Académique Documents
Professionnel Documents
Culture Documents
C H A P T E R S I X T E E N
The Management
OPERATIONAL CONTROL
You cant turn quality on like a spigot. Its a culture, a lifestyle within a company.
A Ford Engineer
For decades, management experts in the United States, including W. Edwards Deming and
J. M. Juran, urged manufacturers to design in quality at the beginning of the process, not to
inspect-in quality at the end of the production line. The quality call-to-arms mainly fell on
deaf ears in the United States, but not in Japan. More than 40 years ago, Juran predicted that
a focus on quality would help turn Japan into an economic powerhouse.
Jurans prediction proved true.1 In the late 1970s and the early 1980s, many U.S. rms
had a rude awakening. Many U.S. executives realized, for the rst time, that Made in the
U.S.A. no longer stood for the best that was available. Once a term of mockery, Made in
Japan became a term synonymous with quality. U.S. executives, especially those working
for rms employing traditional management techniques that had paid off so well a scant 20
years earlier, found themselves searching frantically for answers and desperately seeking to
remain competitive.
U.S. auto manufacturers realized in the late 1970s that Japanese auto manufacturers were
somehow able to sell automobiles that performed better, had far fewer defects, and cost less
than those made in the United States and still earn high returns. Likewise, when Hewlett-
Packard tested the quality of more than 300,000 new computer chips, it found those made by
Japanese manufacturers had zero defects per thousand. Those made by U.S. manufacturers
had 11 to 19 defects per thousand. After 1,000 hours of use, the failure rate of U.S. chips was
27 times higher than those of the Japanese chips. Many industry and government leaders in
the United States saw the handwriting on the wall: Get quality or lose the race.
The world had changed. Global competition gave consumers abundant choices and they
became more cost and value conscious, demanding high-quality products and services.
Firms that failed to pay attention to quality often found eroding market shares and operating
prots.
1
N. Gross, M. Stepanek, O. Port, and J. Carey, Will Bugs Eat Up the U.S. Lead in Software? BusinessWeek, December 6, 1999.
648
BlocherStoutCokinsChen: IV. Operational Control 16. The Management and The McGrawHill
Cost Management: A Control of Quality Companies, 2008
Strategic Emphasis, Fourth
Edition
For decades, ve-star hotels and restaurants have had consumers quality. For example, in Oklahoma (the rst state to set up a rating
lining up to get in. Now comes a new consumer rating: ve-star child- system) close to 60 percent of all child-care slots in the state are in
care. Just as if they were restaurants or hotels, child-care concerns facilities rated in the top two tiers, up from 30 percent in 2003. In Ten-
(both childcare centers and family child-care homes) are being nessee, where provider participation in star ratings is mandatory, 50
assigned star ratings by state regulators. These ratings are fast be- percent of facilities have earned a top rating, up from 30 percent in
coming the linchpin of states drive to raise child-care quality. The 2002. Critics argue, however, that although participation by child-care
ratings systems evaluate facilities on such criteria as low childadult providers is growing, the systems are mostly voluntary; provider par-
ratios, teacher credentials, curriculum, group size, and the safety and ticipation ranges from 10 percent to 60 percent in states where the
richness of the environment. Some of these criteria have in research systems are voluntary.
studies been associated with better outcomes in children. There is Source: S. Shellenbarger, Finding Five-Star Child-Care: States Rate Facilities
some preliminary evidence that the ratings systems are improving in Effort to Boost Quality, The Wall Street Journal (March 23, 2006), p. D1.
Eco-friendly, or green, buildings are one of the most talked-about Green certication is not cheap: costs can run anywhere from
trends in the trillion-dollar U.S. construction industry. Environmental $30,000 to $150,000 for administration and paperwork. Critics argue
quality concerns regarding new-building construction are important: that the existing standards are too lenient and that the scoring system
buildings today account for about one-third of U.S. energy consump- does not give differential weights to what are considered more criti-
tion, 30 percent of greenhouse gas emissions, and 30 percent of raw cal performance criteria.
material use. The U.S. Green Building Council (www.usgbc.org), a BRE, British Research Establishment Limited (www.bre.co.uk) is a
private environmental organization, now provides different levels of U.K. counterpart organization that, among other things, assesses and
green certication for new-building construction, based on six crite- certies both new and existing buildings using the BRE Environmental
ria and the use of a 69-point rating scale. Assessment Method (BREEAM) (see www.breeam.org). This method
The six evaluation criteria (i.e., green categories) are: sustainable is considered in the U.K.s construction and property sectors as the
sites (e.g., public transportation access); water efciency, energy measure of best practice in environmental design and management.
and atmosphere; materials and resources (e.g., use of materials with As worldwide demand for natural resources continues, the man-
post-consumer recycled contents); indoor environmental quality (e.g., agement and control of environmental quality costs will likely take on
carbon dioxide monitoring); and innovation and design process. In ad- increased importance, both in the United States and abroad.
dition to basic certication, higher-performance designations (silver, Source: A. Frangos, Is It Too Easy Being Green?, The Wall Street Journal
gold, and platinum) are awarded. (October 19, 2005), pp. B1, B6.
place that ensure that outputs of the organization satisfy customer quality requirements. Fur-
ther, these standards are intended to apply to all types of businesses, including services such
as transportation, health-care, and banking.
