Sei sulla pagina 1di 19

Taguchi Methods

Computer Aided Engineering Assignment


Department of Computer Engineering
Curtin University of Technology
DISCLAIMER
http://kernow.curtin.edu.au/www/Taguchi/CAE204.HTM

Introduction
The following article is an introduction to the basic principles and philosophies behind
the Taguchi approach to quality control. It will attempt to give an overview on the
Taguchi techniques used in quality control. These include, the concept of Parameter
design, Orthogonal arrays and the process of Brainstorming.

Table of Contents
Introduction
The Taguchi Philosophy
The Fundamental Taguchi Concepts
The Total Loss Function
Total Loss Function Case Study
Taguchi Quality Strategy
Taguchi Techniques using Quality Control
Brainstorming
Orthogonal Arrays
Parameter Design
Conclusion
References
Authors

INTRODUCTION
The word quality cannot have a specific meaning when applied to the manufacturing
industry. This is basically because the word quality changes within the context it is being
used. For a long time, manufacturing industries in European and American countries have
worked from the basis of a tolerance. This tends to suggests that the manufactured item
would be passed as acceptable if its quality was within the specified tolerance range.
Taguchi's response to quality differs rather greatly from the goalpost philosophy of the
European and American countries. The Japanese implementation of Taguchi's concept

sees them working on the principle that when designing a product, it should be designed
with minimum loss, with the relative product being designed as close to the optimum
value as is feasibly possible. This would result in the product being manufactured in
regards to its life cycle and customer satisfaction from the design stages. It would also
mean that less repair work would be required in the long run.
The Authors of the article would like to thank you for showing interest in this article. The
setup has been constructed for ease of navigation. It consists of a table of contents with
links to all sections. Each section at the bottom comprises of convenient links to navigate
between and within the relevant sections. In-text referencing is also implemented
extensively throught the article. To return to the home page of the article, please click on
the Taguchi Home Page icon.

2.0 Taguchi Philosophy


It is being increasingly recognised that the high quality of a product or service and the
associated customer satisfaction are the key for enterprise survival. Also recognised is the
fact that pre-production experiments, assuming properly designed and analysed, can
contribute significantly towards quality improvements of a product. A traditional (but still
very popular) method of improving the quality of a product is the method of adjusting
one factor at a time during pre-production experimentation. In this method, the engineer
observes the result of an experiment after changing the setting of only one factor
(parameter). This method has the major disadvantages of being very costly and
unreliable. The Japanese were the first to realise the potential of another method using
statistical design of experiments (SDE) - originally developed by R. Fisher [1]. SDE, in
contrast to the one factor method, advocates the changing of many factors simultaneously
in a systematic way (ensuring an independent study of the product factors). In either
method, once factors have been adequately characterised, steps are taken to control the
production process so that causes of poor quality in a product are minimised.
In the manufacturing industry, one area of current development is concerned with the
application of modern off-line quality control techniques (pre-production
experimentation and analysis) to product and process engineering. Most of the ideas for
these quality control techniques are derived from W. E. Deming [2]. These ideas were
built upon by Professor Genichi Taguchi. While Deming's main achievements was to
convince companies to shift quality improvements to statistical control of the production
process [3], Taguchi makes a further step back from production to design, to make a
design robust against variability in both production and user environments.
Five major points of the Taguchi quality philosophy are :
In a competitive market environment, continual quality improvements and cost
reductions are necessary for business survival.
An important measurement of the quality of a manufactured product is the total
loss generated by that product to the society.

Change the pre-production experimental procedure from varying one factor at a


time to varying many factors simultaneously (SDE) , so that quality can be built
into the product and the process.
The customer's loss due to poor quality is approximately proportional to the
square of the deviation of the performance characteristic from its target or
nominal value. Taguchi changes the objectives of the experiments and the
definition of quality from "achieving conformance to specifications" to "achieving
the target and minimising the variability.
A product (or service) performance variation can be reduced by examining the
non-linear effects of factors (parameters) on the performance characteristics. Any
deviation from a target leads to poor quality.
Taguchi's main objectives are to improve process and product design through the
identification of controllable factors and their settings, which minimise the variation of a
product around a target response. By setting factors to their optimal levels, a product can
be manufactured more robust to changes in operation and environmental conditions.
Taguchi removes the bad effect of the cause rather than the cause of a bad effect, thus
obtaining a higher quality product.

