Publicações

2016
Marina Soares Almeida, Mônica do Amaral, and Reinaldo Morabito. 2016. “Um estudo sobre localização de terminais intermodais na rede de escoamento da soja brasileira.” Production, 26, 3, Pp. 562-580. Publisher's VersionAbstract

A soja é um dos principais produtos agrícolas brasileiros. Do total colhido, aproximadamente 40% é destinado à exportação. Como as lavouras estão concentradas principalmente na região central no país, essa soja percorre, em geral, grandes distâncias até os portos marítimos. Nesse caso, o transporte intermodal pode tornar mais eficiente e econômica a movimentação de grandes volumes por longas distâncias. Analisa-se, neste estudo, a aplicação de um modelo de programação matemática para apoiar decisões de fluxo de soja e localização de terminais intermodais. Com base em dados secundários, foram propostas três redes alternativas de escoamento, com diferentes áreas de abrangência e possibilidade de utilização de ferrovias ainda em construção ou projeto. Foram considerados os principais estados produtores e movimentadores da soja. Experimentos computacionais resultaram em fluxos de escoamento e localização de terminais condizentes com a realidade. Cenários alternativos da rede também foram analisados, evidenciando o potencial de análise dessa ferramenta de otimização.

Paulo Estevão Teixeira Martins, Wilingthon Guerra Zvietcovich, Thiago Augusto Oliveira de Silva, and Belmir José de Jesus. 2016. “Allocation of power quality monitors by Clonal Algorithm.” In 17th International Conference on Harmonics and Quality of Power (ICHQP), Pp. 980-985. IEEE. Publisher's VersionAbstract
This paper presents the application of Clonal Algorithm technique for detection of Voltage Sags and Swells with few meters, in order to monitor short-circuit conditions occurring in the electrical network. It is considered possible conditions of symmetry. These conditions make the problem even more complicated. All the proposed method steps are described, from the construction of the antibodies, cloning, mutation of clones, maturation, to the selection of novel antibodies. The algorithm starts with a large number of monitors, and then decrease this number to find a configuration with a minimum number of monitors that ensure monitoring of all short-circuit conditions. The evaluation of the methodology performance for the 63-buses network is presented.
George H. G. Fonseca, Haroldo G. Santos, and Eduardo G. Carrano. 2016. “Integrating matheuristics and metaheuristics for timetabling.” Computers & Operations Research, 74, Pp. 108-117. Publisher's VersionAbstract

The High School Timetabling Problem requires the assignment of times and resources to events, while sets of required and desirable constraints must be considered. The most common approach for this problem is to employ metaheuristic methods. This work presents a matheuristic approach that combines a Variable Neighbourhood Search algorithm with mathematical programming-based neighbourhoods for high school timetabling. Computational experiments on well-known benchmark instances demonstrate the success of the proposed hybrid approach, which outperforms the standalone Variable Neighbourhood Search algorithm by far. Additionally, the proposed algorithm was able to improve 15 out of 17 current best known solutions in a very famous benchmark set.

George H. G. Fonseca, Haroldo G. Santos, Eduardo G. Carrano, and Thomas J. R. Stidsen. 2016. “Modelling and Solving University Course TimetablingProblems Through XHSTT.” XI PATAT - International Confenference on Practice and Theory of Auto-mated Timetabling. Udine, Italy. Publisher's VersionAbstract

The XHSTT was proposed as a standard format to express a wide rangeof School Timetabling problems. Although the format is powerful to represent dif-ferent timetabling features, its application to University Course Timetabling (UCT)problems was not formally studied. The goal of this work is to present encodingsfrom Curriculum-Based Course Timetabling (CB-CTT) and Post-Enrolment CourseTimetabling (PE-CTT) to XHSTT and to evaluate how a state-of-art XHSTT solverperforms on these problems. Computational experiments performed suggested that thisapproach is suitable for dealing with UCT: XHSTT solver would be ranked as fourth inCB-CTT track of the Second International Timetabling Competition (ITC2007) andit achieved feasible solutions for most PE-CTT instances within one hour. Althoughthe results do not outperform the best known approaches for these problems, XHSTTsolvers were designed to handle a wide range of features and constraints beyond the ones present in these models, making it able to fit the specific requirements of several universities.