ISO 14000 ISO 14000 is a set of standards that relate to environmental management, that is, what an
is a set of quality standards organization does to minimize harmful effects to the environment. As with ISO 9000, ISO
designed to minimize 14000 is concerned with quality managementprocesses in place that ensure a product will
environmental effects of an have the least harmful impact on the environment, at any stage of its life cycle, either by pol-
organizations outputs. lution or by depleting natural resources.
In sum, ISO standards contribute to making the development, manufacturing, and supply
of products and services more efcient, safer, and cleaner. They make trade between countries
easier and fairer. They provide governments with a technical base for health, safety, and en-
vironmental legislation and they aid in transferring technology to developing countries. ISO
standards also serve to safeguard consumers, and users in general, of products and services
as well as to make their lives simpler. As of this writing, more than 700,000 organizations in
154 countries have implemented ISO 9000 and ISO 14000 standards (see www.iso.ch).
Lower
Higher Higher Faster
Warranty
Turnover Prices Delivery
and
Service
Costs
Higher
Increased
Market
Revenues
Share
Financial Performance
rm has more reliable manufacturing processes and schedules. Improved product quality also
lowers manufacturing costs as the rm reduces or eliminates rework and increases productiv-
ity. Customers are likely to perceive quality products as having higher values, which allows
the rm to command higher prices and enjoy a larger market share. Higher prices and great-
er market shares increase revenues and prots. Improved quality also decreases cycle time.
Faster cycle times speed deliveries, and prompt delivery makes happy customers, creates new
demand, and increases market shares. Higher revenues and lower costs boost net income and
increase the rms return on investment (ROI).
For most years since 1995, the hypothetical Baldrige Stock Index, consisting of publicly
traded U.S. companies that have received the Malcolm Baldrige National Quality Award, has
outperformed the Standard & Poors 500 by a margin of almost three to one. In a series of
papers,8 Hendricks and Singhal compared the performance of 600 quality award-winning com-
panies, including the Baldrige, state (e.g., the Georgia Oglethorpe Award), and other quality
award programs, with the performance of a control group of companies. These researchers found
that the award-winning companies signicantly outperformed the control group in many aspects
of their business, including the value of their common stock, operating income, sales, return on
sales, and asset growth. Saccomano9 reports that companies with effective TQM programs have
higher stock prices, sales, and prots compared to a control sample of rms.
In sum, cost, quality, and time are among the critical factors in successful strategies. Hav-
ing quality products allows rms that compete on differentiation to be effective in sustaining
their strategy. A rm with low costs and quality products provides its customers with products
equal to or better in quality at lower prices. Only with quality products can the rm truly be a
cost leader. Continual improvements in the quality of products and services and in processes
should be a fundamental strategic objective and a major item in the balanced scorecard of
most rms and organizations.
Chapter Preview
In the next section of this chapter, we dene the term quality and then present a conceptual
framework for managing and controlling quality costs. This is followed by a discussion of
nancial performance measures related to quality (relevant cost analysis and cost of qual-
ity [COQ] reports). We then discuss the role of nonnancial quality indicators in the overall
framework. We conclude the chapter with a discussion of a number of techniques that can be
used to identify and analyze quality-related problems.
8
K. B. Hendricks and V. R Singhal, Does Implementing an Effective TQM Program Actually Improve Operating Performance:
Empirical Evidence from Firms That Have Won Quality Awards, Management Science 43 (1997), pp. 12581274; K. B.
Hendricks and V. R Singhal, Firm Characteristics, Total Quality Management, and Financial Performance, Journal of Operations
Management 19 (2001), pp. 269285; and, K. B. Hendricks and V. R Singhal, The Long-Run Stock Price Performance of Firms
with Effective TQM Programs as Proxied by Quality Award Winners, Management Science 47 (2001), pp. 359368.
9
A. Saccomano, TQM Works Over Time, Trafc World, 1998, p. 37.
BlocherStoutCokinsChen: IV. Operational Control 16. The Management and The McGrawHill
Cost Management: A Control of Quality Companies, 2008
Strategic Emphasis, Fourth
Edition
As noted earlier, some organizations have a quality orientation and Boston Scientic Corporation recently reached an agreement
embrace managerial initiatives such as TQM to support this competi- with the U.S. Food and Drug Administration (FDA) in which
tive strategy. For each of the following examples, consider (1) which the company committed itself to an aggressive timeline for
nonnancial performance indicators, or controls, might be instituted resolving quality-control problems. Prior to this agreement, the
to help control quality and (2) what kinds of quality-related costs might FDA had announced that it would withhold approval of some
be involved by failing to control quality: new products from the company until it resolved the issues.
