| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 4.57 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
Enquadrado com a realidade no corte de bobines de aço na empresa Ferpinta S.A., este trabalho apresenta uma abordagem de resolução aos problemas numa perspetiva fundamental de racionalização e otimização dos recursos, orientada para objetivos específicos. Os problemas de corte de stock unidimensional consistem em cortar um conjunto de objetos disponíveis em stock, para satisfazer uma determinada procura de itens de modo a otimizar uma determinada função objetivo. É composta por três objetivos distintos (multiobjectivo), i.e., a minimização de perdas geradas no processo de corte (otimização da matéria-prima), a minimização do número de diferentes padrões de corte (tempo de preparação da máquina) e a minimização do corte de itens para stock (sem consumo imediato definido). É apresentado um modelo de programação linear inteiro para resolver o problema de corte unidimensional. O principal propósito é desenvolver como base na estrutura dos problemas de corte, um modelo de apoio à decisão que auxilie no dia-a-dia o decisor a escolher melhores soluções possíveis. Pretende-se otimizar não só em termos de qualidade de solução mas também em termos de tempo necessário para determinar a solução. Foram realizados testes computacionais com dados gerados aleatoriamente e com dados recolhidos na empresa, de forma a avaliar empiricamente o impacto da utilização do modelo proposto.
Enclosed within the reality on slitting steel coils in Ferpinta S.A., this paper presents an approach on solving problems in an ultimate perspective of downsizing and optimizing resources, focused on specific objectives. The one-dimensional cuttingstock problem consists of cutting a set of objects available in stock to satisfy a given search items, in order to optimize a given objective function. It consists of three different objectives (multi-objective), i.e., to minimize losses generated in the slitting process (optimization of the raw material), to minimize the number of different cutting patterns (machine setup time) and the minimization of cutting items to stock (without defined immediate consumption). It is presented as an integer linear programming model to solve the one-dimensional slitting problem. The main objective is to develop has a base on the structure of the cutting problems, a model of decision to assist in the day-to-day decision-maker to choose the best possible solutions. It is intended to optimize not only on a quality of results but also in terms of the time required to determine the finest solution. Computational tests were performed with randomly generated data and also collected in the company in order to empirically evaluate the impact of using the proposed model.
Enclosed within the reality on slitting steel coils in Ferpinta S.A., this paper presents an approach on solving problems in an ultimate perspective of downsizing and optimizing resources, focused on specific objectives. The one-dimensional cuttingstock problem consists of cutting a set of objects available in stock to satisfy a given search items, in order to optimize a given objective function. It consists of three different objectives (multi-objective), i.e., to minimize losses generated in the slitting process (optimization of the raw material), to minimize the number of different cutting patterns (machine setup time) and the minimization of cutting items to stock (without defined immediate consumption). It is presented as an integer linear programming model to solve the one-dimensional slitting problem. The main objective is to develop has a base on the structure of the cutting problems, a model of decision to assist in the day-to-day decision-maker to choose the best possible solutions. It is intended to optimize not only on a quality of results but also in terms of the time required to determine the finest solution. Computational tests were performed with randomly generated data and also collected in the company in order to empirically evaluate the impact of using the proposed model.
Description
Keywords
Problema de corte de stock Unidimensional Multiobjectivo Otimização Programação linear inteira One dimensional cutting Stock Problem Multi-objetive Optimization Integer linear programming