3.0 The Fundamental Taguchi Concepts


The definition of quality given by the Taguchi methodology is customer orientated.
Taguchi defines quality in a negative manner "Quality is the loss imparted to society from the time the product is shipped" [3]
. This "loss" would include the cost of customer dissatisfaction that leads to the loss of
company reputation. This differs greatly from the traditional producer-orientated
definition which includes the cost of re-work, scrap, warranty and services costs as
measures of quality. The customer is the most important part of the process line, as
quality products and services ensure the future return of the customer and hence improves
reputation and increased market share. In general, there are four quality concepts devised
by Taguchi :
1. Quality should be designed into the product from the start, not by inspection
and screening. (Details of Concept 1)
2. Quality is best achieved by minimising the deviation from the target, not a
failure to confirm to specifications. . (Details of Concept 2)
3. Quality should not be based on the performance, features or characteristics of
the product. . (Details of Concept 3)
4. The cost of quality should be measured as a function of product performance
variation and the losses measured system- wide. (Details of Concept 4)
The above concepts are detailed below:

3.1 Quality Concept One


Quality should be designed into the product from the start, not by inspection and
screening. Quality improvements should occur during the design stages of a product or
process, and continue through to the production phase. This is often called an "off-line"
strategy. Poor quality can't be improved by the traditional process of inspection and
screening (on the production line). According to Taguchi, no amount of inspection can
put quality back into a product; it merely treats the symptom [3]. Quality concepts should
therefore be developed by the philosophy of prevention ; problems are tackled at the
source and not down stream. Taguchi emphasises that quality is something that is
designed into a product, to make it robust and immune to the uncontrollable
environmental factors in the manufacturing phase. This leads us to the next quality
concept of minimising variation in a product. Return to Quality Concepts

3.2 Quality Concept Two


Quality is best achieved by minimising the deviation from the target, not a failure to
confirm to specifications. The product should be designed so that it is robust or immune
to uncontrollable environmental factors - eg. noise, temperature and humidity. This
concept mainly deals with actual methods of affecting quality. Reducing variation is the
key to improving quality. By specifying a target value for critical parameters, and
ensuring manufacturing meets the target value with little deviation, the quality may be
greatly improved. Return to Quality Concepts

3.3 Quality Concept Three


Quality is not based on the performance, features or characteristics of the product.
Adding features to a product is not a way of improving quality, but only of varying its
price and the market it is aimed at. The performance and characteristics of a product, can
be related to quality, but should not be the basis of quality. Instead, performance is a
measure of product capability. Return to Quality Concepts

3.4 Quality Concept Four


The cost of quality should be measured as a function of product performance variation
and the losses measured system-wide. From given design parameters, the deviations from
a target are measured in terms of the overall life cycle costs of the product. This includes
costs or re-work, inspection, warrantry servicing, returns and product replacement. It is
these costs that provide some guidance as to which major parameters need controlling.
Return to Quality Concepts

4.0 The Total Loss Function


In the early 1980s, Dr Taguchi proposed the following statement relating to the quality of
a product:

"Quality is the financial Loss to society after the article is shipped".


This is one of the many concepts which was developed by Dr Taguchi. However, the
above statement somehow depicts inequality, since a loss to society is not a desirable
characteristic. The idea of quality is related to something which is new, beautiful and
good which in the engineering sense, must have many features or functionalities. We can
rearrange the above definition and still retain the basic concept of Quality to denote a
positive attribute as follows:
"Quality is the avoidance of financial loss to society after the product is shipped"
The important point here is the fact that quality is related to monetary loss and not to any
other factors or conditions. Even though the actual loss maybe the loss of functionality to
the product, or other losses such as pollution, time, noise, etc., the overall effect is a
financial loss. It can also be expanded to include the development, and manufacturing
phases of a product.
A poorly designed product begins to impart losses to society from the very start of the
production stage, and continues to do so, until steps are taken to improve its functionality
and performance. There are two major categories of loss to society with respect to the
product quality:
The first category relates to the losses incurred as a result of harmful effects to
society. ( e.g. pollution ).
The second category relates to the losses arising because of excessive variation
in functional performances.
The second category has a dominant impact to the design stages of the product and will
be discussed here.
The conventional method of computing the cost of quality is based on the number of
parts rejected and reworked. This method of quality evaluation is incapable of
distinguishing between two samples, that are both within the specification limits, but with
different distributions of targeted properties.
The following figure illustrates the difference between the conventional method and
Taguchi's view of the loss function.