The FDA alleged that the company had failed to report, or
A recent study published in the November 15, 2005, issue of Can-
delayed reporting, potential safety problems associated with its
cer (a journal of the American Cancer Society) underscores the
products.
difculty of improving screening rates to detect colon cancer,
the third leading cause of cancer deaths.* Based on a review of PeopleSoft, Incorporated, reached an agreement to pay Cleveland
patient charts from individuals associated with a California HMO, State University $4.25 million to settle a lawsuit over computer
fewer than 30 percent of eligible patients over age 50 received any problems that delayed nancial aid to thousands of students. The
of the three types of colon-cancer tests. According to the National university claimed that students often waited months for nancial
Committee for Quality Assurance, a Washington-based nonprot aid because of computer problems that also hindered other ser-
organization that promotes health-care quality, Tufts Health Plan vices for more than two years.
(Waltham, MA) achieved the highest score in the nation, 72 per-
Sources:
cent, for colorectal cancer screening. *
R. L. Rundle, Colon-Cancer Screening Rates Rise Only Slightly, Study Says,
UnumProvident Corporation, a disability-income insurer, paid an The Wall Street Journal (October 11, 2005), p. B1.
$8 million civil penalty and $600,000 court costs to settle a suit D. Gullapalli, UnumProvident Is Set to Pay $8 Million Penalty in California,
brought against the company by the California Department of In- The Wall Street Journal (October 3, 2005), p. C3.
Boston Scientic Sets to Fix Quality Issues, The Wall Street Journal
surance, to resolve allegations that it cheated policyholders by (February 4, 2006), p. A2.
improperly denying claims. This settlement followed an earlier
Software Firm Will Pay CSU $4.25M Settlement, The Wall Street Journal
ne of $15 million paid by the company to the U.S. Labor Depart- (February 4, 2006), p. A2.
ment in a multistate settlement.
Quality For purposes of discussion we dene the term quality to mean the total level of customer satis-
is dened as customer faction with the organizations product or service. Dened in this manner, we can decompose the
satisfaction with the total notion of quality into two broad components: features and performance. The former component
experience of a product or refers to whether the characteristics, attributes, or functionality of the product or service is com-
service, that is, the difference
patible with customer expectationsin short, design quality. Outputs that fail to meet such ex-
between customer desires and
pectations result in quality-of-design failure costs. Conceptually, you can think of design failure
actual performance of the
product or service. as the difference between the actual features of the product (or service) and what the customer
wants. Such failures represent one component of total quality cost. One way to manage (i.e., re-
Design quality duce) design failure is through the use of target-costing procedures, as discussed in Chapter 10.
is the difference between In this chapter, we are concerned with the management and control of the other broad com-
customer desires (for attributes, ponent of quality, performance quality. Performance quality can be dened as the difference
services, functionality, etc.) and
between the design specications of the product and the actual performance of the product.
product design.
Thus, a personal computer whose electronic mouse consistently malfunctions or whose oper-
Performance quality ating system constantly locks up relates to what can be called conformance quality failures. As
is the difference between such, we dene performance quality costs as those related to providing a customers required
actual performance and design level of product or service performance.
specications. Not all customers have the same expectations for a product or service. All 3/8-inch drill bits
can drill 3/8-inch holes. Nevertheless, a rm can manufacture a 3/8-inch drill bit that costs $3
for home use and an industrial-strength drill bit that costs $15. The specications and quality
expectations for the less expensive drill bit are not the same as those for the more expensive
one. The industrial strength drill bit is designed for heavy, continuous use and can be used for,
say, 100 hours before it needs to be replaced. A drill bit for home use, on the other hand, is not
designed for continuous use for long hours and has a shorter expected life of, say, 10 hours.
Expectations for services also differ. A tourist does not expect the same services from a Motel
6 as from a Ritz-Carlton Hotel, although both provide rooms for tourists. A mechanic performs
quality service by changing a cars oil as specied: draining old oil, installing a new oil lter,
lubricating the chassis, and adding clean new oil. The service is a quality service even if the me-
chanic used a regular oil, not a new synthesized oil that improves engine performance, if the cus-
tomer asked for a regular, not a deluxe, oil change. The mechanic has failed to deliver a quality
BlocherStoutCokinsChen: IV. Operational Control 16. The Management and The McGrawHill
Cost Management: A Control of Quality Companies, 2008
Strategic Emphasis, Fourth
Edition
service, however, if the new oil lter falls off the next morning due to improper installation or if
the rell is four or six quarts of oil instead of the ve quarts specied by the manufacturer.
Specifications for
External Suppliers
EXHIBIT 16.3
Customer
Comprehensive Framework Expectations Satisfied
for Managing and Controlling Customers
Quality
Deliver
Set Quality- Perform Work/ Product /Service and
Work
Related Goals Monitor Output/ Monitor Customer
Processes
(i.e., Strategy) Correct Defects Satisfaction
Dissatisfied
Nonfinancial Customers
Taguchi Loss Quality Indicators
Functions,
Six Sigma Statistical
Programs Quality Control
and Run Charts
Thus, in the comprehensive model shown in, Exhibit 16.3, we depict consumer expectations
as the cornerstone of the entire framework. In this sense, then, the model can be viewed as
customer-based. As well, the model attempts to capture (as external failure costs) various
costs associated with dissatised customers.
