COST OF QUALITY AND

TAGUCHI LOSS
FUNCTION
Submitted by
Anup kumar
COST OF QUALITY AND TAGUCHI LOSS FUNCTION
Abstract: An organization must take into account the costs associated with achieving quality because
one of the important objective of continuous improvement programs is to meet customer requirements
at the lowest possible cost. Hence, an organization needs to adopt framework to classify costs and to
focus on eisting cost of quality !"o#$ models. %he objective of this paper is to give a survey of
research literature and models on the topic of "o# and to provide a basic understanding of quality
costs.
Keywords: #uality "osts , "ost of &oor #uality, "ot of 'ood #uality, %he %aguchi #uality (oss
)unction.
1. INTRODUCTION
*any organizations consider improving quality as the best way to enhance satisfaction, to reduce
manufacturing costs and to increase productivity. )or this, the "o# must be reduced. All quality
management consultants tend to have quality cost programs as an integral part of their repertoire
! +uhansa ,odchua, -../$. *onitoring and controlling "o# are becoming critical activities of quality
improvement programs. *anufacturing companies tend to measure visible costs and ignore
significant hidden costs that are difficult to measure such as opportunity costs. An approach for
quantifying the opportunity costs is presented here.
2. Literature review
%hree noted authors on quality management, 0eming, "rosby and 1uran each have a different
attitude to cost of quality !"o#$. 0eming2s view is that cost analysis for quality is a misguided waste of
time and measuring quality costs to seek optimum defect levels is evidence of failure to understand
the problem. "rosby argues that quality costs need to be measured, not for management control, but
for the development of 3quality4 thinking within the organization. %he more popular approach is that of
1uran who advocates the measurement of costs on a periodic basis as a management control tool.
%o collect quality costs a firm needs to adopt a framework to classify costs. %he )eigenbaum
classification system is almost universally accepted !&lunkett 5 0ale, 6788$. )eigenbaum!679:$
identified three cost categories : prevention , appraisal and failure. 1uran divided the failure costs into
two categories: internal failure and eternal failure ! 1uran 5 'ryna, 679.$. *any writers such as
*orse and ,oth !678;$, &onemon !677.$, <oude !677-$, ,oss!677;$ have used 1uran2s classification
system in their research.
%o further facilitate understanding of cost of quality trends over time, the four cost
categories are analyzed and presented in )igure 6
)igure 6: Hypothetical #uality "osts %rends over %ime
2.1. THE COQ MODELS
=n general, "># models are classified into four groups:
1. &?A?) models: &revention costs@ Appraisal costs@ )ailure costs
2. "rosby2s model: "ost of conformance@ "ost of non?conformance
3. >pportunity or intangible cost models: A&revention costs@ Appraisal costs @
)ailure costs @ >pportunity costsBC A"ost of conformance@ "ost of non?
conformance@ >pportunity costsB C A%angibles @ intangiblesB C A&?A?) !failure costs
includes opportunity costs$B
4. &rocess cost models: "ost of conformance @ "ost of non?conformance
2.2. COST O QUALIT! CATE"ORIES
%he cost of quality can be separated into three categories , namely: the cost of
conformance to customer requirementsD the cost of non?conformance to customer
requirementsD basic operational costs.
0efining each element in turn, the cost of conformance is the cost an organization
incurs in meeting the requirements of its customer. A strong element of this cost is
the money a company spends on preventing products or services going wrong or
checking that they are right before they reach the customer. %he costs of non?
conformance are failure costs , the costs incurred by a company in repairing what
has gone wrong. %he third category basic operational costs are the costs an
organization cannot avoid encountering during the normal performance of its
business.
%hose categories are represented in )igure -.
)igure -: "ost of quality categories
2.3. COST O #OOR QUALIT! AND COST O "OOD QUALIT!
%here are two main elements to focus on : the "ost of 'ood #uality !"o'#$ and
the "ost of &oor #uality !"o&#$. %he "o'# relates to costs incurred to assure the
quality of products and prevent poor quality. %he "o'# should be viewed as an
investment in reducing the "o&#, whereas the "o&# can be viewed as a direct
measurement of the failure costs incurred in producing a product. %ogether, these
two variables make up the cost of quality.
CoQ = CoGQ + CoPQ
%he "o&# measures internal and eternal failure costs and the "o'# measures
appraisal and prevention costs.
%he "o&# quantifies traditional quality costs companies measure and includes
rework, returned materials, scrap, and originate both internally and eternally.
Hence, the "o&# variable is divided into two sub?variables: =nternal )ailure "osts
!=)"$ and Eternal )ailure "osts !E)"$.
CoPQ = IFC + EFC
>n the other end of the spectrum, the "ost of 'ood #uality !"o'#$ measures
costs related to areas such as people and technology involved in the production of
high quality products.
%hese costs are much less likely to be accounted for by eecutives and plant
managers in respect to the cost of quality equation. %hey are often measured
disparately rather than in a standardized way that can facilitate enterprise? wide
comparison. %he "o'# has two central components: Appraisal "osts !A"$ and
&revention "osts !&"$.
CoGQ = AC + PC
2.4. THE TA"UCHI QUALIT! LOSS UNCTION $QL% AND THE
HIDDEN COSTS O QUALIT!
%aguchi *ethods was developed by 0r 'enichi %aguchi. =t combined engineering
and statistical methods that achieve rapid improvements in cost and quality by
optimizing product design and manufacturing processes. %here are three
statements that apply for the methods:
we cannot reduce cost without affecting qualityD
we can improve quality without increasing costD we can
reduce cost by reducing variation or by improving quality.
%herefore, when we do so, performance and quality will
automatically improve.
%aguchi defined quality as 3the loss imparted to society from the time the product is
shipped.4 )undamental to this approach to quality engineering is this concept of
loss. He associated loss with every product that meets the customer2s hand. %his
loss include, among other things , consumer dissatisfaction, added warranty costs
to the producer, and loss due to a company2s bad reputation , which leads to
eventual loss of market share.
#uality costs or poor quality costs are usually quantified in terms of scrap and
rework, warranty, or other tangible costs.
Fhat about the hidden costs or long?term losses related to engineering,
management time, inventory, customer dissatisfaction, and lost market shareG "an
we quantify theseG &erhaps, but not accurately. =ndeed we must find a way to
approimate these hidden and long?term losses, because they are the largest
contributors to total quality loss. %aguchi *ethods uses the #uality (oss )unction
!#()$ for this purpose. #() depends on the type of quality characteristic involved
like:
o Hominal?the?best !achieving a desired target value with minima
variation: dimension and output voltage$
o +maller?the?better !minimizing a response: shrinkage and wear$
o (arger?the?better !maimizing a response: pull?off force and tensile
strength$ Attribute !classifying andCor counting data: appearance$
o 0ynamic !response varies depending on input: speed of a fan
drive should vary depending on the engine temperature$
(oss can occur not only when a product is outside the specifications, but also when
a product falls with specifications. )urther, it is reasonable to believe that loss
continually increases as a product deviates further from the target value, as the
parabola !#()$ as shown in )igure ;.
)igure ;: %he %aguchi #uality (oss )unction !#()$
1482