• Entrar
    Ver item 
    •   Página inicial
    • BIBLIOTECA DIGITAL: Teses & Dissertações
    • 40001016030P0 Programa de Pós-Graduação em Métodos Numéricos em Engenharia
    • Dissertações
    • Ver item
    •   Página inicial
    • BIBLIOTECA DIGITAL: Teses & Dissertações
    • 40001016030P0 Programa de Pós-Graduação em Métodos Numéricos em Engenharia
    • Dissertações
    • Ver item
    JavaScript is disabled for your browser. Some features of this site may not work without it.

    Desafios na aplicação de Particle Swarm Optimization em um problema de planejamento de produção de uma olaria

    Thumbnail
    Visualizar/Abrir
    R - D - CAROLINA DE ALMEIDA SANTOS PINOTTI.pdf (3.692Mb)
    Data
    2017
    Autor
    Pinotti, Carolina de Almeida Santos
    Metadata
    Mostrar registro completo
    Resumo
    Resumo: O presente trabalho tem como objetivo principal verificar a aplicabilidade de uma math-heurística em um problema de Planejamento de Produção de uma indústria de tijolos. Foi escolhida a metaheurística Particle Swarm Optimization (PSO) para a aplicação no problema. Depois de realizada uma revisão bibliográfica acerca do assunto percebeu-se que são encontrados poucos trabalhos que aplicam um PSO discreto em um problema de planejamento de produção, o que motivou o trabalho. A math-heurística será aqui a combinação do PSO com a metodologia exata Branch-and-Bound. A combinação deve-se ao fato de que a aplicação proposta é feita com um PSO discreto, ou seja, a matriz possui somente entradas inteiras. O uso do PSO discreto fez com que fosse proposta a utilização de uma metodologia exata para que a função objetivo do problema fosse encontrada, sendo que esta depende da distribuição das quantidades a serem produzidas e esses valores, por sua vez, são encontrados com base na matriz discreta que está sendo movimentada pelo PSO. O problema proposto possui preservação da preparação, capacidade (tempo limite de produção) e demanda a ser atendida. A preservação da preparação é o que dificulta a resolução do problema por heurísticas (como o Greedy-Mod) e também faz com que se encontrem muitas infactibilidades. Os desafios encontrados na aplicação dessa proposta não foram poucos. Merece uma maior discussão a codificação e representação das soluções, tratamento de infactibilidades nas soluções iniciais e na movimentação da nuvem. Estes problemas geraram alto custo computacional. Sugere-se uma tentativa de mudança na forma de aplicação (para a utilização de um PSO contínuo), de forma que as infactibilidades sejam reduzidas e o tempo computacional melhorado.
     
    Abstract: The present work has as main objective to verify the applicability of a math-heuristic in the Production Planning of a brick industry. The Particle Swarm Optimization (PSO) metaheuristic was chosen for the application in the problem. After a bibliographical review about the subject, it was noticed that not too many works were found that apply a discrete PSO in a problem of production planning, one of the reasons that motivated this work. The math-heuristic will be the combination of the metaheuristic PSO with the Branch-and-Bound exact methodology (performed by an optimizer). The combination is due to the fact that the proposed application is made with a discrete PSO, that is, the matrix to be moved has only discrete inputs. The use of discrete PSO has meant that the use of an exact methodology is proposed so that the objective function of the problem is found, which depends on the distribution of the quantities to be produced and these values, in turn, are found based on the discrete matrix that is being moved by the PSO. The proposed problem has preservation of the preparation, capacity (time limit of production) and demand to be met. Preservation of the preparation is what makes it difficult to solve the problem by heuristics (such as Greedy-Mod) and also causes many infeasibilities. Deserves a further discussion the codification and representation of solutions, treatment of infeasibilities in the initial solutions and in the movement of the swarm. These problems generated a high computational cost. An attempt is made to change the manner of application (for the use of a continuous PSO), so that the infeasibilities are reduced and the computational time improved.
     
    URI
    https://hdl.handle.net/1884/52836
    Collections
    • Dissertações [102]

    DSpace software copyright © 2002-2022  LYRASIS
    Entre em contato | Deixe sua opinião
    Theme by 
    Atmire NV
     

     

    Navegar

    Todo o repositórioComunidades e ColeçõesPor data do documentoAutoresTítulosAssuntosTipoEsta coleçãoPor data do documentoAutoresTítulosAssuntosTipo

    Minha conta

    EntrarCadastro

    Estatística

    Ver as estatísticas de uso

    DSpace software copyright © 2002-2022  LYRASIS
    Entre em contato | Deixe sua opinião
    Theme by 
    Atmire NV