2015
Thiago A.O.Silva, Mauricio C. de Souza, Rodney R. Saldanha, and Edmund K.Burke. 2015. “Surgical scheduling with simultaneous employment of specialised human resources.” European Journal of Operational Research, 245, 3, Pp. 719-730. Publisher's VersionAbstract
Surgical scheduling is a challenging problem faced by hospital managers. It is subject to a wide range of constraints depending upon the particular situation within any given hospital. We deal with the simultaneous employment of specialised human resources, which must be assigned to surgeries according to their skills as well as the time windows of the staff. A particular feature is that they can be assigned to two surgeries simultaneously if the rooms are compatible. The objective is to maximise the use of the operating rooms. We propose an integer model and integer programming based heuristics to address the problem. Computational experiments were conducted on a number of scenarios inspired by real data to cover different practical problem solving situations. Numerical results show that relaxations provide tight upper bounds, and relax-and-fix heuristics are successful in finding optimal or near optimal solutions.
2013
Bruno N. Gomes, Alexandre X. Martins, and Ricardo S. de Camargo. 4/2013. “An Efficient Genetic Algorithm for the Design of Hub-and-Spoke Networks” 17 (4). Link para acessoAbstract
We propose an efficient genetic algorithm (GA) for the design of hub-and-spoke networks with single allocation. The creation of the initial population is based on the greedy randomized search procedure, which provides high quality individuals. Furthermore, new crossover and mutation operators were implemented in order to improve the solution over the evolutionary process. Additionally, a local search procedure is applied in the best individuals. The adapted GA is tested in the Australian Post (AP) and Civil Aeronautics Board (CAB) data sets and clearly outperforms four other evolutionary algorithms of the literature, both in solution quality and CPU time.
2012
Cruz R. C., Silva T. C. B., Souza M. J. F., Coelho V. N., Mine M. T., and Martins A. X. 12/1/2012. “GENVNS-TS-CL-PR: A heuristic approach for solving the vehicle routing problem with simultaneous pickup and delivery.” In Elsevier Electronic Notes in Discrete Mathematics Volume 39, 1 December 2012, Pages 217-224, 39: Pp. 217-224. Link para acessoAbstract
This work addresses the Vehicle Routing Problem with Simultaneous Pickup and Delivery (VRPSPD). Due to its complexity, we propose a heuristic algorithm for solving it, so-called GENVNS-TS-CL-PR. This algorithm combines the heuristic procedures Cheapest Insertion, Cheapest Insertion with multiple routes, GENIUS, Variable Neighborhood Search (VNS), Variable Neighborhood Descent (VND), Tabu Search (TS) and Path Relinking (PR). The first three procedures aim to obtain an good initial solution, and the VND and TS are used as local search methods for VNS. TS is called after some iterations without any improvement through of the VND. The PR procedure is called after each VNS iteration and it connects a local optimum with an elite solution generated during the search. The algorithm uses an strategy based on Candidate List to reduce the number of solutions evaluated in the solution space. The algorithm was tested on benchmark instances taken from the literature and it was able to generate high quality solutions.
Alexandre X. Martins, Christophe Duhamel, Philippe Mahey, Rodney R. Saldanha, and Mauricio C.de Souzae. 9/2012. “Variable neighborhood descent with iterated local search for routing and wavelength assignment.” In Elsevier, 9th ed., 39: Pp. 2133-2141. Elsevier. Link para acessoAbstract
In this work we treat the Routing and Wavelength Assignment (RWA) with focus on minimizing the number of wavelengths to route demand requests. Lightpaths are used to carry the traffic optically between origin-destination pairs. The RWA is subjected to wavelength continuity constraints, and a particular wavelength cannot be assigned to two different lightpaths sharing a common physical link. We develop a Variable Neighborhood Descent (VND) with Iterated Local Search (ILS) for the problem. In a VND phase we try to rearrange requests between subgraphs associated to subsets of a partition of the set of lightpath requests. In a feasible solution, lightpaths belonging to a subset can be routed with the same wavelength. Thus, the purpose is to eliminate one subset of the partition. When VND fails, we perform a ILS phase to disturb the requests distribution among the subsets of the partition. An iteration of the algorithm alternates between a VND phase and a ILS phase. We report computational experiments that show VND-ILS was able to improve results upon powerful methods proposed in the literature.