Fig 4.1 Taguchi and The Conventional Loss Functions.


The graph depicts the loss function as a function of deviation from an ideal, or the target
value of a given design parameter. Here m represents the target value, or the most
desirable value, of the parameter under consideration. This parameter may be critical
dimension, colour of the product, surface finish or any other characteristics that
contributes to the customer's conception of quality.
UAL and LAL in the figure 4.1 represent upper and lower acceptable limits of a design
parameter, respectively. Normally, the product is functionally acceptable if the value of
the specified parameter is within the range between the UAL and LAL limits. No societal
loss is assumed to occur and the product is accepted for further processing. However, if
the product lies outside these limits, it is either discarded or subjected to salvage
operations. Every attempt is made to control the manufacturing process to maintain the
product within these limits.
However, according to Taguchi, there is no sharp cut-off in the real world. Performance
begins to gradually deteriorate as the design parameter deviates from its optimum value.
Therefore, he proposed that the loss function be measured by the deviation from the ideal
value. This function is continuous as shown by the dotted line in Figure 4.1. Product
performance begins to suffer when the design parameters deviate from the ideal or the
target value. This loss function takes the following basic quadratic form:
L(x) = k(x - m)
Where L is the loss in dollars (money), m is the point at the which the characteristic
should be set, x is where the characteristic actually is set, and k is a constant that depends
on the magnitude of the characteristic and the monetary unit involved.
This basic loss function is used if no other function based on data is available. However,
when no market research data is available, the next best option is to use the quadratic loss
function.

From the quadratic loss function, the total loss increases parabolically as the deviation
from the target value increases. This loss represents a continuous function. This indicates
that by making a product within the specification limits, it does not necessary mean that
the product is of good quality, since good quality is now defined as keeping the product
characteristic on target with low variation. Taguchi emphasis on optimum customer
satisfaction is by developing products which meet the target value on a consistent basis.
Thus, the most important aspect of Taguchi's quality control philosophy is the
minimisation of variation around the target value.

5.0 Total loss function Case Study


To illustrate the advantages of the loss function, a case study which places emphasis on
low variations of the target value to minimise cost and loss of the product will be
examined.

5.1 Loss function in Developmental Design


The loss function can be used to compute the advantages of being on target with low
variation for the distribution of a product characteristic. The following figures describe
the relationship between the output voltage and the gain of a power transistor in a
regulated power supply circuit. This information is commonly available from transistors
manuals and data sheets which are published by the manufacturers of the components.

Fig 5.1.0
From the figure, it can be seen that for a design specification of 115V, it would be
necessary to use a transistor with a gain of 20, which would cost approxiamateley 25
cents. The cost of the electronic component depends on the tolerance and the power

handling capability. The 25 cent transistor has a tolerance of 30% which will be
assumed to be three standard deviation away from the target value. Hence one standard
deviation is equivalent to 10% in tolerance.
In figure 5.1.0, it can also be seen that the variation in the gain, is transmitted to the
variation in the voltage. If a normal distribution of the gain is assumed , then a normal
distribution of voltage will be obtained. While it is centred around the target of 115V, it is
also possible to have a voltage as low as 109 and as high as 121V. If a higher tolerance
transistor is used then the tolerance would be reduced. However, this would mean a
higher cost in the overall product. To produce the true regulated power supply could then
cost up to four times the original design cost.
However, a more cost effective approach is to use the portion of the voltage Vs gain
curve that is less steep. In this way, the large variation in gain to the output voltage is not
transmited. This part of the curve is around the 40-HFE point. Even with the 30%
tolerance range, the variation in voltage around this point is 2 Volts. Hence, it is seen
that designing in the constant region of the relationship will reduce the variation.
The distributions for transistor A and transistor B are shown in the following diagram:

Figure 5.1.1 Distribution Curve for Transistor A and B


The figure shows the output of both transistors, with the output of transistor B shifted and
superimposed over the output of transistor A. From inspection, transistor A, with is wide
and flat distribution curve, will exhibit a large variation in its output voltage. Whereas
transistor B will exhibit the opposite characteristics.
To calculate the expected loss, integration of the area of the loss function with the area of
the distribution must be preformed. This may be done numerically, point by point, or by
combing the distribution function with the loss function.
The results shows that the expected loss (EL) which is related to the standard deviation k
and the location of the average of the distribution to be:
EL = k[ (AL - m ) + S ]

The losses for the two transistors can then be calculated as follows:
For transistor A:
EL = 0.444[ ( 115 -115) + 2 ]
EL = $1.78
For transistor B:
EL = 0.444[ (124 -115) + 0.33 ]
EL = $36.01
The loss for transistor B is excessive because it is 9 volts of the target. This loss can be
minimised by using a higher current limiting resistor which would shift the entire voltage
versus gain curve as follows:

Figure 5.1.2 New Gain Vs Voltage Curve


The loss for transistor B now becomes:
EL = 0.444[ (115 -115 ) + 0.33 ]
EL = $0.048
Hence the above case study has shown how Taguchi quality engineering by design
methodology can be put into practice to minimise the output variation. Since loss is a
function of the variation, reducing the variation will ultimately reduce the loss.

6.0 Taguchi Quality Strategy


Taguchi viewed poor quality as a lack of consistency in the ingredients of the product.
Because of this inconsistency, the manufactured products may not satisfy the

quality(product variation) and specification(target value) demands of the customers. The


product mean value is off target and the variation around the mean is large(Total Loss
function.) Thus, methodologies and techniques have been developed by Taguchi to
reduce the elements of variation:
Deviation from the targeted value, these include the total loss function and
parameter design techniques and
Variations with respect to others in the group. Which is dealt with in Orthogonal
Arrays

6.1 Quality Strategy


Taguchi observed that variation in product specification, is the primary cause for rejection
of a product. The cure for this quality loss is to reduce the variation, and effort should be
directed toward zero variation, zero defect. The Taguchi approach for reducing variation
in the product is a two-step process :
Manufacture the product in the best manner most of the time. (less deviation
from the target)
Produce all products as identically as possible. (less variation between the
products)
This approach is a general methodology by Taguchi in quality control. Since the quality
of a product, or process, may be difficult to define in quantitative terms.
Quality is what the customers perceive it to be.
Thus, quality varies from customer to customer, and from product to product. But, in
general, lack of product consistency is the major factor in the perception of poor quality.
Taguchi quality strategy is to improve quality in the product design stage by :
Making the design less sensitive towards influence of uncontrollable factors.
Optimizing the product design.

Taguchi Techniques in Quality Control


The following techniques are used in Taguchi quality control.

Brainstorming
Orthogonal Arrays

Parameter Design

Brainstorming

The brainstorming stage is perhaps the most important stage of the whole Taguchi
procedure. At this stage, clear statements of the problems are established, the objectives,
the desired output characteristics, the methods of measurement and the appropriate
experiments are designed. Brainstorming is an activity which promotes group
participation, encourages creative thinking and generates many ideas in a short period of
time.
The experiment stage is where resulting data is examined and an interpretation of results
is conducted. The variability control factors are determined and the optimal values are
selected so that the variability of the product is minimised. The target control parameters
are also determined during this stage, and their settings are selected for a desired mean
response. Predictions are also made at this stage of the Taguchi procedure.
To confirm the predicted results, a confirmatory experiment is conducted. This is an
essential stage in order to confirm that the new parameter settings do provide optimal
performance. Confirmation removes any concerns of the incorrect choice of parameters,
experimental design or assumptions of product response. After which, corrective action
can then be followed. A study of the new system should take place and any improvements
require standardising. After successful improvements, the optimal settings will, from then
on, become a standard.

