• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Doctoral Thesis
DOI
10.11606/T.45.2009.tde-09072013-090428
Document
Author
Full name
Miriam Harumi Tsunemi
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
São Paulo, 2009
Supervisor
Committee
Esteves, Luís Gustavo (President)
Ho, Linda Lee
Leite, Jose Galvao
Louzada Neto, Francisco
Rodrigues, Josemar
Title in Portuguese
Um modelo Bayesiano semi-paramétrico para o monitoramento ``on-line" de qualidade de Taguchi para atributos
Keywords in Portuguese
Inferência Bayesiana não-paramétrica
mistura de Processos Dirichlet
Monitoramento ``on-line" de Taguchi para atributos
Abstract in Portuguese
Este modelo contempla o cenário em que a sequência de frações não-conformes no decorrer de um ciclo do processo de produção aumenta gradativamente (situação comum, por exemplo, quando o desgaste de um equipamento é gradual), diferentemente dos modelos de Taguchi, Nayebpour e Woodall e Nandi e Sreehari (1997), que acomodam sequências de frações não-conformes assumindo no máximo três valores, e de Nandi e Sreehari (1999) e Trindade, Ho e Quinino (2007) que contemplam funções de degradação mais simples. O desenvolvimento é baseado nos trabalhos de Ferguson e Antoniak para o cálculo da distribuição a posteriori de uma medida P desconhecida, associada a uma função de distribuição F desconhecida que representa a sequência de frações não-conformes ao longo de um ciclo, supondo, a priori, mistura de Processos Dirichlet. A aplicação consiste na estimação da função de distribuição F e as estimativas de Bayes são analisadas através de alguns casos particulares
Title in English
A semi-parametric model for Taguchi´s On-Line Quality-Monitoring Procedure for Attributes
Keywords in English
mixture of Dirichlet Processes
nonparametric Bayesian Inference
Taguchi's On-Line Quality-Monitoring Procedure for Attributes
Abstract in English
In this work, we propose an alternative model for Taguchi´s On-Line Quality-Monitoring Procedure for Attributes under a Bayesian nonparametric framework. This model may be applied to production processes the sequences of defective fractions during a cycle of which increase gradually (for example, when an equipment deteriorates little by little), differently from either Taguchi's, Nayebpour and Woodall's and Nandi and Sreehari's models that allow at most three values for the defective fraction or Nandi and Sreehari's and Trindade, Ho and Quinino's which take into account simple deterioration functions. The development is based on Ferguson's and Antoniak's papers to obtain a posteriori distribution for an unknown measure P, associated with an unknown distribution function F that represents the sequence of defective fractions, considering a prior mixture of Dirichlet Processes. The results are applied to the estimation of the distribution function F and the Bayes estimates are analised through some particular cases.
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
Publishing Date
2013-07-10
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
Centro de Informática de São Carlos
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2020. All rights reserved.