Financial Component
You will notice that the reporting of quality cost information is a key element of the compre-
hensive framework shown in Exhibit 16.3. In fact, we depict cost information in four separate
categories to give prominence to the different types of quality costs that organizations incur.
This nancial approach to the management and control of quality, known as cost of quality, is
dealt with in greater detail later in the chapter.
Feedback Loops
You will notice that the comprehensive framework illustrated in Exhibit 16.3 contains a number
of feedback loops, designed to inform future decisions and to support an organizations overall
goal of continuous improvement. Thus, for example, the entire model continually helps the
organization better understand customer expectations and, in turn, set appropriate quality
goals for the organization.
In recent years, many major pharmaceutical companies have discov- people. The cultural shift to Six Sigma allows companies to get
ered the benets of using Six Sigma principles to eliminate manu- their employees more engaged. Tying rewards to accomplish-
facturing process variation, defects, and inefciencies. A smaller ments is particularly important to instituting such a culture change.
number of such companies are applying Six Sigma to Research and Competitive advantage: early adopters of Six Sigma in the pharma-
Development (R&D), in addition to the manufacturing function. Some ceutical industry stand to gain competitive advantage. Tradition-
aggressive companies, however, are applying the concept to func- ally, cost-cutting and eliminating process variation (two targets of
tions across the entire value chain of activities. Among the benets Six Sigma) have not been widely embraced in the industry. Thus,
cited by pharmaceutical companies regarding Six Sigma are the early adopters of this approach can gain at least temporary com-
following: petitive advantage in an increasingly competitive environment.
Changing economics of the industry: the Medicare Modernization For Six Sigma to work, most consultants believe that top manage-
Act (January 2006) will likely motivate increased use of generic ment support and commitment are keythat is, that Six Sigma can be
equivalents. For companies that have a thin pipeline of new drugs used as a leadership tool. In order to change the culture of an organiza-
or major drugs going off patent, the only way to enhance prot- tion to support Six Sigma, signicant personnel training costs are likely.
ability (at least in the short run) is to focus on cost controls and Still the nancial return of such implementations can be signicant. For
process efciencies, both of which are supported by the use of example, Eli Lilly estimates that its cumulative benet to date from the
Six Sigma. use of Six Sigma, over 160 projects, is approximately $250 million.
Maximizing employee value: the biggest asset for knowledge- Source: N. DAmore, Six Sigma Adds Up for Pharma, MedAdNews 25, no. 2
based organizations, such as pharmaceutical companies, is (February 1, 2006), p. 18.
functional teams, more or less on a consulting project basis. In the design stage of the project,
the Six Sigma team denes the problem and the scope of the problem (i.e., species the deliv-
erables of the project). In the measure stage, the team collects relevant process performance
data. In the analyze stage, the team tries to uncover root causes of an underlying quality
problem. This is followed by the improve stage, in which proposed solutions to the underlying
problem(s) are generated and then implemented. Finally, in the control stage of the project,
appropriate controls are put in place to ensure that the identied problem does not recur.
Motorola, Inc. pioneered the concept of Six Sigma as a structured approach for assessing
and improving both product and service quality. Today, this approach has gained notoriety and
credibility because of its adoption by rms such as Allied Signal and General Electric. The
term Six Sigma actually comes from statistics: in a normal distribution, the area outside of
+/ six standard deviations from the mean is very small. From a control standpoint, we can
express this area in terms of relative number of defects. One interpretation of a Six Sigma
quality expectation is approximately 3.4 defects per million items produced.15
The move from, say, a 3-sigma to a 6-sigma quality level is dramatic. For example, sup-
pose your bank tracks the number of errors associated with checks written on the bank by its
customers. If the bank nds, say, 12 errors per 1,000 checks processed, this is equivalent to an
error rate of 12,000 per millionsomewhere between 3.5 and 4 sigma levels! As Evans and
Lindsay point out,16 a change from 3 to 4 sigma represents a 10-fold improvement in quality; a
change from 4 to 5 sigma, a 30-fold improvement; and a change from 5 to 6 sigma, a 70-fold
improvement. For this reason, Six Sigma is not likely the goal for all processes and operations.
The appropriate quality expectation is a function of the strategic importance of the process and
the anticipated costs of taking the process to a higher level of quality.
Many organizations today are using Six Sigma principles to improve be dened as a transaction invoiced at a price lower than the one
manufacturing efciency and to lower costs. Others are using Six Pricing had approved); Measure (the team developed a map of the
Sigma to improve service processes. Sodhi and Sodhi (2005) provide pricing process, which included six sequential steps; in theory, the
a recent example of a global manufacturer of industrial equipment process was straightforward, but in practice shortcuts were often
that applied Six Sigma rigor to increase revenues. taken and the quality of information available at various steps was
The company in question offers a diverse product line, with many deemed decient); Analyze (the team used a cause-and-effect matrix
products manufactured to customer specication. Each sale, there- at each of the six steps to depict possible causes for lack of control);
fore, has its own individually approved discount and hence its own Improve (the goal here was to decrease the number of unapproved
invoiced price. With tens of thousands of sales transactions per year, prices without creating an onerous approval process); and, Control
the task of making sure that each invoice accords with the list and (in the present case, the company set up a monthly review process to
approved prices is indeed daunting. ensure that the company was experiencing higher transaction prices,
The company had already experienced success in applying Six fewer pricing exceptions, and no loss of market share).
