Achieving Customer Specifications Through Process Improv.docx

Achieving Customer
Specifications Through
Process Improvement
Using Six Sigma: Case Study
of NutriSoil – Portugal
AMÂNDIO PEREIRA BAÍA, POLYTECHNIC OF GUARDA –
PORTUGAL AND UDI –
RESEARCH UNIT FOR INLAND DEVELOPMENT OF
GUARDA
© 2015, ASQ
Tolerance limits are essential in production process
management, as they determine consumer satis-
faction. The use of statistical quality control tools
allows for process improvement and the cpk index
enables its measurement. Above all, the adoption
by businesses of lean tools has been crucial in
reducing the variation of a process or a product,
satisfying the consumer’s specifications, eliminat-
ing defects, reducing operating costs, and, in short,
increasing profitability.
The NutriSoil Company in Portugal, a small and
medium-sized enterprise (SME), sells fertilizer in
bags. The company has had problems with its filling
process due to excess weight of the bags. Results show
that by implementing Six Sigma combined with the 5S
program, NutriSoil achieved an improvement in its
cpk index for this process, which increased consumer

satisfaction and a highly significant cost savings. This
resulted in increased competitiveness.
Key words: cp and cpk capability index, process
capability, process improvement, Six Sigma, SME,
statistical quality control
INTRODUCTION
The basic objective of this study is to explore, using
a case study, the benefits of implementing the strat-
egy of Six Sigma combined with the 5S program in
NutriSoil, a Portuguese small and medium-sized enter-
prise (SME) struggling to retain profitability. NutriSoil
had high production costs, a situation that is common
to many SMEs.
The specifications or tolerance limits define the dif-
ference between acceptable and unacceptable products,
and producing within these limits is critical to con-
sumer satisfaction. The ability to consistently distribute
products within specifications determines whether the
supplier will continue to do business with the con-
sumer. A company can improve a production process
by efficiently coordinating the specifications and the
design process. Process capability measures how the
process meets specifications.
True process capability cannot be determined
until x-bar and R control charts have reached opti-
mum quality improvement without significant
investment in new equipment. A key aspect of process
improvement is to recognize that regardless of the
depth of this monitoring, there is always variation.
This variation is well defined when a process is sta-
tistically controlled. A modern definition of quality

48 QMJ VOL. 22, NO. 2/© 2015, ASQ
Achieving Customer Specifications Through Process
Improvement Using Six Sigma: Case Study of NutriSoil –
Portugal
www.asq.org 49
dispersion permitted, that is, with specification limits. It
does, however, have a flaw; it assumes that the process
average is centered in the range of specifications. In
fact, it is not always so.
On the other hand, the cpk index takes into account
the level of acceptable risk, the product variation,
and gives a quick overview of the actual process per-
formance. It measures the capability of the current
process. This measure is in line with the positioning of
the average case.
Sigma describes the variability of a process that
produces similar products or services (Dedhia 2005). A
quality level sigma provides an indicator of the occur-
rence frequency of defective items, whereby a higher
quality level indicates a process with less possibility
of creating defective items. Consequently, with the
increased level of sigma quality, the product reliability
improves the need for testing and inspection declines,
product cycle time decreases, costs are lower, and con-
sumer satisfaction is increased.
Six Sigma is a process capability condition defined
as the ability of a process to produce a good product.
It establishes a relationship between product specifi-

cations and process variability, measured in terms of
process capability indices: cp and cpk. A process that
operates the Six Sigma has a cp = 2 and a cpk = 1.5
(Kumar 2002). Six Sigma is an initiative that aims to
eliminate the defective items in any product, process,
or transaction.
Motorola was the first organization to use the term
“Six Sigma.” In 1992, it reduced defective item levels
by a factor of 150 (The History of Six Sigma 2007).
Honeywell began this program in the early 1990s and
is said to have saved more than 600 million euros
in 1999 (Pande, Neuman, and Cavanagh 2003).
A study from Lucier and Sheshadri (2001) shows that
General Electric was able to save $2 billion in three
years, after implementing Six Sigma in 1996 (Antony
and Banuelas 2002). Reports of Buss and Ivey (2001),
Feo and Bar-El (2002), McClusky (2000), and Weiner
(2004) also show the benefits gained by companies
such as Raytheon, Dow Chemical, DuPont, Texas
Instruments, Johnson and Johnson, Toshiba, Boeing,
and others.
states that “quality is inversely proportional to vari-
ability” (Montgomery 2009). It is recognized that
variability reduction directly results in lower costs.
Variability reduction means fewer repairs, fewer con-
sumer complaints, less rework, and reduced time
waste, all of which imply less effort and money spent
during the process.
Six Sigma is a powerful business strategy that
employs a disciplined approach to tackle process vari-
ability using a rigorous application of statistical and
nonstatistical tools and techniques (Hamon 2010;
Evans and Lindsay 2005). Six Sigma is an initiative

