Tamanhos de lotes capacitados com preparação de máquina em sistemas produtivos ininterruptos

• Ano de defesa: 2007
• Autor(a) principal: Alfredo Carvalho de Castro
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: Set-up; Planejamento de produção; Tamanho de lotes; Programação inteira mista; Fundição; Engenharia de produção
• Link de acesso: http://hdl.handle.net/1843/AMCN-8A2GF5

This work deal with the economic capacitated lot-sizing problem. It is focused on the problems that has setup times and costs with production ramp-up. The type of industry studied has parallel uni-mode machines in a multi-product dynamic demand. The costs and capacity are dynamic and time discrete. This effort is justified by the frequency of this type of problem and the impact of the production plan on the gross margin. This case can be found on material conforming industries, high-pressure die casting and injection mold facilities. In these cases the costs are mostly associated to commodities, in a lower added value situation were storage costs strongly affects the overall costs. Also in this type of industry the capital invested on machinery, installations and tooling demand a great share of the total costs. The main contribution of this work is the development of a method to treat the precedence factor when more the one product is produced on the same time period. This dissertation is organized as follows: Initially it is presented an introduction focusing the application this work on the corporate context. On the first chapter is presented a literature research around the main authors that have publish about Lot-Sizing. The second chapter the main problem of this dissertation is presented and a model is developed to treat this problem. On the third chapter the model performance is analyzed on real world instances and other ones developed based on the literature. The fourth chapter is presented a constructive heuristics approach to the problem. And at the end this work is concluded.