PPG - Pesquisa Operacional

URI Permanente para esta coleção

https://hdl.handle.net/11600/61102

Navegar

Agora exibindo 1 - 20 de 53

Acesso aberto (Open Access)
Uma Math-Heurística Para O Problema De Dimensionamento De Lotes Com Máquinas Paralelas E Setup Carry-Over
(Universidade Federal de São Paulo (UNIFESP), 2018-02-09) Fernandes, Carla Cristina Doescher [UNIFESP]; Nascimento, Maria Cristina Vasconcelos [UNIFESP]; Universidade Federal de São Paulo (UNIFESP)
This Master’s Dissertation deals with the Lot Sizing Problem (LSP) in Parallel Machines with multiple items are considered and different machines that produce the same items and that have capacity restrictions. The items can be produced in any machine and at the beginning of the production of each item the time and cost of setup of the used machine is incurred. In addition, it is considered the possibility of taking advantage of setup at the end of each period: the setup carry-over, for cost reduction and, implicitly, for the definition of a production planning that reduces possible environmental impacts caused by the preparation of machines. Despite several studies in the literature for LSP in parallel machines, few of them introduce the efficient solution methods to solve it considering the setup carry-over. In this sense, to solve this problem efficiently, in this Dissertation, it is proposed its study from the point of view of modeling and efficient solution method. First, three mixed integer programming models were adapted for the LSP in Parallel Machines. After experiments, two models were defined that were used for the proposed solution method based on modeling (math-heuristics). The math-heuristic based on the well-known heuristic relax-and-fix with local branching constraints was modeled whose structure was inspired by the Greedy Randomized Search Procedures (GRASP). Experiments, in which the solutions obtained by the CPLEX solver and the math-heuristic are compared, demonstrated the efficiency of the proposed method.
Acesso aberto (Open Access)
Metaheuristics With Random Keys And Local Search For The Vehicle Routing Problem With Private Fleet And Common Carrier
(Universidade Federal de São Paulo (UNIFESP), 2018-02-23) Higino, William [UNIFESP]; Chaves, Antonio Augusto [UNIFESP]; Universidade Federal de São Paulo (UNIFESP)
The Vehicle Routing Problems (VRPs) have been target of a high number of studies in the Operational Research area, given its applicability on several fields. Among its categories are the Vehicle Routing Problems with Profits. Those problems are characterized by the lack of obligatoriness in the service of all customers. Instead, a profit or prejudice rate to the service of each customer is defined. This category presents the Vehicle Routing Problem with Private Fleet and Common Carrier (VRPPFCC). In this problem, besides the traditional vehicle routing to serve customers, considering demand and capacity, there is the possibility of outsourcing partly the service, considering the profitability in such process. This study applies two meta-heuristics based on random keys, Biased Random Keys Genetic Algorithm (BRKGA) and Unified Marginal Distribution Algorithm (UMDA) on the solution of the VRPPFCC. It also combines such meta-heuristics with variations of Random Variable Neighborhood Descent (RVND), Self-Adaptive Variable Neighborhood Descent (SAVND), and additional conceived local search methods, in order to further explore the search space. Aiming to make a better use of computational resources in local searches, the Clustering Search (CS) hybrid method is used, seeking to improve the obtained solutions quality by managing the application of the local search procedure, evaluating promising regions of the search space. Computational tests are performed with available instances in the literature, and the method results and behaviors are compared. Finally, conclusions are made based on the achieved results
Acesso aberto (Open Access)
Abordagem Multicritério Para A Classificação De Áreas Para Doenças Transmitidas Pelo Mosquito Aedes Aegypti
(Universidade Federal de São Paulo (UNIFESP), 2018-03-02) Aguiar, Emerson Santos [UNIFESP]; Belderrain, Mischel Carmen Neyra [UNIFESP]; Universidade Federal de São Paulo (UNIFESP)