that aims to eliminate the defective items in any prod-
uct, process, or transaction.
Even though Six Sigma has proved to be effective in
many situations, other lean tools such as the 5S frame-
work also offer the ability for organizations to improve
rapidly. Lean thinking emphasizes maximizing value
through reduction of waste, variation, and overburden
within processes (Womack 2011). Lean promotes the
attitude of make it better now, make it perfect later
(Toussaint and Gerard 2010).
LITERATURE REVIEW
The cp and cpk process capability indices represent
the ability to combine people, machines, methods,
materials, and measurements to produce a product
that consistently meets the requirements or expecta-
tions of the consumer. The process capability indices
continue to be used as process tools, even though
there is “a growing recognition that these tools are
limited and, in particular, the capability standard
indices are appropriate only with measurements that
are independent and distributed in a reasonably nor-
mal manner” (Rodriguez 1992, 176). The popularity
of process capability indices, although in many cases
these indices are flawed tools, has led to ongoing
research in this area, which has recently been evalu-
ated by Wu, Pearn, and Kotz (2009).
However, a high cp value does not guarantee that
a production process is within the limits of the speci-
fications because the cp value does not imply that
the current state of the process coincides with the

Portugal
50 QMJ VOL. 22, NO. 2/© 2015, ASQ
systems. As far as metrics are concerned, when a pro-
cess is operating at a Six Sigma level, it produces
noncompliance (defective items or errors) at a rate of
not more than 3.4 defective items per million oppor-
tunities. As a methodology, Six Sigma leads to an
improvement of the business process focusing on the
understanding and management of customer expecta-
tions (Brewer and Eighme 2005; Rudisill and Clary
2004). As a management system, Six Sigma is used to
ensure that critical efforts for improvement developed
through the methodology and metrics are aligned with
the company’s business strategies. In the late 1990s,
about two-thirds of Fortune 500 organizations had
undertaken Six Sigma initiatives with the aim of reduc-
ing costs and achieving improvements.
Although Six Sigma proved to be efficient when
applied to big companies, few reports provide success
factors, guidelines, tools, and techniques for imple-
menting Six Sigma in the context of SMCs (Antony
2011; Antony, Kumar, and Madu 2005; Wessel and
Burcher 2004).
METHODOLOGY
cp and cpk Capability
Index and Six Sigma
In their study the authors used cp and the cpk as the
main metrics in this cycle of the improvement pro-
cess. If they compare the cp process metric with the cpk
metric, they are able to detect which is the opportunity

improvement that exists in the process. By reducing
process variation, the supplier increases the ability to
meet the specifications and also reduces the number of
products that do not meet the specifications. A cpk = 2
represents a Six Sigma level, with 3.4 defective items
per million opportunities, with quality costs less than
10 percent of sales and configures a benchmark of
“world class” (Mike 1998).
It should be noted that although a process pro-
duces a quality feature with a capability index
higher than 2.5, the needless precision makes it
very expensive. The process capability must be
Many of the success stories on the implementation
of Six Sigma belong to large organizations, particu-
larly multinationals. Very few publications refer to
the implementation of Six Sigma in SMEs. Six Sigma
has been criticized by people saying that it requires
a large investment and resource-intensive programs
that only large companies can provide (Calcutti
2001). However, an SME may have fewer complica-
tions than a large company in terms of company size,
the nature of its projects, effort for building teams,
and training employees, so it can be argued that
the implementation of Six Sigma in SMEs is easier,
except for the cost of investment.
The Six Sigma philosophy consists of manage-
ment by facts and not by opinions (Nanda and
Robinson 2011). Antony, Kumar, and Madu (2005)
found that Six Sigma provides executives and manag-
ers with the strategies, methods, tools, and techniques
to change their organizations. Many organizations
recognized that the Six Sigma methodology provides
a set of practices designed to improve the produc-

