Monitoreo de procesos con datos censurados
Cargando...
Archivos
Autores
Conto López, Romario Ademir
Director
Tipo de contenido
Document language:
Español
Fecha
Título de la revista
ISSN de la revista
Título del volumen
Documentos PDF
Resumen
En ocasiones durante el monitoreo de procesos no es posible realizar mediciones exactas sobre las unidades observadas, por cuestiones bien sea de costos y/o tiempo, por tanto es necesario trabajar con observaciones censuradas y por ello no es apropiado aplicar cartas de contol tipo Shewhart tradicionales ya que la censura disminuye la potencia para detectar cambios en los parámetros del proceso que se está monitoreando. Ante esta situación, es recomendable la implementación de cartas de control que consideren el tipo de censura y la distribución estadística de los datos, con el fin de detectar rápidamente posibles cambios en el proceso que puedan ocasionar situaciones de fuera de control estadístico.
Abstract: Sometimes during process monitoring is not possible take exact measurements of the units observed, therefore it should be necessary to work with censored observations. Then it is not appropriate a direct application of traditional Shewhart charts, because in the presence of censuring there is a lower power to detect changes in the process parameters being monitored. According to this situation, we consider the implementation of control charts to take in to account the censorship and a statistical distribution for the data, with the purpose of detect, quickly, possible changes in the process that can cause situations out of statistical control.
Abstract: Sometimes during process monitoring is not possible take exact measurements of the units observed, therefore it should be necessary to work with censored observations. Then it is not appropriate a direct application of traditional Shewhart charts, because in the presence of censuring there is a lower power to detect changes in the process parameters being monitored. According to this situation, we consider the implementation of control charts to take in to account the censorship and a statistical distribution for the data, with the purpose of detect, quickly, possible changes in the process that can cause situations out of statistical control.