In face the current scenario of outbreaks and epidemics of Zika, Chikungunya and Dengueâs outlined by the spread of these viruses around the world, recognizing the epidemic at the outset, through risk or vulnerability assessment, is essential for the planning of mitigation and control actions over the impact of these diseases on society. As follows, the objective of this work was to assist the epidemiological surveillance of diseases associated with the Aedes aegypti mosquito in society by constructing maps of vulnerability from a multicriteria spatial decision model. Initially, a bibliographic study was designed to identify the determinants of the vulnerability of areas to Zika, Chikungunya and Dengue; these factors were used as a basis for structuring the criteria used in the multicriteria model. Given the importance that the evaluated areas presented satisfactory and balanced performances in most of the criteria, the use of the non-compensatory multicriteria method PROMETHEE II was considered to construct the map of vulnerability. The proposed approach was applied in a case study to classify neighborhoods of a city in northeastern Brazil. The response provided by the approach was presented spatially on thematic maps created in a geographic information system. The results suggest that the proposed model is useful as a decision support tool in the epidemiological surveillance of Aedes aegypti-related diseases, allowing a definition of the vulnerability of areas related to Aedes aegypti, a more effective definition of the vulnerability of areas to diseases transmitted by Aedes aegypti, as well as allowing assistance in a more efficient decision-making process on disease mitigation and control measures. The model supports the process from the investigation of the factors determining the vulnerability of the areas, assists in the evaluation of the health conditions of the population through the performance of the districts, both individually and in an aggregated way, and, finally, helps in the design and evaluation of actions to prevent and control diseases.
Acesso aberto (Open Access)
Uso De Malhas Triangulares Em Métodos De Programação Linear Sequencial Aplicados À Otimização Topológica
(Universidade Federal de São Paulo (UNIFESP), 2018-07-30) Aguiar, Marila Torres De [UNIFESP]; Bueno, Luis Felipe Cesar Da Rocha [UNIFESP]; Universidade Federal de São Paulo (UNIFESP)
Problems of Topological Optimization aim to obtain an ideal configuration of a structure, taking into consideration aspects such as stiffness, displacements, intensity and position of force applications and material availability. Usually, these problems are formulated as large nonlinear optimization problems. In this work, we present research possibilities related to the efficiency of the methods of the Sequential Linear Programming types and in Topological Optimization problems with irregular domain. The main contribution is the adaptation of the proposed code in (SENNE, 2009) to deal with regular and irregular triangular finite elements. For this adaptation, the whole theoretical part of the problem modeling is redone and numerical tests are performed to validate the developed algorithm and analyze its efficiency. In addition, a bibliographical review on the use of Topological Optimization in the aerospace sector is presented and the resolution of a problem related to the optimal filling of an aircraft wing is presented.
Somente Metadadados
Modelo Multicritério De Apoio À Decisão Para O Problema De Alocação De Berços Dinâmico
(Universidade Federal de São Paulo (UNIFESP), 2018-11-09) Gandra, Stefane Rego [UNIFESP]; Azevedo, Anibal Tavares De [UNIFESP]; Universidade Federal de São Paulo (UNIFESP)
For the purpose of a reduction of logistics bottlenecks, to support decision making and to improve the efficiency of port terminals, the aim of this project is to build a methodology that may contribute to the solution of the discrete dynamic berth allocation problem (DDBAP). At first, the purporse of the research was to identify the difficulties and to raise possibilities for improvements for the DDBAP. Dispite of the diversity of studies about this subject, it was verified that until this moment there was no methodology that could evaluate if there is a soluction to a DDBAP when it is applied to other objective functions. Also, a methodology that can consider the relavance of the opinion e the expertise of the decision maker during the process of modelling and solving the problem. In this sense, we first selected the most pertinent objective functions in the literature regarding DDBAP and so, considering a small case study, these models were implemented in CPLEX. Posteriorly, genelarized mathematical formulation was proposed, where the objective functions are conditioned to the same constraints. Then, by obtaining the optimum solution of each model (vessel sequencing), the evaluation was performed in terms of the other objective functions. As an alternative to evaluate the results for each solution, the proposed methodology was added the PROMETHEE II multicriteria method to assist the prioritization of the alternative that has the best evaluation in relation to all the objective functions. The results showed the importancy of the use of the method to resolve and aid the decision making to the problem presented.
