Loading...
Publication
Planos fatoriais e método de Taguchi : aplicação ao estudo de resinas compostas fotopolimerizadas, usando o R
| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 1.5 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
Esta dissertação aborda técnicas statísticas de planejamentos de experimentos com ênfase em planos fatoriais fracionados com níveis mistos. Planos de Taguchi e planos racionados obtidos com funções dos pacotes AlgDesign, planorR e qualityTools do R são aplicados a um caso real. O objetivo é analisar os efeitos produzidos à microdureza de uma resina composta ativada por fotopolimerizador, testando-se dois tipos de resina, quatro fotopolimerizadores em quatro diferentes profundidades. Os resultados dos planos fracionados são comparados com os decorrentes de um planejamento fatorial completo e também da literatura da área. Eles indicam que estas técnicas de frações de plano experimental foram efetivas para a análise proposta e se apresentam como alternativas viáveis. Os três fatores (resina, fotopolimerizador e profundidade) influenciam na microdureza. A partir do plano de Taguchi foi possível estabelecer uma combinação ótima de fatores e níveis que maximiza a microdureza. Foram também investigados pacotes e funções para as aplicações propostas e desenvolvidas algumas rotinas em linguagem R para aplicação do Método de Taguchi.
This dissertation discusses statistical techniques of design of experiments with emphasis on fractional factorial designs with mixed levels. Taguchi designs and fractionated factorial designs were obtained with functions of the R packages AlgDesign and planorR and were applied to a real case. The objective is to analyze the effects on the micro hardness of a composite resin, testing two types of resin, and four photopolymerizers in four different depths. The analyses of fractionated designs are compared with the results of a full factorial design and the literature. The results indicate that these fractional techniques were effective for the analysis of the experiment and present themselves as viable alternatives. The three factors (resin, photopolymerizers and depth) influence on the micro hardness. From the Taguchi design it was possible to establish a combination of factors and levels that maximizes the micro hardness. Packages and functions for the proposed applications were also investigated and some routines in the R language were developed for implementing the Taguchi method.
This dissertation discusses statistical techniques of design of experiments with emphasis on fractional factorial designs with mixed levels. Taguchi designs and fractionated factorial designs were obtained with functions of the R packages AlgDesign and planorR and were applied to a real case. The objective is to analyze the effects on the micro hardness of a composite resin, testing two types of resin, and four photopolymerizers in four different depths. The analyses of fractionated designs are compared with the results of a full factorial design and the literature. The results indicate that these fractional techniques were effective for the analysis of the experiment and present themselves as viable alternatives. The three factors (resin, photopolymerizers and depth) influence on the micro hardness. From the Taguchi design it was possible to establish a combination of factors and levels that maximizes the micro hardness. Packages and functions for the proposed applications were also investigated and some routines in the R language were developed for implementing the Taguchi method.
Description
Dissertação de Mestrado em Estatística, Matemática e Computação apresentada à Universidade Aberta
Keywords
Estatística Métodos estatísticos Método Taguchi Medicina dentária Qualidade Experimentação Design of experiments Fractional factorial design Taguchi design Composite resins R language
Pedagogical Context
Citation
Henning, Elisa - Planos fatoriais e método de Taguchi [Em linha] : aplicação ao estudo de resinas compostas fotopolimerizadas, usando o R. [Lisboa] : [s.n.], 2014. 101 p.