1/23/2012. “A Multi-Objective Evolutionary Algorithm Based on Decomposition for Optimal Design of Yagi-Uda Antennas.” In IEEE Transactions on Magnetics, 2nd ed., 48: Pp. 1/23/2012. IEEE.Abstract
This paper presents a multi-objective evolutionary algorithm based on decomposition (MOEA/D) to design broadband optimal Yagi-Uda antennas. A multi-objective problem is formulated to achieve maximum directivity, minimum voltage standing wave ratio and maximum front-to-back ratio. The algorithm was applied to the design of optimal 3 to 10 elements Yagi-Uda antennas, whose optimal Pareto fronts are provided in a single picture. The multi-objective problem is decomposed by Chebyshev decomposition, and it is solved by differential evolution (DE) and Gaussian mutation operators in order to provide a better approximation of the Pareto front. The results show that the implemented MOEA/D is efficient for designing Yagi-Uda antennas.
2011
Eduardo Uchoa, Túlio A. M. Toffolo, Mauricio C. de Souza, Alexandre X. Martins, and Ricardo Fukasawa. 11/29/2011. “Branch-and-cut and hybrid local search for the multi-level capacitated minimum spanning tree problem.” Networks. Link para acessoAbstract
We propose algorithms to compute tight lower bounds and high quality upper bounds (UBs) for the multilevel capacitated minimum spanning tree problem. We first develop a branch-and-cut algorithm, introducing some new features: (i) the exact separation of cuts corresponding to some master equality polyhedra found in the formulation; (ii) the separation of Fenchel cuts, solving LPs considering all the possible solutions restricted to small portions of the graph. We then use that branch-and-cut within a GRASP that performs moves by solving to optimality subproblems corresponding to partial solutions. The computational experiments were conducted on 450 benchmark instances from the literature. Numerical results show improved best known (UBs) for almost all instances that could not be solved to optimality. © 2011 Wiley Periodicals, Inc. NETWORKS, 2012
Alexandre Xavier Martins, Christophe Duhamel, Mauricio Cardoso de Souza, Rodney Rezende Saldanha, and Philippe Mahey. 2011. “A VND-ILS Heuristic to Solve the RWA Problem.” Network Optimization, Pp. 577-582. Link para acessoAbstract
The Routing and Wavelength Assignment problem occurs in optical networks where communication requests fulfilled without inducing wavelength conflicts. We propose a VND-ILS algorithm which is a hybridization of both the ILS metaheuristic and the VND local search. Three neighborhood structures are defined for the VND as well as a perturbation step in the ILS. The efficiency of our approach is illustrated on classical realistic RWA instances. The computational results show our method outperforms some of the best existing methods in the literature.
2009
Alexandre X. Martins, Mauricio C. de Souza, Marcone J. F. Souza, and Túlio A. M. Toffolo. 4/2009. “GRASP with hybrid heuristic-subproblem optimization for the multi-level capacitated minimum spanning tree problem.” Journal of Heuristics, 15, 2, Pp. 133–151. Link para acessoAbstract
We propose a GRASP using an hybrid heuristic-subproblem optimization approach for the Multi-Level Capacitated Minimum Spanning Tree (MLCMST) problem. The motivation behind such approach is that to evaluate moves rearranging the configuration of a subset of nodes may require to solve a smaller-sized MLCMST instance. We thus use heuristic rules to define, in both the construction and the local search phases, subproblems which are in turn solved exactly by employing an integer programming model. We report numerical results obtained on benchmark instances from the literature, showing the approach to be competitive in terms of solution quality. The proposed GRASP have in fact improved the best known upper bounds for almost all of the considered instances.
Thiago Augusto Oliveira de Silva and Leonardo P. Santiago. 2009. “New product development projects evaluation under time uncertainty.” Pesquisa Operacional, 29, 3, Pp. 517-532.Abstract
