Gráficos de X com esquemas especiais de amostragem para o controle de médias que oscilam ou para o controle de processos que geram observações autocorrelacionadas

• Ano de defesa: 2014
• Autor(a) principal: Franco, Bruno Chaves [UNESP]
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Estadual Paulista (Unesp)
• Palavras-chave em Português: Autocorrelação (Estatistica); Amostragem (Estatistica); Computação gráfica; Desempenho - Medição; Graficos aleatorios; Autocorrelation (Statistics)
• Link de acesso: http://hdl.handle.net/11449/111071

Control charts proposed by Shewhart assume that the measures of the quality characteristic are independent and identically distributed and the mean of the process does not change until the occurrence of a special cause that moves the mean from its target value. In this thesis are considered two different situations of established by Shewhart: (1) the process mean has an oscillatory behavior that is independent of the special cause, or, the special cause shifts the process mean without altering its oscillatory behavior, (2) observations within the rational subgroups are dependent. Both, the dependence of observations and oscillatory behavior of the process mean are described by an AR (1) model. For better use of the X control chart to control wandering process, this study suggested samples with equally spaced items according to the instant they are produced and selected from the production line. Moreover, for control process, which the observations are autocorrelated, this study suggested samples with observations coming from different rational subgroups. The proposed samples schemes improve the performance of X control charts