Orthogonal Arrays
A product can be designed and manufactured based on a set of specifications demanded
by the customer. Each specification has a required parameter value or values, which the

manufactured product must be able to satisfy. Thus, the manufacturing process must be
capable of producing the designed parameters, which is termed as the targeted value,
according to the customer's specifications. Unfortunately in reality, manufacturing
processes are far from ideal. Products manufactured tends to give a distribution that has a
mean value slightly different from the targeted value. Thus, one of the main technique
used in Taguchi's quality control is to reduce the variation around the targeted value.
According to Taguchi, the quality of a group of products can be improved by achieving
its end product specifications distribution as close to the target value as possible. This
concept can be realised by designing and building the quality into the product itself.
Hence, Taguchi employs design experiments using specially constructed table, known as
"Orthogonal Arrays (OA)" to treat the design process, such that the quality is build into
the product during the product design stage.
Discussions of the various aspects of Orthogonal Arrays(OA) can be found in the
following links:
The Approach of OA
Comparison to the Conventional Approach
OA Analysis
Application of OA
Advantages and Disadvantages of OA

The Approach of Orthogonal Arrays


An experiment during the product design stages, involves the materials used in
manufacturing the experimental product which affects the final quality outcome. Factors
such as variations in the chemical ratio, the level of ingredients used, and how the
product is formed together, will contribute to the variation in the targeted value of the
final product.
Orthogonal Arrays(OA) are a special set of Latin squares, constructed by Taguchi to lay
out the product design experiments. By using this table, an orthogonal array of standard
procedure can be used for a number of experimental situations. Consider a common 2level factors OA as shown in table 1 below :

Table 1. An orthogonal array of L8.


The array is designated by the symbol L8, involving seven 2-level factors, zeros and
ones. The array has a size of 8 rows and 7 columns. The number (zeros/ones) in the row
indicate the factor levels (be it a fluid viscosity, chemical compositions, voltage levels,
etc.) and each row represents a trial condition. The vertical columns represents the
experimental factors to be studied. Each of the assigned columns contain four levels of
zeros(0), and four levels of ones(1), these conditions, can combine in four possible ways,
such as (0,0), (0,1), (1,0), (1,1,), with 27 possible combinations of levels. The columns
are said to be orthogonal or balanced, since the combination of the levels occurred the
same number of times, when two or more columns, of an array are formed. Thus, all
seven columns of an L array, are orthogonal to each other.
The OA facilitates the experimental design process by assigning factors to the appropriate
columns. In this case, referring to table 1, there are at most seven 2-level factors, these
are arbitrarily assigned factors A, B, C, D, E, F, and G to columns 1, 2, 3, 4, 5, 6, 7 and 8
respectively, for an L8 array. From the table, eight trials of experiments are needed, with
the level of each factor for each trial-run as indicated on the array. The experimental
descriptions are reflected through the condition level. For example, 0 may indicates the
factor is not applied, and 1 represents the factor that is fully applied. The factors may be
variation in chemical concentration, material purity, mechanical pressure and so on. The
experimenter may use different designators for the columns, but the eight trial-runs will
cover all combinations, independent of column definition. In this way, the OA assures
consistency of the design carried out by different experimenters. The OA also ensures
that factors influencing the end product's quality are properly investigated and controled
during the initial design stage.

Comparison to the Conventional Approach


The method of investigating all possible combinations and conditions in an
experiment(involving multiple factors) is traditionally known as factorial design. The
factorial design is based on the theory, that for a full factorial design, the number of
possible designs, N (number of trails), is :
N = Lm
where L = number of levels for each factor
m = number of factors involved
Thus, if the qualities for a given product depended on three factors, the variation of 2level conditions can be limited to a number of design experiments of 23, which equals 8
trials. If the same method is carried out on the conditions based on table 1, for 27, then
128 trials would be needed. Moreover, the method of level combinations laid out is not
specified in the factorial design process. This may lead to different results on the same
experimental subject each time a trial is conducted. Thus, Taguchi's Orthogonal Array is
able to simplify and standardised the factorial designs, in a manner that will yield
consistent data results and similar outcomes, even though the trials are carried out by

different experimenters. Thus, two different investigators will have similar conclusion
and a standard design methodology.
The concept of standard design methodology and uniform results through OA analysis is
very important, since it allows the manufacturer to produce two products of the same
quality standards, using the same materials, but with differences in the manufacturing
process. This is possible since, through OA experimental analysis, the quality influencing
factors of a product can be identified, controlled, and hence compensated during the early
product design stage. Thus, the quality of the product itself, rather than depending on the
manufacturing process, is able to "adapt" to the manufacturing process.
Taguchi's OA is considered to be more superior than the traditional factorial design
method since :
The factorial design experiment is not efficient in handling large number of
factor variables.
Taguchi's OA experiments, on a product design yield similar and consistent
results, although the experiment can be carried out by different experimenters.
The OA table allows determination of the contribution, of each quality
influencing factor.
OA allows easy interpretation of experiments with a large number of factors.

