analyze the performance of
countries. The objective is not to generate a final classification of countries, but to identify the
Olympic powers. As the choice of Olympic powers is subjective and has variation according to the
decision maker, we used the concepts of h-index and h-core. We applied the methodology to the
Olympic cycle comprising the 2016 Summer Olympic Games, in Rio de Janeiro, and the Winter
Olympic Games of Pyeongchang, in the year of 2018. As a result, the Games achieved a h-index
equal to 5 and 5 countries as designated as Olympic powers: United Kingdom, United States,
Germany, China and Russia.
participation of countries, although there is no official ranking released by the International
Olympic Committee. This study proposes an alternative way to analyze the performance of
countries. The objective is not to generate a final classification of countries, but to identify the
Olympic powers. As the choice of Olympic powers is subjective and has variation according to the
decision maker, we used the concepts of h-index and h-core. We applied the methodology to the
Olympic cycle comprising the 2016 Summer Olympic Games, in Rio de Janeiro, and the Winter
Olympic Games of Pyeongchang, in the year of 2018. As a result, the Games achieved a h-index
equal to 5 and 5 countries as designated as Olympic powers: United Kingdom, United States,
Germany, China and Russia.
Research Interests: Olympic Games and H Index
The cross evaluation technique in Data Envelopment Analysis (DEA) was developed based on the model with constant returns of scale, DEA CRS (Constant Returns to Scale). However, in many applications, it is not possible to guarantee the... more
The cross evaluation technique in Data Envelopment Analysis (DEA) was developed based on the
model with constant returns of scale, DEA CRS (Constant Returns to Scale). However, in many
applications, it is not possible to guarantee the proportionality between inputs and outputs, and this
may limit the use of the cross evaluation model. Due to this, it is proposed in this paper the use of
the variable returns model, DEA VRS (Variable Returns to Scale) oriented to input, together with
the cross evaluation model. Nonetheless, it is possible to find lower efficiencies when using DEA
VRS and, in extreme cases, to find negative efficiencies. In order to address this problem, solutions
suggested in the literature are presented and a new approach is proposed as well. Considering that
one of the characteristics of the DEA VRS model is to be more benevolent with the decisionmaking
unit (DMU) than the DEA CRS model, the matrices of the cross-evaluation of the two
models will be compared in order to generate a final matrix. In this study, the Brazilian electricity
sector will be used to apply the presented approach.
model with constant returns of scale, DEA CRS (Constant Returns to Scale). However, in many
applications, it is not possible to guarantee the proportionality between inputs and outputs, and this
may limit the use of the cross evaluation model. Due to this, it is proposed in this paper the use of
the variable returns model, DEA VRS (Variable Returns to Scale) oriented to input, together with
the cross evaluation model. Nonetheless, it is possible to find lower efficiencies when using DEA
VRS and, in extreme cases, to find negative efficiencies. In order to address this problem, solutions
suggested in the literature are presented and a new approach is proposed as well. Considering that
one of the characteristics of the DEA VRS model is to be more benevolent with the decisionmaking
unit (DMU) than the DEA CRS model, the matrices of the cross-evaluation of the two
models will be compared in order to generate a final matrix. In this study, the Brazilian electricity
sector will be used to apply the presented approach.
Research Interests:
Research Interests:
Research Interests:
Despite standard Data Envelopment Analysis (DEA) models having been widely used in assessing efficiency in education, a few studies use Network DEA (NDEA) models in educational evaluation. In this paper, we proposed an alternative to the... more
Despite standard Data Envelopment Analysis (DEA) models having been widely used in assessing efficiency in education, a few studies use Network DEA (NDEA) models in educational evaluation. In this paper, we proposed an alternative to the official evaluation performed every three years by CAPES (Brazilian agency for post-graduation programs regulation) using an NDEA model. The use of NDEA is justified because depending on the point of view, some variables can be considered either inputs or outputs. The use of NDEA avoids the need to decide whether a variable is an input or output of the entire process. This happens because a variable can be both an output for one stage and an input for another. Our relational NDEA model computes both productivity and quality assessments together with global efficiency using bibliometric data. Evaluación de los programas de post-grado usando un modelo de análisis envolvente de datos en red Resumen A pesar de que los modelos estándar del Análisis Envolvente de Datos (DEA) han sido ampliamente utilizados en la evaluación de la eficiencia en educación, existen pocos estudios que utilizan modelos DEA en red (Network DEA – NDEA) en la evaluación educativa. En el presente trabajo, se ha propuesto una alternativa a la evaluación oficial realizada a cada tres años por la CAPES (agencia brasileña para la regulación de los programas de post-graduación) mediante un modelo DEA en red. El uso de NDEA se justifica ya que dependiendo del punto de vista algunas variables pueden ser consideradas como entradas o como salidas. El uso de NDEA evita la necesidad de decidir si una variable es una entrada o una salida de todo el proceso. Esto ocurre porque una variable puede ser tanto una salida para una etapa y una entrada para otro. Nuestro modelo relacional NDEA evalúa tanto la productividad como la calidad junto con la eficiencia global, a partir de datos bibliométricos. Palabras clave: educación; análisis envolvente de datos; DEA en redes; eficiencia educacional; evaluación bibliométrica.
Research Interests:
Resumo A avaliação da produção acadêmica é um tópico de grande relevância, pois é um dos critérios levados em conta no desenvolvimento de políticas de fomento pelo Ministério da Educação. Um dos índices mais populares para a quantificação... more
Resumo A avaliação da produção acadêmica é um tópico de grande relevância, pois é um dos critérios levados em conta no desenvolvimento de políticas de fomento pelo Ministério da Educação. Um dos índices mais populares para a quantificação da produção científica é o índice h, que contabiliza as citações recebidas pelos artigos de determinado pesquisador, e é capaz de resumir sua história de produção científica em um número. Ao mesmo tempo, isso pode ser entendido como uma limitação, porque descarta muitos detalhes do registro de citação, por exemplo, o tempo de vida acadêmica do pesquisador. Neste artigo é proposta uma nova metodologia para avaliação da produção docente através do cálculo da eficiência de cada pesquisador utilizando um modelo DEA, com tempo de vida acadêmica como input e índice h como output. Constatou-se que esta metodologia é mais adequada para este tipo de avaliação, principalmente quando comparada ao índice m-quociente, uma vez que considera retornos variáveis de escala. Palavras-chave: Avaliação da Produção Docente. Análise Envoltória de Dados. Índice h.