Acesso aberto (Open Access)
Uma Aplicação De Meta-Aprendizagem Nas Cotações Euro/Dólar
(Universidade Federal de São Paulo (UNIFESP), 2018-12-07) Duarte, Felipe De Almeida [UNIFESP]; Sato, Renato Cesar [UNIFESP]; Universidade Federal de São Paulo (UNIFESP)
For the investor, knowing when to buy or sell a asset is a decision of extreme importance to make profits in financial markets. However, the same decision-making process, the investor should choose the ideal model for the study and analysis of the data series worked. Due to the large number of models available, choosing an ideal model often makes the task difficult, especially for inexperienced investors. To assist in such a decision, meta-learning can be an ideal tool, by making suggestions of models through applications in similar past data. This tool has already been used for the problem of algorithm selection and presents good results in the selection of study models of time series. In this work, we seek to use the method of analysis and analysis of time series (concrete euro/dollar quotations), through a ranking algorithm that suggests the best models. The results were satisfactory, presenting good predictions for the suggested models.
Acesso aberto (Open Access)
NSGA-II para a solução eficiente de problemas de roteamento multiobjetivo em redes de sensores sem fio
(Universidade Federal de São Paulo (UNIFESP), 2019-01-28) Jeske, Marlon [UNIFESP]; Rosset, Maria Cristina Vasconcelos Nascimento [UNIFESP]; Universidade Federal de São Paulo (UNIFESP)
In the last years it has been observed an increase in the use of applications and new technologies for wireless sensor networks (WSNs), which have the function of performing monitoring tasks in different environments. In this way, WSNs allow information from the physical environment to be connected to the internet, making them an essential part of the Internet of Things (IoT) concept. A WSN is composed basically of low-cost micro- devices with limited energy (namely sensors), able to collect, transmit and receive data. It is of fundamental importance to find ways and solutions that correspond to the intrin- sic needs related to the technological limitations of the network components, such as the maximization of the network energy efficiency (lifetime), minimization of the packet loss rate, maximization of connectivity coverage, among others characteristics called metrics of QoS (Quality of Service). This Master Thesis addresses a bi-objective routing problem in WSN recently proposed in the literature, which has as optimization criteria two con- flicting metrics of QoS, the most emphasized in the literature: residual energy efficiency and packet delivery reliability. The heuristic approach employed in the literature was not able to obtain results for large-scale environments. In addition, for small and large-scale environments, the literature heuristics provided slightly dense Pareto curves, making it difficult to evaluate the results. In this way, to solve the problem in this Thesis was deve- loped the multiobjective evolutionary algorithm Elitist Non-dominated Sorting Algorithm (NSGA-II). Simulation results show that the solution of the problem using NSGA-II has better efficiency in terms of solution quality and computational time when compared to literature heuristics and exact approaches. Besides that, due to the good performance obtained in small-scale environments through the evolutionary approach, it was possible to solve the bi-objective problem in large-scale environments with solutions in a viable computational time.
Acesso aberto (Open Access)
Strategic Options Development and Analiysis (SODA): a survey of fields of application (1989-2018)
(Universidade Federal de São Paulo (UNIFESP), 2019-05-27) Abuabara, Leila [UNIFESP]; Belderrain, Mischel Carmen Neyra [UNIFESP]; Universidade Federal de São Paulo (UNIFESP)
Strategic Options Development and Analysis (SODA) is a well-established problem structuring method (PSM) that has been addressing problematic situation of shared concern for at least 30 years within the discipline of Operational Research (OR). The aim of this study is to assess the ways OR practitioners and academics have been implementing SODA methodology in different fields of knowledge and practice. Hence, we will consider studies associated with the practical application of SODA published from 1989 (the publication date of the first collection of PSMs, the book Rational Analysis for a Problematic World) up to 2018. We will examine the level of each application, whether as a purely SODA application, or as a hybrid. Within these contexts, we will also investigate which elements of the methodology have been used. Our findings suggest that SODA methodology through its associated technique of cognitive mapping has been used especially to improve other methods through an integrated use. Our findings also confirm SODA as an originally participative methodology that provides for dialogue, reflection, learning, consensus and commitment. SODA seems to be adaptable to different contexts, approaches, situations and cultures and, in whole or in part, its use has grown over time. In addition, its main mission has been limited to modeling the problematic situation and providing a common understanding of it to participants, leaving aside the group negotiation support.