Sigma principles to its manufacturing operations. In fact, several The overall result? The original goal was to increase sales rev-
individuals within the company had earned Six Sigma certications enues by $500,000 for the year. In just six months, however, revenues
(Green Belt, Black Belt). The company then decided to apply, on a had increased by a whopping $5.8 million, most of which went directly
pilot basis, a Six Sigma approach to its price-setting process. to the bottom line. As such, the company is now rolling out Six Sigma
The project in question involved a cross-disciplinary team (IT, pricing across the entire organization.
sales, pricing, nance, and marketing) and ve Six Sigma steps, re- Source: M. S. Sodhi and N. S. Sodhi, Six Sigma Pricing, Harvard Business
ferred to as DMAIC: Dene (the team decided that a defect should Review (May 2005), pp. 135142.
EXHIBIT 16.4
Goalpost Conformance
Loss No Loss Loss
Total quality management (TQM) is a key strategic and operational issue has identied implementation guidelines that can assist managers in the
for most rms, as their customers continue to have higher expectations for process. Some rms such as General Electric (http://ge.com), Honeywell
product and service quality. Because it involves most if not all the activities (http://honeywell.com/), and Weyerhaeuser (http://weyerhaeuser.com/)
in the rm, the implementation of TQM is usually a complex and difcult take additional steps to ensure the success of their quality initiatives. What
process. The full implementation of TQM may take several years. The IMA do you think these additional steps might include?
EXHIBIT 16.5
Absolute Conformance
(Robust Quality Approach)
Loss Loss
.5
Target Value
18
Evans and Lindsay, The Management and Control of Quality, pp. 112113.
BlocherStoutCokinsChen: IV. Operational Control 16. The Management and The McGrawHill
Cost Management: A Control of Quality Companies, 2008
Strategic Emphasis, Fourth
Edition
EXHIBIT 16.6
Color Density of Sony TV Sets
Manufactured in the San Diego
Plant and a Japanese Plant
Japanese San Diego
Source: J. R. Evans and W. M. Lindsay, The plant plant
Management and Control of Quality, 6th ed.
(South-Western, 2005), p. 113.
The average quality cost (loss) per unit of the San Diego plant, however, was $0.89 higher
than that of the Japanese plant. One reason for the higher quality cost for units produced at the
San Diego plant was the need for more frequent eld service. Customers are more likely to
complain when the density is farther away from the target value. Although the plant in Tokyo
had a higher rejection rate, it experienced lower warranty and repair costs for its products.
For rms desiring to attain long-term protability and customer satisfaction, absolute confor-
mance is the better approach.
The extension of absolute performance standards to estimate Taguchi quality loss functions
is covered in Appendix A.
* Opportunity costs
BlocherStoutCokinsChen: IV. Operational Control 16. The Management and The McGrawHill
Cost Management: A Control of Quality Companies, 2008
Strategic Emphasis, Fourth
Edition
Product redesign and process improvement. Costs incurred to evaluate and improve
product designs and operating processes to simplify manufacturing processes or to reduce
or eliminate quality problems.
Quality circle Quality circles. Costs incurred to establish and operate quality control circles to identify
is a small group of employees quality problems and to offer solutions to improve the quality of products and services.
from the same work area that
meet regularly to identify and
solve work-related problems
Appraisal Costs
and to implement and monitor Appraisal (detection) costs are costs devoted to the measurement and analysis of data to
solutions to the problems. determine conformity of outputs to specications. These costs are incurred during produc-
tion and prior to deliveries to customers. Through measurement, analysis, and monitoring of
Appraisal (detection) costs
manufacturing processes and examination of products and services prior to delivery, rms
are expenditures devoted to the
measurement and analysis of
identify defective items and ensure that all units meet or exceed customer requirements.
data to determine conformity of Appraisal costs include the following:
outputs to specications. Test and inspection cost. Costs incurred to test and inspect incoming materials, work in
process, and nished goods, and the cost incurred to inspect machinery; also, eld-testing
of products at the site of the consumer.
Test equipment and instruments. Expenditures incurred to acquire, operate, or main-
tain facilities, software, machinery, and instruments for testing or appraising the quality of
products, services, or processes.
