| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 4.68 MB | Adobe PDF |
Advisor(s)
Abstract(s)
Esta dissertação descreve o trabalho desenvolvido durante o estágio curricular numa empresa do setor alimentar, com o objetivo da criação de um modelo mais eficiente de distribuição de produto acabado. Para a obtenção do novo modelo de distribuição, foi desenvolvida uma ferramenta de apoio à decisão baseada num modelo matemático de programação inteira para o roteamento dos veículos da frota própria da empresa, enquanto que, para a frota subcontratada a mesma ferramenta baseia-se num método de duas fases. Desta forma, relativamente à frota subcontratada, a primeira etapa consiste numa repartição lógica das encomendas a efetuar, através de um método de clustering e, em seguida, o mesmo modelo matemático criado para o roteamento da frota própria, efetua um roteamento ótimo para cada cluster obtido. A ferramenta de apoio à decisão provou ser capaz de atingir os objetivos propostos e sugere reduções de custo de transporte entre 18,71% e 48,20%, face ao modelo de distribuição atual da empresa. Relativamente ao custo por quilograma (€/kg), foi também obtida uma margem de poupança entre os 18,71% e 48,20%, face ao modelo atual. Deste modo, acordo com os resultados, no pior dos cenários calculados, a ferramenta criada sugere lucros anuais até 83000 €, podendo estenderem-se até aos 377000 € de acordo com o cenário mais otimista. Foi ainda desenvolvido um ficheiro de interface com o utilizador em Microsoft Excel para permitir que a ferramenta de apoio à decisão seja utilizada, sem qualquer dificuldade, por qualquer operador da empresa. Deste modo, como resultado deste projeto, a ferramenta de apoio à decisão desenvolvida permite a criação de um novo modelo de distribuição de produto acabado mais eficiente e com a gestão do transporte centralizada na empresa.
This dissertation describes all the work developed during a curricular internship, in a company that operates in the food industry, with the objective of creating a more efficient model of finished goods distribution. To obtain the new distribution model, a decision support tool was developed based on a mathematical model of integer programming for the routing of the company's own fleet, while for the subcontracted fleet the same tool is based on a two-phased method. In this way, the first step of the method applied to the subcontracted fleet consists in a logical division of the orders to be made, through a clustering method and then, the same mathematical model created for the routing of the own fleet, is used to obtain an optimal routing for each cluster created. The decision support tool proved to be able to reach the initial proposed objectives and suggests reductions of the transportation cost between 18.71% and 48.20%, given the current distribution model of the company. About the cost per kilogram (€/kg), a saving margin of 18.71% and 48.20% was also obtained, compared to the company’s current model of distribution. According to the results, in the worst-case scenario, the tool created suggests annual profits up to € 83000 and can be extended up to € 377000 according to the most optimistic scenario. A user interface file was also developed in Microsoft Excel to allow the decision support tool to be used, without any difficulty, by any operator of the company. The developed decision support tool allows the creation of a new model of finished goods distribution much more efficient and with the management of the distribution centralized in the company.
This dissertation describes all the work developed during a curricular internship, in a company that operates in the food industry, with the objective of creating a more efficient model of finished goods distribution. To obtain the new distribution model, a decision support tool was developed based on a mathematical model of integer programming for the routing of the company's own fleet, while for the subcontracted fleet the same tool is based on a two-phased method. In this way, the first step of the method applied to the subcontracted fleet consists in a logical division of the orders to be made, through a clustering method and then, the same mathematical model created for the routing of the own fleet, is used to obtain an optimal routing for each cluster created. The decision support tool proved to be able to reach the initial proposed objectives and suggests reductions of the transportation cost between 18.71% and 48.20%, given the current distribution model of the company. About the cost per kilogram (€/kg), a saving margin of 18.71% and 48.20% was also obtained, compared to the company’s current model of distribution. According to the results, in the worst-case scenario, the tool created suggests annual profits up to € 83000 and can be extended up to € 377000 according to the most optimistic scenario. A user interface file was also developed in Microsoft Excel to allow the decision support tool to be used, without any difficulty, by any operator of the company. The developed decision support tool allows the creation of a new model of finished goods distribution much more efficient and with the management of the distribution centralized in the company.
Description
Keywords
Logística Distribuição Ferramenta de apoio à decisão Roteamento de veículos Clustering Logistics Distribution Decision support tool Vehicle routing