Resumo: A acirrada competitividade entre as empresas de manufatura, presente no cenário atual, exige a busca por sistemas produtivos cada vez mais eficientes e que estejam inseridos num ambiente de constante aprimoramento. Visando à redução de estoques, e agilidade na produção, o conceito de produção enxuta se apresenta como um conjunto de importantes ferramentas operacionais. Desde que surgiram os conceitos associados à produção enxuta, vários estudos que tratam da utilização eficaz de Linhas de Montagem de Modelos Mistos - LMMM se concentram no sequenciamento de tais linhas. Neste trabalho, linhas de montagem múltiplas para modelos mistos, em ambiente de produção enxuta, são estruturas operacionais nas quais vários fornecedores internos abastecem várias linhas de montagem de modelos mistos simultaneamente, de modo que todas as linhas possam receber peças ou subconjuntos de todos os fornecedores. Para otimizar este sistema, a sequência de programação deve buscar o consumo constante de peças ou subconjuntos, minimizando, assim, o dimensionamento dos kanbans e estoques intermediários e, ainda, nivelar a carga de trabalho em cada posto, minimizando as paradas de linha. Baseado no clássico Problema de Monden, que determina o sequenciamento de uma única linha, este trabalho desenvolve um modelo computacional utilizando a Meta-heurística Simulated Annealing e a Simulação a Eventos Discretos para o sequenciamento do sistema de abastecimento cruzado com várias linhas. O modelo proposto busca o sequenciamento das linhas de montagem que mais aproxima o consumo real de um consumo ideal constante e atende a um padrão de eficiência predefinido para as linhas / Abstract: The fierce competition between manufacturing companies, in this current scenario requires the search for production systems more efficient, and which are embedded in an environment of continuous improvement. Seeking to reduce inventories, and speed of production, the concept of lean production is presented as a range of important operational tools. Since the rise of concepts associated with lean production, several studies dealing with the effective use of assembly lines mixed models - LMMM focus on the sequencing of such lines. In this work, multiple assembly lines for mixed models, in an environment of lean production, are operating structures where several domestic suppliers supply many assembly lines mixed models simultaneously, so that all lines can get parts or subsets of all suppliers. To optimize this system, the programming sequence must seek constant consumption of parts or subassemblies, thus minimizing the scaling of kanbans, and intermediate stocks and also level the workload at each station to minimize line stoppages. Based on the classic problem Monden, which determines the sequencing of a single line, this paper develops a computational model using the Simulated Annealing metaheuristic and Discrete Event Simulation for the sequencing of the supply system with crossed several lines. The proposed model seeks the sequencing of assembly lines that most approximates the actual consumption of a consumption ideal constant and respects a predefined standard of efficiency for the lines / Orientador: Fernando Augusto Silva Marins / Coorientador: Jose Arnaldo Barra Montevechi / Banca: Antonio Augusto Chaves / Banca: Jorge Muniz Júnior / Banca: Dagoberto Alves de Almeida / Banca: Rodrigo Scarpel / Doutor

Master of Science

Master of Engineering

No description available.

Thesis (M.S.)--Massachusetts Institute of Technology, Alfred P. Sloan School of Management, 1980. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND DEWEY. / Bibliography: leaves 128-129. / by Bruce William Lamar. / M.S.

Thesis (Ph.D.) -- McMaster University, 1997. / Includes bibliographical references. Also available via World Wide Web.

Thesis (doctoral)--Technische Hochschule Darmstadt, 1995. / Includes bibliographical references (p. [283]-300) and index.

Thesis (doctoral)--Technische Hochschule Darmstadt, 1995. / Includes bibliographical references (p. [283]-300) and index.

Volvo trucks can be ordered with many options to meet customers' demands. This challenges the efficiency of the manufacturing process, especially at the final assembly line where bare chassis are customised with parts and accessories. In the future, assembly lines may be more flexible to allow for assembly of different parts at higher efficiency. This thesis presents problem areas in traditional assembly lines and proposes a proof-of-concept for future flexible assembly line sequencing, as well as a computer simulation tool with the capability to evaluate variances in production capacity when trucks of different size and parts are assembled in consecutive order. Virtual models of flexible assembly lines are constructed as part of a software solution and used to simulate production sequences of varying truck configurations. Data collected from all simulations show a correlation between production capacity and the order in which vehicles are produced. The assembly line configuration itself has also shown to greatly impact efficiency and might lead to an improvement of at least 39%, as well as limitations of tools and workers with specific capabilities. However, the presented performance numbers do not represent all possible simulation outcomes, which beyond the assembly line configuration also depend on things like product complexity and the assembly sequence of individual products.

With the trend towards greater product customization and shorter delivery time, the use of mixed model assembly lines is increasing. A line balancing approach is needed that can address the complex nature of the mixed model line and produce near optimal solutions to problems of realistic size. Due to the combinatorial nature of the line balancing problem, exact solution techniques are limited to small problems. Heuristic methods, on the other hand, are often too simplistic to find good solutions. Furthermore, many of the existing techniques cannot be expanded to handle the mixed model problem. Simulated Annealing (SA) is a search methodology which has exhibited good results when applied to combinatorial optimization problems. In fact, researchers have found that SA is able to find near-optimal solutions while its processing time increases only as a polynomial function of problem size. However, none of the applications found in the literature fully explore the technique's ability to handle a highly-constrained problem such as line balancing. / Master of Science