Ford Motor Company unveiled the 2001 model of its best-selling sport- When asked by nancial analysts to comment on the cost of delay
utility vehicle, the Ford Explorer, in late 2000. The 2001 model added a and repairing defects, Nasser responded, Pick a number. It is over
host of new safety features that enhanced the most popular SUV on $1 billion. The delay was expensive, but Ford executives say the cost
the market since its introduction a few years earlier. Ford expected of xing warranty claims later would be far higher. One defect caught
the new model to increase the rms market share and to add sub- by engineers was an internal steering-column switch that might have
stantial amounts to its bottom line. Yet, three months after the rede- led motorists to start the engine in the drive position. Left uncor-
signed Explorer began rolling off the assembly line not a single one of rected, this problem had the potential of resulting in big-time safety
the 5,000 built was in dealer showrooms. Instead, they were parked recalls. What was the root cause of the problem? It was traced to a
outside factories in St. Louis and Louisville while Ford engineers supplier who used too much solder on a $1 circuit board. When you
pored over them looking for defects. Jacques Nasser, CEO of the Ford get to the bottom of it, they are that trivial, says a company ofcial of
Motor Company, ordered factory managers to hold off on shipping the such glitches. But when you let them escape, they are just huge.
new Explorer until engineers had the opportunity to correct quality Source: N. Muller, Putting the Explorer under the Microscope, Business-
problems. Week, February 12, 2001, p. 40.
Ideally, each quality cost should have its own account so that quality cost information is
readily apparent, not buried in myriad accounts. These quality cost accounts are the source of
quality cost information.
Report Format
A report on cost of quality is useful only if its recipients understand, accept, and can use the
content of the report. COQ reports can be prepared in different ways. Each rm should se-
lect and design a reporting system that (1) can be integrated into its information system and
(2) promotes TQM. Among considerations in establishing a quality cost report system are
proper stratications of quality cost reports by product line, department, plant, or division, and
the time periods of the reports so that the rm can easily identify the origins of quality costs.
To facilitate assessment of the magnitude of quality costs and their impact, rms often express
cost of quality in percentages of net sales (or total operating costs) for the period.
A cost of quality matrix, as illustrated in Exhibit 16.8, is a convenient and useful tool in
reporting quality costs. With columns identifying functions or departments and rows delineat-
ing COQ categories, a cost of quality matrix enables each department, function, process, or
product line to identify and recognize the effects of its actions on the cost of quality and to
pinpoint areas of high-quality costs.
Design
Engineering Purchasing Production Finance Accounting Other Totals % of Sales
Prevention costs
Quality planning
Training
Other
Appraisal costs
Test and Inspect
Instruments
Other
Internal failure costs
Scrap
Rework
Other
External failure costs
Returns
Recalls
Other
Totals
19
Adapted from IMA Statement No. 4R.
BlocherStoutCokinsChen: IV. Operational Control 16. The Management and The McGrawHill
Cost Management: A Control of Quality Companies, 2008
Strategic Emphasis, Fourth
Edition
What type of nonnancial performance data are available to passen- quality for each airline on a timely basis, using objective, performance-
gers of U.S. domestic airlines? How can such airlines benchmark their based data.
operating performance in terms of critical success factors? Since AQRs for the top three (of 17) and bottom three domestic airlines
1991 such data are provided on an annual basis in what is called the for 2005 and 2004, as well as composite data (based on AQRs for 17
Airline Quality Rating (AQR) report. The AQR reports for each domestic domestic airlines) are as follows:
carrier monthly performance data in four major categories, based on
data obtained from the U.S. Department of Transportations monthly
Air Travel Consumer Report (www.dot.gov/airconsumer/). 2005 AQR 2004 AQR
The AQR is a summary performance measure calculated as a Score Rank Score Rank
weighted average of performance in four categories important to
consumers, as follows: Jet Blue 0.88 1 0.59 1
Air Tran 0.99 2 0.76 2
(+8.63 OT) + ( 8.03 DB) + ( 7.92 MB) + ( 7.17 CC) Independence Air 1.05 3 N/A
AQR = U.S. Airways 2.77 15 1.55 12
(8.63 + 8.03 + 7.92 + 7.17)
COMAIR 2.96 16 3.27 15
where Atlantic Southeast 4.68 17 4.10 16
OT = On-time arrival Industry average 1.73 1.38
DB = Denied boardings
MB = Mishandled baggage
CC = Customer complaints Source: B. D. Bowen and D. E. Headley, 2006 Airline Quality Rating (April
Respective weights = 8.63, 8.03, 7.92, and 7.17 2006), available at www.aqr.aero/aqrreports/AQR2006nal.pdf
The creators of the above model state that the AQR provides both
consumers and industry watchers a means for looking at comparative
the basis for improving operations. Because these measures relate to physical processes,
they focus attention on precise problem areas that need attention.
Such information is more timely than nancial measures of qualityin the extreme, these
measures of quality can be reported on a real-time basis (i.e., instantaneously as operations
occur). Nonnancial quality indicators provide immediate short-run feedback on whether
quality-improvement efforts have, in fact, succeeded in improving quality.
These nonnancial performance measures can be useful predictors (i.e., leading indicators)
of future nancial performance.
specication. The vertical measure has a specied allowable range of variations, which are
referred to as upper and lower limits, respectively. Exhibit 16.10 contains control charts for
manufacturing 1/8-inch drill bits in three workstations.
