Melhorando o Desempenho de Algoritmos Evolutivos por Meio de Mineração de Dados: uma Aplicação na Área de Petróleo
DOI:
https://doi.org/10.5540/tema.2005.06.01.0021Abstract
A solução de problemas das classes NP-Completo e NP-Hard tem sido um grande desafio para pesquisadores da área de Otimização Combinatória. Um dos caminhos promissores tem sido reunir conceitos das áreas de Otimização e Inteligência Artificial (IA), procurando conjugar a rigidez dos métodos exatos de otimização com a flexibilidade dos métodos de busca em IA para desenvolver o que chamamos de técnicas inteligentemente flexíveis. As metaheurísticas são resultados desta união e podem ser vistas como heurísticas genéricas que procuram reduzir limitaçôes históricas encontradas em heurísticas tradicionais, como por exemplo, a dificuldade em escapar de ótimos locais muitas vezes ainda distantes de um ótimo global. Neste trabalho são apresentadas alternativas para melhorar o desempenho de Algoritmos Evolutivos no que se refere à qualidade das soluções geradas para o Problema de Roteamento de Unidades Móveis de Pistoneio. São propostos módulos de busca local e mineração de dados para serem incorporados a um Algoritmo Genético (AG). Os resultados demonstraram uma melhora significativa com relação ao AG que continha somente os operadores genéticos.References
[1] R. Agrawal, T. Imielinski e A. Swami, Mining association rules between sets of items in large databases, em “Proc. of the ACM SIGMOD Conf. on Management of Data”, Washington, DC, USA, pp. 207-216, 1993.
R. Aiex, M.G.C. Resende e C.C. Ribeiro, Probability distribution of solution time in GRASP: An experimental investigacion, J. of Heuristics, 8 (2002), 343-373.
D.J. Aloise, J.A. Barros e M. Souza, A genetic algorithm for a Oil Retrieval System, (In Portuguese), em “Proc. of the XXXII Brazilian Symposium on Operations Research”, São Paulo, Brasil, 2000.
J. Beasley, “Population Heuristics”, Handbook of Applied Optimization, Oxford University Press, Oxford, 2002, pp. 138-157.
J.L. Bresina, Heuristic-biased stochastic sampling, em “Proc. of the Thirteenth National Conference on Artificial Inteligence”, Portland, pp. 271-278, 1996.
A.A. Freitas, A survey of evolutionary algorithms for data mining and knowledge discovery, em “ Advances in Evolutionary Computation”(A. Ghosh e S. Tsutsui, eds.), Springer-Verlag, 2002.
F. Glover, M. Laguna e R. Marti, Fundamentals of Scatter Search and Path Relinking, Control and Cybernetics, 39, No. 3 (2000), 653-684.
D.E. Goldberg, “Genetic Algorithms in Search Optimization & Machine Learning”, Addison-Wesley, Menlo Park, 1989.
J.H. Holland, “Adaptation in Natural and Artificial Systems”, University of Michigan Press, Ann Arbor, 1975.
P.Moscato, “On Evolution, Search, Optimization Algorithms andMartial Arts: Towards Memetic Algorithms”, Report 826, Caltech Concurrent Computation Program, California Institute Technology, 1989.
G. Reinelt, TSPLIB - A Traveling Salesman Problem Library, ORSA J. Comput., 3 (1991), 376-384.
M.G.C. Resende e C.C. Ribeiro, Greedy randomized adaptive search procedures, em “Handbook of Metaheuristics”(F. Glover e G. Kochenberger, eds.), Kluwer, pp. 219-249, 2002.
Downloads
Published
How to Cite
Issue
Section
License
Authors who publish in this journal agree to the following terms:
Authors retain copyright and grant the journal the right of first publication, with the work simultaneously licensed under the Creative Commons Attribution License that allows the sharing of the work with acknowledgment of authorship and initial publication in this journal.
Authors are authorized to assume additional contracts separately, for non-exclusive distribution of the version of the work published in this journal (eg, publish in an institutional repository or as a book chapter), with acknowledgment of authorship and initial publication in this journal.
Authors are allowed and encouraged to publish and distribute their work online (eg, in institutional repositories or on their personal page) at any point before or during the editorial process, as this can generate productive changes as well as increase impact and the citation of the published work (See The effect of open access).
This is an open access journal which means that all content is freely available without charge to the user or his/her institution. Users are allowed to read, download, copy, distribute, print, search, or link to the full texts of the articles, or use them for any other lawful purpose, without asking prior permission from the publisher or the
author. This is in accordance with the BOAI definition of open access
Intellectual Property
All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License under attribution BY.