| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 4.75 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
Uma importante função dos sistemas de controlo da produção é o controlo de entradas (input
control) em produção. Isto envolve a sequenciação dos trabalhos em carteira, i.e., que
aguardam lançamento em produção e a sua seleção e lançamento, atendendo a critérios
relacionados com o próprio trabalho (e.g., data de entrega) e/ou com o sistema de produção
(e.g., limites de carga ou de tempo nos centros de trabalho). A seleção e lançamento é
usualmente feita com recurso a um designado mecanismo de controlo da produção. Diversos
mecanismos têm sido desenvolvidos e propostos na literatura para este fim, como o CONWIP,
entre outros.
Uma outra importante função dos sistemas de controlo da produção é a geração de ordens de
produção e/ou compra, para a qual o sistema de planeamento e controlo da produção DDMRP
foi recentemente proposto. No presente trabalho estudamos a integração do DDMRP para a
geração de ordens com o CONWIP para o seu lançamento em produção, no que concerne às
regras de sequenciação a adotar. Neste âmbito, diversas regras têm também sido desenvolvidas
e propostas na literatura. Contudo, estas têm sido essencialmente desenvolvidas a pensar na
produção por encomenda (make-to-order). A investigação realizada neste âmbito em contextos
de produção para stock (make-to-stock) é ainda reduzida.
Pretende-se com este trabalho rever as regras de sequenciação que têm sido propostas na
literatura para lançamento e avaliar o seu desempenho no âmbito do sistema DDMRP. Para tal,
recorreu-se à simulação de eventos discretos como ferramenta de investigação, considerando
critérios de desempenho como o stock médio de produto acabado e o nível de serviço ou fill
rate.
Os resultados obtidos, permitem concluir que o desempenho das regras de sequenciação é
afetado por diversos fatores, tais como, a configuração do sistema de produção, as quantidades
das ordens de reposição, entres outros. De uma forma geral, não é possível identificar uma
regra que seja superior às outras, para todos os critérios e em todos os cenários experimentais.
Contudo, pode-se concluir que a regra sugerida na bibliografia no contexto do DDMRP
apresenta algumas limitações, as quais são exploradas no contexto deste trabalho.
An important function of production control systems is input control in production. This involves the sequencing of jobs in the backlog, that is, waiting to be released into production, and their selection and release, taking into account criteria related to the job itself (e.g., delivery date) and/or to the production system (e.g., load or time limits at work centers). The selection and release are usually done using a so-called production control mechanism. Several mechanisms have been developed and proposed in the literature for this purpose, such as CONWIP, among others. Another important function of production control systems is the generation of production and/or purchase orders, for which the DDMRP production planning and control system has recently been proposed. In this paper we study the integration of DDMRP for generating orders with CONWIP for launching them into production, in terms of the sequencing rules to be adopted. In this context, several rules have also been developed and proposed in the literature. However, these have been developed mainly with make-to-order production in mind. The research done in this area in the context of make-to-stock production is still limited. This work aims to review the sequencing rules that have been proposed in the literature and evaluate their performance in DDMRP systems context. To do so, discrete event simulation was used as a research tool, considering performance criteria such as Finished Goods Stock average and service level or Fill Rate. The results allow us to conclude that the performance of the sequencing rules is affected by various factors, such as the configuration of the production system and the quantities to be ordered for replenishment orders, among other factors. In general, it is not possible to identify a rule that is superior to the others, for all the criteria and in all the experimental scenarios. However, it can be concluded that the rule suggested in the literature in the context of DDMRP has some limitations, which are explored in the context of this work.
An important function of production control systems is input control in production. This involves the sequencing of jobs in the backlog, that is, waiting to be released into production, and their selection and release, taking into account criteria related to the job itself (e.g., delivery date) and/or to the production system (e.g., load or time limits at work centers). The selection and release are usually done using a so-called production control mechanism. Several mechanisms have been developed and proposed in the literature for this purpose, such as CONWIP, among others. Another important function of production control systems is the generation of production and/or purchase orders, for which the DDMRP production planning and control system has recently been proposed. In this paper we study the integration of DDMRP for generating orders with CONWIP for launching them into production, in terms of the sequencing rules to be adopted. In this context, several rules have also been developed and proposed in the literature. However, these have been developed mainly with make-to-order production in mind. The research done in this area in the context of make-to-stock production is still limited. This work aims to review the sequencing rules that have been proposed in the literature and evaluate their performance in DDMRP systems context. To do so, discrete event simulation was used as a research tool, considering performance criteria such as Finished Goods Stock average and service level or Fill Rate. The results allow us to conclude that the performance of the sequencing rules is affected by various factors, such as the configuration of the production system and the quantities to be ordered for replenishment orders, among other factors. In general, it is not possible to identify a rule that is superior to the others, for all the criteria and in all the experimental scenarios. However, it can be concluded that the rule suggested in the literature in the context of DDMRP has some limitations, which are explored in the context of this work.
Description
Keywords
Production control Sequencing CONWIP DDMRP Simulation Controlo da Produção Sequenciação Simulação