Localização de concentradores orientada a custos com economia de escala variável

• Ano de defesa: 2009
• Autor(a) principal: Alan Alcides Viana
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: Decomposição de Benders; Localização de concentradores; Programação linear inteira mista; Engenharia de produção
• Link de acesso: http://hdl.handle.net/1843/AMCN-8ALGZ4

The present work shows a new approach for Hub Location Problems. Our goal is to determinate, in a given network, the installation and the location of the hubs and the commodities flow (consumer goods, digital data transmission, passengers) to supply the demand of a set of locations, by a set of distribution sites, with the lowest cost possible.The hub structure acts like transshipment centers that connects several places. So, when a significant quantity of fluxes passes thought a couple of hubs, scale economies arises in the transport cost. It is important to underline that the hub installation is a strategic decision of extreme significance, because its costs, generally, are much bigger than any other costs in the network. Another important factor to be observed is the required service level. When the network is cost driven, there is a need to explore the scale economies provided by the inter-hub connection. If service driven, matters the agility and the comfort that the commodities are transported. In this work, the proposition is to bring global efficiency in the network, exploring to the maximum its logistic structure, with flexibility to choose other routes that doesnt passes thought hubs. A new mixed integer linear programming model is introduced to describe problem in subject, remembering that it is a hard problem to solve. Motivated by solving it in a computational feasible time we applied the Benders Decomposition algorithm, that has a good behavior solving Hub Location Problems in other works.