Acesso aberto (Open Access)
Uma Análise do Índice Geral de Cursos usando DEA
(Universidade Federal de São Paulo (UNIFESP), 2019-07-05) Favali, Joao Danilo [UNIFESP]; Milioni, Armando Zeferino [UNIFESP]; Universidade Federal de São Paulo (UNIFESP)
The National Institute for Educational Studies and Research “An ́ısio Teixeira” (Inep), an autarchy linked to the Ministry of Education (MEC), is responsible for providing information and statistics for the definition of educational policies. Within the context of higher education, there is the National System of Evaluation of Higher Education (Sinaes), which aims to evaluate institutions, courses and student performance. One of the informative tools of Sinaes is the General Courses Index (IGC), an indicator that seeks to measure the quality of each Institute of Higher Education (IES). The IGC synthesizes in a single value the average grades of undergraduate, masters and doctoral of the referred IES and is used, at least by the media, to rank them. Nevertheless, the methodology for obtaining the IGC carries an apparent paradox: it is possible for an IES to have undergraduate, masters and doctoral grades respectively higher than those of another IES, but with a lower IGC than the last one. This fact is accentuated by the methodology does not present theoretical justifications for the construction of the IGC, especially regarding the weighting factors and conversion of the grades. The purpose of this present work, therefore, is to use Data Envelopment Analysis (DEA) on the data of the grades - in a first moment, these grades shall be the provided by Inep, later they will be normalized - in order to propose alternative rankings. The traditional DEA-CCR model and the cross efficiency approach were used, with two variations of secondary objectives, that revealed different rankings than the IGC and did not present the mentioned paradox.
Acesso aberto (Open Access)
Feature selection for characterization of continuous optimization functions
(Universidade Federal de São Paulo (UNIFESP), 2019-08-27) Silva, Guilherme Ribeiro Da [UNIFESP]; Scarpel, Rodrigo Arnaldo [UNIFESP]; Universidade Federal de São Paulo (UNIFESP)
The field of automatic algorithm selection has received increased attention in the past years. Machine Learning techniques are now able to predict with high accuracy the best set of algorithms for a given problem instance. In this present dissertation, we establish the steps for the selection of features for characterization of continuous functions. Via a Dimensionality Reduction approach, only the features that best preserve the information of the dataset are selected. Then, the selected features were compared to already established sets of metrics from the literature and the results are analyzed.
Acesso aberto (Open Access)
Aplicação de algoritmos não supervisionados em dados eleitorais
(Universidade Federal de São Paulo (UNIFESP), 2019-12-03) Polizeli, Mateus Vendramini [UNIFESP]; Bueno, Luis Felipe Cesar Da Rocha [UNIFESP]; Universidade Federal de São Paulo (UNIFESP)
Given the incessant search of society for clarity in government spending, management efficiency and transparency using the public agency, the structuring of works that allow a thorough investigation to efficiently monitor these actions becomes relevant. From an initial study in the literature, it was verified the existence of a series of controls and disclosure of accountability of sectors and public agencies. However, despite initiatives such as these, there is still little work considering further investigation to capture possible irregularities in the policy instrument. Thus, the objective of this project is to study some mechanisms for detecting anomalies associated with the 2018 electoral candidate data set. The proposed methodologies are based on unsupervised algorithms K-Means and Isolation Forest in an attempt to create a decision support tool for regulators to direct human resources for research. A combination of these algorithms, referred to here as KM+IF, is also suggested in order to improve accuracy and decrease the error rates associated with the models. The results observed in this project indicate that the proposal KM+IF shows good performance for situations where the variables of interest are available. However, it may yield unsatisfactory results when they are not available. In the case study for the set of electoral candidates, the overall result of the KM+IF algorithm was lower than the individual result of the K-Means and Isolation Forest techniques.