Assume that a rm has determined all drill bits must be within 0.0005 inch of the specied
diameter. All units from workstation A are within the specied range (0.0005), and no further
investigation is necessary. Three units from workstation B are outside the specied rangean
indication that not all operations in workstation B are in control. Management should investigate
the cause of the aberration to prevent further quality failures. Although all units manufactured
by workstation C are within the specied range acceptable to the rm, the control chart reveals
that quality characteristics of workstation C are moving upward. (Using it in this manner, the
A run chart control chart is often referred to as a run chart. A run chart shows the trend of observations over
shows trends in quality measures time.) Management may want to launch an investigation because the trend suggests that in the
over time. near future the operation will most likely produce drill bits outside the specied range.
When the central line and the limits in a control chart are determined through a statistical
process, the control chart is a statistical quality control (SQC) chart or statistical process
control (SPC) chart. The control charts presented in Exhibit 16.10 are SQC (or SPC) charts
if the line in the center, 0.125, is determined by calculating the arithmetic mean (, read mu)
of the observations and the limits, 0.1255 and 0.1245, are determined based on the standard
deviation (, read sigma) of the observations. For example, the standard deviation of the drill
bits is, say, 0.00025 and the rm has determined that variations within two standard devia-
tions are acceptable. Thus the limits are 2, or 0.125 2 0.00025, which are 0.1255
and 0.1245 for upper and lower limits, respectively.
A rm sets the upper and lower control limits based on experience, technology, customer
expectation, and cost and benet analysis that determine the extent of variations within which
the rm is willing to accept or tolerate. The purpose of a control chart is to distinguish between
random and nonrandom variations. A process is considered to be in statistical control if no
sample observation is outside the established limits. Variations that fall within the established
limits are deemed random variations so that no further investigation is needed. Observations
outside the limits may signal quality failures.
.125"
Units
WORKSTATION B
.125"
Units
WORKSTATION C
.125"
Units
BlocherStoutCokinsChen: IV. Operational Control 16. The Management and The McGrawHill
Cost Management: A Control of Quality Companies, 2008
Strategic Emphasis, Fourth
Edition
However, for observations within the established limits to be considered random, the ob-
servations have to show no apparent patterns or runs, with an approximately equal number
of observations above and below the center line and most points nearing the center line. A
process may be out of control if the observations show trends, cycles, clusters, or sudden shifts
hugging the center line or the control limits.
Control charts are useful in establishing the state of control and monitoring processes. Post-
ing control charts in a common area facilitates early detection of quality problems, promotes
awareness of workers on the quality status of their products or services, and encourages active
participation in efforts to raise quality.
Taking Corrective Action
Once control charts indicate that a process may be out of control, what techniques are avail-
able for diagnostic control purposes, that is, to guide corrective action? Histograms, Pareto
charts (diagrams), and cause-and-effect diagrams are useful techniques for diagnosing causes
of quality problems and identifying possible solutions to these problems.
Histogram
A histogram A histogram is a graphical representation of the frequency of attributes or events in a given set
is a graphical representation of of data. Patterns or variations that are often difcult to see in a set of numbers become clear in a
the frequency of attributes or histogram. Exhibit 16.11 contains a histogram of factors that contribute to the quality problems
events in a given set of data. identied by a rm that makes chocolate mousse.
The rm has experienced uneven quality in one line of its product. The rm identies six
contributing factors to the quality problem: substandard chocolate, improper liqueur mixture,
uneven egg size, uneven blending speed, variant blending time, and improper refrigeration
after production. It identied 210 batches as having poor quality. The histogram in Exhibit
16.11 suggests that variations in egg size may be the largest contributor to the quality problem,
followed by uneven speed in blending ingredients.
Pareto Diagram
A Pareto diagram
is a histogram of the frequency A Pareto diagram is a histogram of factors contributing to a specied quality problem, ordered
of factors contributing to a from the most to the least frequent. Joseph Juran observed in the 1950s that a few causes usu-
quality problem, ordered from ally account for most of the quality problems, thus the name Pareto.20 See the Pareto diagram
the most to the least frequent. of the chocolate mousse quality problem in Exhibit 16.12.
EXHIBIT 16.11 70
Histogram of Quality Problem:
Contributing Factors 60
50
Frequency
40
30
20
10
0
(1) (2) (3) (4) (5) (6)
Causes of Poor Quality
Key:
(1) Quality of chocolate (4) Blending speed
(2) Liqueur (5) Blending duration
(3) Egg size (6) Improper refrigeration
20
V. Pareto, a nineteenth-century Italian economist, observed that 80 percent of the wealth in Milan was owned by 20 percent of
its residents.