The development time is one of the key factors that contribute to the new product development success. In spite of that, the impact of the time uncertainty on the development has been not fully exploited, as far as decision supporting models to evaluate this kind of projects is concerned. In this context, the objective of the present paper is to evaluate the development process of new technologies under time uncertainty. We introduce a model which captures this source of uncertainty and develop an algorithm to evaluate projects that incorporates Monte Carlo Simulation and Dynamic Programming. The novelty in our approach is to thoroughly blend the stochastic time with a formal approach to the problem, which preserves the Markov property. We base our model on the distinction between the decision epoch and the stochastic time. We discuss and illustrate the applicability of our model through an empirical example.
2006
Edilaila Fernandes Moraes, José Maria Carmo Bento do Alves, Marcone Jamilson Freitas Souza, Ivo Eyer Cabral, and Alexandre Xavier Martins. 2006. “Um modelo de programação matemática para otimizar a composição de lotes de minério de ferro da mina Cauê da CVRD .” Revista Escola de Minas 59 (3). Link para acessoAbstract
Esse trabalho tem seu enfoque no problema de blendagem de produtos de minério de ferro, estocados nos pátios da mina Cauê, da Companhia Vale do Rio Doce, em Itabira, Minas Gerais, para a composição de lotes. Propõe-se um modelo de programação linear por metas que visa a determinar os locais de retomada dos produtos estocados, de tal forma que a mistura atenda aos limites de especificações de qualidade e quantidade preestabelecidos pelo cliente e satisfaça as restrições operacionais dos pátios. O modelo de programação matemática desenvolvido foi implementado no modelador e otimizador LINGO 9.0, interfaceando com planilhas do EXCEL 2000, possibilitando a utilização e exportação de dados em um ambiente familiar à empresa de mineração. O sistema desenvolvido foi validado comparando-se os resultados obtidos com os produzidos manualmente pela empresa. Os resultados computacionais apresentados comprovaram que é possível prover uma melhora na composição dos lotes através do modelo proposto.
2005
Aloísio Castro Gomes de Júnior, Marcone Jamilson Freitas Souza, and Alexandre Xavier Martins. 11/1/2005. “Um Algoritmo Simulated Annealing Eficiente para o Problema de Roteamento de Veículos com Janela de Tempo.” In XXV Encontro Nac. de Eng. de Produção. Publisher's VersionAbstract
Este trabalho apresenta um algoritmo eficiente, baseado na metaheurística Simulated Annealing (SA), para resolver o Problema de Roteamento de Veículos com Janela de Tempo. Esse problema tem como objetivo determinar as rotas de custo mínimo para uma frota de veículos de mesma capacidade, atendendo à demanda de um conjunto de clientes dentro de um intervalo de tempo determinado, chamado janela de tempo. A metodologia proposta, denominada SA-RAI, incorpora ao algoritmo Simulated Annealing clássico, mecanismos auto-adaptativos para determinação da temperatura inicial e número de iterações em uma mesma temperatura. Ainda nesta metodologia, quando a temperatura atinge um valor limiar, a mesma é reaquecida um certo número de vezes, possibilitando escapar de ótimos locais. Além disso, ela conta com uma fase de intensificação. Sempre que uma melhor solução é encontrada, ela é submetida a um procedimento de refinamento, visando ao seu melhoramento. A metodologia foi aplicada a 168 instâncias-teste da literatura e 13 novos melhores resultados foram encontrados.
Aloísio Castro Gomes de Júnior, Marcone Jamilson Freitas Souza, and Alexandre Xavier Martins. 2005. “Simulated annealing aplicado à resolução do problema de roteamento de veículos com janela de tempo.” ANPET - Associação Nacional de Pesquisa e Ensino em Transportes, 13, 2. Link para acessoAbstract
Este trabalho apresenta um algoritmo eficiente, baseado na metaheurística Simulated Annealing (SA), para resolver o Problema de Roteamento de Veículos com Janela de Tempo. Esse problema tem como objetivo determinar as rotas de custo mínimo para uma frota de veículos de mesma capacidade, atendendo à demanda de um conjunto de clientes, para os quais o atendimento somente é possível dentro de um intervalo de tempo determinado, chamado janela de tempo. A metodologia proposta, denominada SA-RAI, incorpora ao algoritmo Simulated Annealing clássico, mecanismos auto-adaptativos para determinação da temperatura inicial e número de iterações em uma mesma temperatura. Nesta metodologia, quando a temperatura atinge um valor limiar, a mesma é reaquecida um certo número de vezes, possibilitando escapar de ótimos locais. Além disso, ela conta com uma fase de intensificação. Sempre que uma melhor solução é encontrada, ela é submetida a um procedimento de refinamento, visando ao seu melhoramento. A metodologia foi aplicada a 168 problemas-teste da literatura e 13 novos melhores resultados foram encontrados.

Páginas