Acesso aberto (Open Access)
Antecipação de Mudança de Regimes na Fatia Diária de Voos Atrasados e Cancelados no Aeroporto Internacional de Guarulhos
(Universidade Federal de São Paulo (UNIFESP), 2019-12-09) Teixeira, Rosana Batista [UNIFESP]; Scarpel, Rodrigo Arnaldo [UNIFESP]; Universidade Federal de São Paulo (UNIFESP)
Flight delays and cancellations are frequent occurrences in most airports around the world. In Brazil the deregulated increase in air traffic caused flight concentration in some airports, enabling the occurrence of delays and cancellations due to congested days. The Guarulhos International Airport is the most affected by delays. Therefore, the goal of this work is to anticipate the occurrence of congested days at Guarulhos International Airport employing clustering and classification approaches to identify a regime change in the daily share of delayed and canceled flights. The built model is composed of a Hidden Markov Models as a clustering approach and the classification methods Classification and Regression Tree, Random Forest, and Support Vector Machine. The accuracy of the prediction model was considered satisfactory, and it anticipates the regime change in a daily share for one period ahead.
Acesso aberto (Open Access)
Objetivos Baseados em Valores na Definição de Ações do SSM: Estudo de Caso na Inclusão de Estudantes NEE
(Universidade Federal de São Paulo (UNIFESP), 2019-12-10) Francozo, Rafael Verao [UNIFESP]; Belderrain, Mischel Carmen Neyra [UNIFESP]; Universidade Federal de São Paulo (UNIFESP)
Real world problems are complex, and this complexity translates into the absence of elements that allow us to define objectives, constraints, and resources, in addition to, what the actual problem to be considered is. Such problems may require broad changes in underlying policies, assumptions, and objectives, making it necessary to define clearly the real problems to be addressed and the true objectives to be reached. Education for students with special educational needs is a complex problem with context-dependent dynamics that are difficult to address. This study developed an approach to identify and define undesirable states in a given problem situation. In addition, the designed models led to a more desirable state in accordance with the objectives of those involved. To achieve this objective, an approach based on problem structuring methods was built to: 1) Focus on the context of a problematic situation under consideration; 2) Define objectives based on stakeholder values; and 3) Steer the situation towards change in line with the defined objectives. The constructed approach was applied in a teaching unit to address the issue of education for students with special educational needs. The application of the approach resulted in benefits such as the clear definition of how education for these students is understood in the institution, with special attention to the target audience regardless of clinical report. It also helped to identify the real problems and goals, by deconstructing established concepts. This study contributes to the specialized literature on Operational Research by guiding systemic planning for objectives based on stakeholder values. This contribution helps to set limits, guide, and prioritize the planning of necessary changes. As a social contribution, the study provides a structured means to identify the necessary practices for the effective inclusion of students with special educational needs, which helps to make society more inclusive, humane, and supportive.
Acesso aberto (Open Access)
Seleção Otimizada de Enlaces Instáveis em Redes de Comunicação ATN/IPS
(Universidade Federal de São Paulo (UNIFESP), 2019-12-13) Sernagiotto, Marco Aurelio [UNIFESP]; Rosset, Maria Cristina Vasconcelos Nascimento [UNIFESP]; Universidade Federal de São Paulo (UNIFESP)
The aeronautical communications in the new concept of ATN-BR (Aeronautical Telecommunication Network -- Brazil) bring a new communication model, exclusively digital. As a result, network management becomes essential to ensure the continuity of communications. In this context, the heterogeneous network selection techniques cited in the literature are a way to perform the datalink selection based on criteria that meet the requirements of communication services and users of this aeronautical network. The aim of this document is to present an integrated optimization model to select aeronautical network links, maximizing network usage while minimizing the difference in capacity utilization between datalinks (load balancing). To achieve this goal, multicriteria evaluation techniques are combined with a multiobjective optimization method. For the analysis of subjective criteria the AHP method (Analytic Hierarchy Process) is used and for the objective criteria, the utility functions are used. The criteria are combined with the SAW (Simple Additive Weighting) compensatory multicriteria method to evaluate possible datalink-packet combinations. And ɛ-Restricted multiobjective optimization provides Pareto-optimal boundary values as the balancing levels (ɛ) are set for the network. The datalink selection problem is approached through a case study, with data generated based on the real ATN-Br data. The results obtained in this case study indicate that there is a trade-off value that maximizes profit use and minimizes the difference in link usage (difference between available bandwidth used rates) without causing an increase in packet loss. Thus, the presented solution offers network managers a reference for optimized operation of ATN-Br, ensuring the continuity of the most critical (priority) communications, even in scenarios of link instability.