Orthogonal Array Analysis


The results obtained from the orthogonal array are then analysed to achieved the
following objectives :
To estimate the contribution of individual quality influencing factors in the
product design stage.
To gain the best, or optimum, condition for a process, or a product, so that good
quality characteristics can be sustained.
To approximate the response of the product design parameters under the
optimum conditions.
The contribution of individual quality influencing factors, is the deciding key of the
control to be enforced on the product design. A commonly applied statistical treatment The Analysis of Variance (ANOVA) - is used to analyse the results of the OA experiment
in product design, and to determine how much variation each quality influencing factor
has contributed. By studying the main effects of each of the factors, the general trends of
the influence factors, towards the product, or process, can be characterised. The
characteristics can be controlled, such that a lower, or a higher, value in a particular

quality influencing factor, produces the preferred result. Thus, the levels of influencing
factors , to produce the best results, can be predicted.
There are two different methodologies in carrying out the complete OA analysis. A
common approach is to analyse the average result of repetitive runs, or a single run,
through ANOVA analysis as discussed. Another approach, which is a better method for
multiple runs, is to use signal (S) to noise (N) ratio (S/N) for the same steps in the
analysis. The objective of S/N analysis, is to determine the most optimum set of the
operating conditions, from variations of the influencing factors within the results. The
signals, in this case, will be those factors which are invariant. The noise are those
influencing factors which are active. Details regarding the methods of OA results analysis
using ANOVA and signal-to-noise ratio can be referred to article [1].

Application of Orthogonal Array


Taguchi's OA analysis is used to produce the best parameters for the optimum design
process, with the least number of experiments. The OA manages to transform a quality
concept into the product design. The OA method is able to treat quality influencing
factors at discrete levels, and often this method save time, and indirectly reduces the cost
of hardware testing. Thus, the OA is usually applied in the design of engineering
products, test and quality development, and process development. All applications
involved have a common objective, that is to use Taguchi's OA method to build the
quality into a product at the initial design stage.

Advantages and Disadvantages of Orthogonal Array


The advantages of OA, are such that they can be applied to experimental design
involving a large number of design factors. The OA design experiments, analysis, and
cost guidance based on the loss function have made this approach more attractive. The
limitation of OA is that it can only be applied at the initial stage of the product/process
design system. There are some situations whereby OA techniques are not applicable, such
as a processes involving influencing factors that vary in time and cannot be quantified
exactly.

Parameter Design
When a product is said to be optimum, it implies that the product has achieved most of
the target values set out by the quality measure. Taguchi tries to reduce the variation
around the target, not by eliminating the cause of variation, since totally removing the
cause of variation can be expensive in an industrial setting. The variation is reduced by

adjusting the levels of the influencing factors, and controlling the variation of other
factors, which is the approach of the parameter design technique.
To illustrate this concept, consider an example involving only one factor.
An electronic device which controls the luminance of a light bulb, was influenced
significantly by the applied voltage. The investigator, wishing to select the right voltage,
investigate the luminance of the light bulb at several input voltages. The influence of the
voltage variation on the luminance of the light bulb can be shown in the figure .

Fig 3. Voltage variation on luminance


If the working range of the applied voltage is between VC and VD as shown in figure 3,
obviously, voltage B to B" would give a variation around targeted value B, but would
minimally affect the intensity of the luminance. Voltage B is attractive, since small
fluctuations in the applied voltage ( B to B" ), will have no significant effects on the
quality of the luminance as perceived by the user. So, through the parameter design
process, the investigator is able to choose a parameter, which is least influenced by the
variation factors.
Thus, the performance characteristics of a product can be affected by two factors, namely
design parameters, and sources of noise. The design parameters are those nominal
product values which are selected by the engineers. Sources of noise are those variables
that cause the deviations of actual nominal product value. The objective of parameter
design experiments are to identify the settings of the design parameters, at which the
noise factor influence is at its minimum. The detail analysis on the parameter design
experiments can be found in reference [1].
In summary, the following points were investigated:

The concept of poor quality from the Taguchi's perspective.