BlocherStoutCokinsChen: IV. Operational Control 16. The Management and The McGrawHill
Cost Management: A Control of Quality Companies, 2008
Strategic Emphasis, Fourth
Edition
A Pareto diagram not only discloses the frequency of factors associated with a quality
problem but also provides a useful visual aid. A Pareto diagram includes a curve that shows
the cumulative number of causes, as shown in Exhibit 16.12. Using a Pareto diagram, man-
agement can separate the few major causes of quality problems from the many trivial ones
and identify areas that contribute most to poor quality. Thus, management can focus its efforts
on areas that are likely to have the greatest impact on quality improvement. For example, the
cumulative line in Exhibit 16.12 shows that improper egg size and erratic blending speed
account for 110 quality problems in manufacturing chocolate mousse. To improve quality,
management would most likely demand that all suppliers deliver eggs uniform in size and that
operating personnel regulate the speed of blenders.
Cause-and-Effect Diagram
A cause-and-effect diagram The cause-and-effect, or sh-bone, diagram organizes a chain of causes and effects to sort
is used to identify potential out root causes of an identied quality problem. Karou Ishikawa discovered that for situa-
causes of quality problems. tions with myriad factors the number of factors that inuenced a process or contributed to
a quality problem were often overwhelming. He developed cause-and-effect diagrams as an
organizing aid.21
A cause-and-effect diagram consists of a spine, ribs, and bones. At the right end of the hori-
zontal spine is the quality problem at hand. The spine connects causes to the effect, the quality
problem. Each branch or rib pointing into the spine describes a main cause of the problem.
Bones pointing to each rib are contributing factors to the cause. In Exhibit 16.13 we illustrate
the general structure of a cause-and-effect diagram.
Typical main causes for quality problems in manufacturing operations are
Machines
Materials
Methods
Manpower
Some users refer to the four main categories as 4M.
140
Frequency
120
100
80
60
40
20
0
(1) (2) (3) (4) (5) (6)
21
K. Ishikawa, Guides to Quality Control, 2nd ed. (Tokyo: Asian Productivity Organization, 1986).
BlocherStoutCokinsChen: IV. Operational Control 16. The Management and The McGrawHill
Cost Management: A Control of Quality Companies, 2008
Strategic Emphasis, Fourth
Edition
Quality
Problem
Cause
In Exhibit 16.14 we show a cause-and-effect diagram for the quality problems in the manu-
facturing of chocolate mousse. The rm identied these main causes for the 20 percent rejec-
tion rate:
Machines
Equipment not properly calibrated
Timer functions erratically
Materials
Suppliers delivered wrong or irregular-size eggs
Low-quality chocolate
Wrong liqueur used
Methods
Improper refrigeration of ingredients
Ingredients not added at proper time or in prescribed sequence
Inappropriate preheating
Manpower
Hiring of new workers without proper experience and not giving adequate training
Workers failed to follow instructions
Many rms have found brainstorming an effective technique in constructing cause-and-effect
diagrams.
Advanced technology is generating quality gains that help U.S. drying of nished foam cores. The facility quickly achieved a 30 per-
manufacturers distinguish themselves. For example, by digitizing the cent efciency gain over its predecessor and boosted capacity by
control of its factory, privately owned Latex Foam International (LFI) 50 percent, all in a smaller space with less than two-thirds of the
boosted its capacity, productivity, and quality. At a cost of $35 mil- workforce. The manufacturing system lets LFI track every mattress,
lion, LFI built a state-of-the-art digital plant at Shelton, Connecticut. right to when robots prod them to test for rmness with numerical
LFIs engineers can monitor all the factorys operationsfrom the precision.
mixing of latex and the distribution of liquid rubber into molding beds Source: Adam Aston, The Flexible Factory, BusinessWeek, May 5, 2003, p. 91.
by mantis-like hanging robots to the heating, cooling, cleaning, and
Summary In todays global competition, with short product life-cycles and rapidly changing technolo-
gies and consumer tastes, rms can sustain long-term survival and protability only by manu-
facturing quality products and rendering quality services.
Providing quality is the best strategy for attaining long-term protability. Businesses offer-
ing quality products and services gain market shares over the years; studies show that quality
is positively related to nancial performance.
A quality product or service meets or exceeds customer expectations at a price customers
are willing to pay. To achieve quality products or services, many rms adopt total quality
management, which requires continuous efforts by everyone in an organization to understand,
meet, and exceed the expectations of both internal and external customers.
How can accounting add value to the organization by supporting quality-related initiatives
of management? We propose, in Exhibit 16.3, a comprehensive framework that can be used
to manage and control quality for a business. This framework begins, and ends, with the goal
of meeting customer expectations. That is, the framework implies an iterative or continu-
ous process. One primary role in this process for accounting is to provide relevant nancial
information. We identify two such examples: relevant cost (and revenue) data for evaluating
spending and investments in quality and the preparation of cost of quality (COQ) reports.
Such nancial information regarding quality is supplemented with internal and external non-
nancial measures of quality. To detect poor quality (i.e., out-of-control processes), these data
can be analyzed using run or control charts. Histograms, Pareto diagrams, and cause-and-
effect diagrams can then be used for diagnostic control purposes, that is, to identify the source
of quality problems in order to inform appropriate corrective action.
Management accountants, with training and expertise in analyzing, measuring, and report-
ing information, can help design and implement the type of comprehensive control system
depicted in Exhibit 16.3.
Appendix A