Acesso aberto (Open Access)
Novas formulações de fluxo para problemas de otimização combinatória
(Universidade Federal de São Paulo, 2020-02-18) Silva, Tiago Tiburcio da; Chaves, Antônio Augusto; Yanasse, Horacio Hideki; http://lattes.cnpq.br/6187221670775160; http://lattes.cnpq.br/4973949421738244; http://lattes.cnpq.br/1675935201364161
Neste trabalho aborda-se o Problema de Minimização de Trocas de Ferramentas (PMTF) e o Problema do Caixeiro Viajante Multiproduto com Prioridades (PCVMP). O PMTF consiste em determinar um sequenciamento de tarefas, de tal modo que a quantidade de trocas de ferramentas entre as tarefas seja a menor possível. Cada tarefa requer um conjunto de ferramentas distinto, e supõe-se que cada um destes conjuntos não contenha mais ferramentas do que suporta a máquina. Já o PCVMP consiste em determinar uma rota de entrega de mercadorias considerando ao mesmo tempo, o cliente e o vendedor, ou seja, minimizando os custos totais do vendedor e maximizando as preferências dos clientes. Neste estudo tem-se como objetivo modelar, baseado em fluxo multicommodity, os problemas citados. Modelos matemáticos de otimização foram propostos assim como alguns resultados teóricos foram desenvolvidos. No caso do PMTF, o melhor modelo proposto foi comparado com os modelos existentes na literatura, mostrando um melhor desempenho tanto em quantidade de instâncias resolvidas na otimalidade, quanto no valor da relaxação linear e no tempo de execução. Mostrou-se que o valor da relaxação linear nos modelos propostos corresponde a diferença entre a quantidade de ferramentas e a capacidade da máquina. Algumas matheurísticas baseadas em busca por proximidade e um método exato enumerativo considerando eliminação de simetria foram propostos e comparados com os resultados da literatura. Já no caso do PCVMP, o modelo proposto se mostrou eficiente em resolver instâncias de pequeno e médio porte. Duas metaheurísticas, BRKGA e BRKGA adaptativo, ambas com busca local, também foram propostas para o PCVMP, apresentando bons resultados.
Acesso aberto (Open Access)
Sistema integrado de manutenção produtiva a partir de otimização para contextos industriais de pequeno e médio porte
(Universidade Federal de São Paulo (UNIFESP), 2020-02-19) Picanco, Ailson Renan Santos [UNIFESP]; Salles Neto, Luiz Leduino De [UNIFESP]; Universidade Federal de São Paulo
Industrial maintenance is the operational management of production process equipment that aims to ensure that machines operate as required, minimizing damage from unscheduled downtime. Observing the demands of the market and the objects of study in this area, a dilemma was found between investigations and research into increasingly complex problems and; the need for simplified solutions with low cost and easy implementation by organizations. From this gap, the objective of the work is to propose an integrated maintenance system based on the construction of optimization models under uncertainty and multicriteria decision analysis, oriented to small and medium industrial contexts. The research is based on the need to integrate optimization models for maintenance aimed at small and medium industries, as it understands that businesses of this dimension are relevant to the economy in terms of the contribution of GDP and in the generation of jobs, but that present demands for increased productivity and process development through management methods and models. In this sense, the system seeks to fill gaps in the choice of type of equipment maintenance, construction of the sequencing of activities, dimensioning of spare parts and component analysis based on reliability, with a focus on optimizing costs and the availability of productive resources. The implementation capacity, that is, technical and economic feasibility, of applying the models are fundamental boundary conditions, which guide the modeling process. Models were built to determine the global criticism for choosing the equipment allocated for preventive or corrective management, through a multicriteria approach combining cognitive map for extracting the criteria, TAW for the attribution of weights and clustering; and TOPSIS for the criticality ranking. A model for minimizing expected costs was built from the preventive maintenance schedule, as well as a cost minimization model for the corrective management of spare items. They were applied in two exploratory case studies developed in an intermodal sugar terminal and in a construction and machinery rental company, whose results point to a reduction in expected costs of 20 to 35%. Finally, as a facilitating measure for the application process, an implementation protocol was developed, punctuating the collection procedures, evaluation of variables and decisions, as well as analysis of risks and potential improvements.