The types of variations possible.
Taguchi's quality characteristics.
The quality characteristics evaluation criteria.
Taguchi's quality strategy in the product design process.

Conclusions
This article has investigated the Taguchi philosophy in quality control. And discussed the
difference between Taguchi's new concepts, in quality control, and the traditional
goalpost philosophy. It has shown that in Taguchi's view, the measurement of the quality
of a manufactured product, is the total loss generated by that product to the society.
Taguchi defines the quality control of a given product as "achieving the targeted value
and minimising the variability around the target value", instead of "achieving
conformance to the specification".
The Taguchi quality strategies discussed are derived from several experiment techniques
used in the product design stage to implement the quality concepts into the product.
Taguchi proposed the idea of brainstorming to define the processes and factors which
create the product, followed by a series of experimental implementation, outlined by
Taguchi's quality concepts, to determine the optimum parameters to be implemented in
the parameter design stage of the product.
Taguchi suggested the use of Orthogonal Arrays(OA), in the experiment implementation
stage, to investigate and predict noise factors which might affect the quality of a given
product during the product manufacturing phase. Through OA experiment analysis, the
quality influencing factors of a product can then be identified, controlled, and hence
compensated during early product design stage.

Glossary Terms
Total Loss Function
The loss function is a function of deviation from the ideal value of a given design
parameter.
Orthogonal Arrays
An important analysis tool in designing quality into the design process.

Parameter Design
The approach whereby the influential factors are adjusted to control the variation
of other factors.

REFERENCES.
[1] Howard Gitlow, Shelly Gitlow, Alan Oppenheim and Rosa Oppenheim.1989. "Tools And Methods
For The Improvement Of Quality". Von Hoffman Press, Inc. Boston.
[2] Claire G. Meisonheime. 1992 "Improving Quality : A Guide To Effective Programs". Aspen
Publishers, Inc. Gaitherburg. .
[3] A.V. Feigenbaum.1991 "Total Quality Control". 3rd edition. Mcgraw Hill, Inc. New York.
[4] Ranjit Roy. "A Primer On The Taguchi Method". Van Nostrand Reinhold. New York. 1990.
[5] Thomas B. Barker. 1990 "Engineering Quality By Design : Interpreting The Taguchi Approach".
Marcel Dekker, Inc. New York. .Figure 5.1.0(14), 5.1.1(16), 5.1.2(13)
[6] Genichi Taguchi.1988 "Introduction To Quality Engineering : Designing Quality Into Products And
Processes". Asian Productivity Organisation. Tokyo. .
[7] R. Kacker. 1986 "Taguchi's Quality Philosophy : Analysis And Commentary". Quality Progress.
December. PP 21-29.
[8] Kwok-Leung Tsui, 1988, "Strategies for Planning Experiments using Orthogonal Arrays and
Confounding Tables", Quality and Reliability Engineering International, Vol. 4. pp 113-122.
[9] K. Dehnad, 1989 "Quality Control, Robust Design, and the Taguchi method" , Wadsworth &
Brooks/Cole, California.
[10] 1988"Introduction to Taguchi methods", Engineering, Jan.
[11] G. Taguchi,1988, "Introduction to Quality Engineering: Designing Quality into Products and
Processes", Asian Productivity Organisation, Japan.
[12]T.B. Barker, 1990 "Engineering Quality by Design", Marcel Dekker, Inc. N.Y.
[13] R. Roy, 1990"A primer on the Taguchi Method", Van Nostrand Reinhold, N.Y.
[14] C.Maynard,1995 "Quality Engineering", Curtin University Handout Notes for Computer Aided
Engineering.

Authors
The Authors of this paper are all currently studying at Curtin University Of Technology
in Western Australia. We are Final Years Students doing the Information and
Communications Engineering Degree.
Article assembled and placed on web by

Katleen.F.Ritchie.
Research for article provided by
Yeow Nam Ng
Don Black
Khanh Luu
Latest Revision done by :
Hartono Tedjokusumo (2nd yr Computer Technology)

Taguchi Home Page

Potrebbero piacerti anche