Análise do desempenho do gráfico de controle do desvio do nominal em pequenos lotes de produção

• Ano de defesa: 2021
• Autor(a) principal: Meira, Márcio Ricardo Morelli de
• Orientador(a): Oprime, Pedro Carlos
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de São Carlos; Câmpus São Carlos
• Programa de Pós-Graduação: Programa de Pós-Graduação em Engenharia de Produção - PPGEP
• Palavras-chave em Português: Gráficos de controle; Pequenos lotes de produção; Desempenho de gráficos de controle
• Palavras-chave em Inglês: Control charts; Small batches; Performance of control charts
• Área do conhecimento CNPq: ENGENHARIAS::ENGENHARIA DE PRODUCAO
• Link de acesso: https://repositorio.ufscar.br/handle/20.500.14289/14727

Since originally proposed by Shewhart in the 1920s, control charts are widely used by manufacturing companies to monitor and control production processes. The purpose of the control charts is to detect special causes so that corrective actions are taken, whenever necessary, ensuring stability and capacity to the process. Over the years, production systems have changed. The concept of lean manufacturing has gained more space and the use of Shewhart control charts in these environments has been questioned. Specific methods were developed from there, due to the possibility of fundamental assumptions violation and unavailability of data for an accurate parameters estimation. Among the main specific methods, there is the deviation from nominal control chart. However, Montgomery (2014) highlights some essential aspects, such as, for example, the influence of process variation on its performance. This work studies the effect of parameters estimation and the influence of process variation on the deviation from nominal control chart performance through the Average Run Length (ARL) method and proposes some adjustments on the control limits so that its use is optimized in small production batches. For this, the research method used was modeling and computer simulation. The results found validated the propositions presented by the literature, verifying that the control charts built with estimated parameters, instead of known ones, present a lower performance than expected. There was a significant influence of the process variation on the performance of the control chart. However, it has also been demonstrated that small adjustments on the control limits can make it enable on lean manufacturing environments.