Acesso aberto (Open Access)
Método para agregação de Tradeoff interativo e flexível para decisão em grupo
(Universidade Federal de São Paulo (UNIFESP), 2020-11-30) Mondadori, Jorge Augusto Pessatto [UNIFESP]; Belderrain, Mischel Carmen Neyra [UNIFESP]; Universidade Federal de São Paulo
This work is based on the need to develop a flexible and interactive method for group decision that considers the strength of preference of each decision maker during the process of evaluating scale constants, within the context of a multicriteria decision, without necessarily considering the weight of each decision maker during the process. In order to achieve these objectives, the concept of flexible and interactive Tradeoff, which starts from the Theory of Multi-attribute Value, was used and helps individuals to assign values for scale constants in a multicriteria model with reduced cognitive effort. To assist in this process, considering multiple decision makers, a hybrid method was applied between FITradeoff and the construction of second-order preference functions. Such method allows decision makers to continuously interact with the flexible Tradeoff procedure, while at each choice cycle, the heuristic is adjusted based on the strength of preference for each decision maker, in each choice cycle. This hybrid method allows decision makers to aggregate judgment while making individual choices, and not just adding the final preference and its vectors or individual weight spaces. In addition to structuring the elicitation model for scale constants, the hybrid method allows the investigation of the construction of second order preference functions in an additive aggregation model within the context of multicriteria decision. As a final contribution, there is the structured method, with an execution algorithm, considering the union of these functionalities. An example is implemented, in which decision makers were able to participate in the process, interacting with the previously structured model, converging to a single choice decision at the end.
Acesso aberto (Open Access)
Otimização sob incerteza: um novo método computacional, uma nova medida de robustez e aplicações
(Universidade Federal de São Paulo, 2021-03-10) Butkeraites, Renan Brito Cano [UNIFESP]; Salles Neto, Luiz Leduino de [UNIFESP]; Lodwick, Weldon A.; http://lattes.cnpq.br/6568555460229412; http://lattes.cnpq.br/3728820959678712; http://lattes.cnpq.br/3494006054065503
A abordagem mais popular atualmente para ajudar o tomador de decisão durante o tratamento da incerteza em problemas de otimização linear e inteira tem o revés de necessitar de um parâmetro de controle de incerteza que nem sempre é de trivial escolha. Este trabalho apresenta três avanços nesta área: 1) Desenvolvimento de uma abordagem para resolver problemas de otimização sob incerteza, aplicando o método desenvolvido em um problema de otimização linear da Indústria 4.0 (empacotamento de largura de banda) e comparar com uma abordagem da literatura; 2) Modelagem e resolução de um problema não linear de localização 2D de alvos sob incerteza utilizando geometria de distância, comparando com uma abordagem da literatura e; 3) Criação de uma medida de qualidade baseada em factibilidade e validar em um problema de otimização linear clássico da literatura. O método de resolução de problemas de otimização sob incerteza proposto nesta tese, chamado de SIROM (Sampling-based multi-objective Iterative Robust Optimization Method), foi arquitetado para encontrar automaticamente soluções para problemas de otimização sob incerteza utilizando as informações obtidas durante um processo de simulação e agrupamento de soluções por semelhança. Sua aplicação ao problema de largura de banda obteve em 92,5% das instâncias resultados melhores ou iguais aos resultados da aplicação no método da literatura para o mesmo problema. A modelagem do problema de localização 2D de alvos considerou uma definição diferente do tipo de incerteza usada no modelo clássico da literatura e sua resolução encontrou menor erro de localização em 59% dos casos, além de menor tempo de execução. Além disso, foi criada uma nova medida de robustez, chamada Hardness, que considera a factibilidade de uma solução para qualquer realização dos parâmetros incertos, que, por considerar a diferença do grau de robustez entre as soluções, demonstrou ser mais clara do que as medidas de qualidade da literatura.
