Thesis (MScEng)--University of Stellenbosch, 2001. / ENGLISH ABSTRACT: Statistical experimental design techniques are powerful tools that are often approached with suspicion
and apprehension by experimenters. The trend is to avoid any statistically structured and designed
experimentation program, and to rather use the traditional method of following ones "gut feel". This
approach, more often than not, will supply a satisfactory solution, but there is so much more
information availablefor the same amount of effort.
This thesis strives to outline the method and application of the Taguchi methodology of experimental
design. The Taguchi method is a practical, statistical experimental design technique that does not rely on
the designer's knowledge of the complex statistics typicallyneeded to design experimental programs, a
fact that tends to exclude design of experiments from the averageengineers' toolbox. The essence of the
statistical design of experiments is this: The traditional method of varying one variable at a time and
investigating its effect on an output is no longer sufficient. Instead all the input variables are varied at
the same time in a structured manner. The output trends resulting from each input variable are then
statisticallyextracted from the data in the midst of the variation.
Taguchi method achieves this by designing experiments where every level of every input variable occurs
an equal number of times with every level of every other input variable. The experimental designs are
represented in orthogonal arrays that are chosen and populated by the experimenter by following a
simple procedure.
Four case studies are worked through in this text and, where possible, compared to the "traditional"
approach to the same problem. The case studies show the additional information and time savings
availablewith the Taguchi method, as well as clearlyindicating the importance of using a stable system
on which to do the experiments. The Taguchi method generated more information in fewer
experiments than the traditional approaches as well as allowing analysis of problems too complex to
analysewithout a statisticaldesign of the experimentation procedure. / AFRIKAANSE OPSOMMING: Statistiese eksperimentele ontwerptegnieke is besonder kragtige instrumente wat baie keer met agterdog
deur ekspermenteerders beheen word. Die neiging is om enige statistiese gestruktureerde and ontwerpte
eksperimentele program te vermy, en om liewer die tradisionele metode, wat op 'n mens se intuïsie
staatmaak, te gebruik. Hierdie benadering sal baie keer 'n bevredigende oplossing gee, maar daar is veel
meer inligting vir dieselfde hoeveelheid inspanning verkrygbaar, wanneer die Taguchimetode gebruik
word.
Hierdie tesis strewe om die metode en toepassing van die Taguchimetodologie van eksperimentele
ontwerp voor te lê. Die Taguchimetode is 'n praktiese statistiese eksperimentele ontwerptegniek .wat nie
op die ontwerper se kennis van komplekse statistiek om eksperimentele programme te ontwerp berus
nie. Hierdie komplekse statistiek neig ook om eksperimentele ontwerp van die gemiddelde
ingenieursvaardigehede uit te sluit. Die kern van statistiese eksperimentele ontwerp is die volgende: Die
tradisionele metode van een veranderlike op 'n slag te varieer om die effek op die uitset te ondersoek, is
onvoldoende. In plaas daarvan, word al die insetveranderlikes gelyktydig gevarieer in 'n gestruktureered
manier. Die neigings van elke veranderlike is dan statisties ontleed van die data ten midde van die
variasie van al die ander veranderlikes.
Die Taguchimetode bereik die ontwerpte eksperimente deur elke vlak van elke insetveranderlik in 'n
gelyke aantal keer met elke vlak van elke ander insetveranderlike te varieer. Hierdie is verteenwoordig
deur ortogenale reekse wat gekies en gevul is deur 'n eenvoudige wisselpatroon te volg.
Vier gevallestudies is deurgewerk en, waar moontlik, vergelyk met die tradisonele siening van dieselfde
probleem. Die gevallestudies wys hoe toereikbaar die additionele inligting in die Taguchimethode
toepassings is. Hulle beklemtoon ook die belangrikheid van 'n stabiele sisteem waarop die eksperimente
berus. Die Taguchimetode het meer inligting verskaf met minder eksperimente as die tradisionele
toenaderings, en ook toegelaat dat die analise van probleme, te kompleks om te analiseer sonder om 'n
statistiese ontwerp van eksperimentele prosedure te volg, opgelos kon word.
Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:sun/oai:scholar.sun.ac.za:10019.1/52475 |
Date | 12 1900 |
Creators | Green, Jeremy James |
Contributors | Taylor, A. B., Stellenbosch University. Faculty of Engineering. Dept. of Mechanical and Mechatronic Engineering. |
Publisher | Stellenbosch : Stellenbosch University |
Source Sets | South African National ETD Portal |
Language | en_ZA |
Detected Language | Unknown |
Type | Thesis |
Format | 155 p. : ill. |
Rights | Stellenbosch University |
Page generated in 0.0019 seconds