Acesso aberto (Open Access)
Seleção e alocação de ações baseadas em análise fundamentalista e técnica e MCDA
(Universidade Federal de São Paulo, 2021-06-16) Salvadori, Bruno Lopes [UNIFESP]; Belderrain, Mischel Carmen Neyra; http://lattes.cnpq.br/ 3574552458175171; http://lattes.cnpq.br/9268285242645478
O interesse da população brasileira na área financeira vem crescendo nos últimos anos, esse aumento pode ser evidenciado pelo aumento de 400% do número de pessoas físicas que começaram a investir na bolsa de valores de São Paulo nos 10 últimos anos. Mohanram (2005) afirma que os analistas de mercado possuem muitas estimativas ingênuas, o que afeta o fator de sucesso de uma estratégia de investimento. Assim, constata-se que há necessidade de novos estudos e novas técnicas que auxiliem os investidores em suas tomadas de decisões. Exposta essa necessidade, o presente trabalho propõe um método de elaboração de uma carteira de ações e seu gerenciamento utilizando métodos multicritério de apoio à decisão, sendo esses métodos pautados por indicadores fundamentalistas e técnicos além do uso da teoria de carteira de Markowitz. O método proposto foi aplicado em um período de 12 meses entre 2019 e 2020 acumulando resultados superiores ao Ibovespa e a retornos oferecidos por fundos de investimentos existentes no mercado.
Acesso aberto (Open Access)
Nova abordagem no Método de Análise e Solução de Problemas (MASP)
(Universidade Federal de São Paulo, 2021-07-05) Salvadori, Thais Silvério [UNIFESP]; Belderrain, Mischel Carmen Neyra; http://lattes.cnpq.br/ 3574552458175171; http://lattes.cnpq.br/4999109403480044
A competição acirrada do mercado faz com que empresas busquem continuamente o aprimoramento de seus processos, produtos e/ou serviços, com o intuito de reduzir perdas e retrabalhos, garantindo altos padrões de custo-benefício, qualidade e prazos. Contudo, a identificação do problema, análise e solução é um processo bastante complexo, no qual necessita de uma equipe multidisciplinar e um processo eficiente para uma solução eficaz. Assim, o objetivo desse estudo é agregar ao método de análise e solução de problemas (MASP) uma etapa mais robusta de estruturação e utilizar um método multicritérios para priorização das ações. O método proposto utiliza o método MASP (Método de Análise e Solução de Problemas) e suas respectivas ferramentas da qualidade com foco para análise e solução do problema e; faz a introdução de métodos da área de Pesquisa Operacional para melhor identificação e estruturação de um problema complexo, por meio do método PFM (Problem Focused Method), além do método AHP (Analytic Hierarchy Process) para a priorização das ações. O método proposto foi aplicado em uma unidade fabril durante a implementação de um novo processo, na qual a implementação durou mais de dois anos. No início das produções havia um alto nível de rejeição de lotes devido a defeitos associados ao produto, na qual a conformidade do lote baseia-se em uma percentagem máxima de defeitos por lote. O defeito estudado neste trabalho teve como alvo uma aceitação máxima de 1,00%. Com o método proposto conseguiu-se alcançar essa uma taxa menor que taxa máxima de defeito por lote no processo produtivo.

Navegar

Navegando PPG - Pesquisa Operacional por Data de Publicação

Resultados por página

Opções de Ordenação