tion process, and the methodology quickly spreads to
different functional areas such as marketing, engi-
neering, purchasing, distribution, and administrative
support (Ray and Das 2009; 2010).
Magnusson, Kroslid, and Bergman (2003) found
that Six Sigma is a business process that allows com-
panies to dramatically change and improve their basic
organization through the design and monitoring of
daily business activities so as to minimize waste and
resources while consumer satisfaction is increased in
some of its components. Franken (2007) emphasizes
that when the internal operations of a company are
not well structured, this company will find it difficult
to create value and be highly competitive. Six Sigma
begins as a focus for improving internal operations,
but Zucker (2007) notes that the work of Six Sigma is
usually done through cross-functional teams to man-
age the project. The benefits are well documented for
manufacturing industries and growing service indus-
tries (Wright and Basu 2008).
In the last two decades Six Sigma has evolved
from a focus on metrics to a level of methodology and
finally to the design and development of management
Portugal
www.asq.org 51
500 and 1,000 kilograms (big bags)
and plastic bags of 20 kilograms.

All bags have a label containing
formula, lot, date, and other infor-
mation from the manufacturer.
This SME was established in
2005, employs 32 workers, and
is committed to developing and
producing fertilizer. The main
customers of NutriSoil are super-
markets and other stores geared
for farming and gardening. The
company annual ly produces
40 million tons of fertilizer. Socio-
environmental issues are extremely
important for NutriSoil; therefore,
the company’s involvement with
the environment is a constant con-
cern. The NutriSoil production process is presented in
summary form in Figure 1. All raw materials come
from domestic suppliers and arrive in bulk to NutriSoil
by road. The material is crushed by a shredding
machine and is sent for composting. Then the nutri-
ents and other supplies are blended and the fertilizer
goes to the silo where it is stored. Later the weighing is
done and the formulation is forwarded to the bagging
machines where it is packed.
NutriSoil operates in a highly competitive
market, where the cost-price factor is crucial. To
improve profitability, NutriSoil adopted a simplified
version of the Six Sigma process (see Figure 2) in an
attempt to reduce production costs. This sketch was
developed after meetings with the company owner
and with top and middle-level managers. Also, the
operators’ concerns were fully considered in the

definition of all procedures.
Define
A team was created consisting of operators, pro-
duction and quality engineers, the marketing
department, and the company owner. This team spent
many hours in the production area to collect data
and understand the production and packing mode
re-evaluated periodically to ensure that the process
average has not changed and that process variation
has not increased. The minimum recommendation
for re-evaluation is six months.
The NutriSoil Company implemented a Six Sigma
strategy to pursue its goal of reducing costs, which is
vital to its survival. The company sought to:
• Exert a continuous effort to achieve stable and pre-
dictable results of a process (reduction of process
variation), which is of vital importance to the suc-
cess of a business
• Recognize that production and business processes
have characteristics that can be measured, analyzed,
improved, and controlled
• Obtain commitment to improvement from the entire
organization, particularly from top management
Case Study –
NutriSoil Company
The name of the NutriSoil Company is fictional to
preserve confidential information in an industrial
context, but the case is real. NutriSoil is a company
that produces and sells organic fertilizer in bulk bags of

Figure 1 NutriSoil production process diagram
Green Waste
Destroying
Composting
Mixing
Silage
DispatchPacking
Big bags
30%
20 kg
50%
Bulk
20%
• Visual inspection
• Manual sorting
• Piles formation
• Temperature and
humidity control
• Revolving
NutriSoil
©
20

15
, A
SQ
Portugal
52 QMJ VOL. 22, NO. 2/© 2015, ASQ
the Pareto diagram to determine the importance of the
causes and the cause and effect diagram to detect the
causes of variation. Control charts for average (x-bar)
and the amplitude (R) were used to determine the
assignable causes of variation.
Improve
A program of organization and standardization, 5S
(see Figure 3), was gradually and systematically
implemented with a view to increasing the safety and
efficiency of labor and productivity.
The purpose of the 5S methodology is to improve
efficiency through proper disposal of materials
(separate what is necessary from the unnecessary),
organizing, cleaning and identification of materials
and spaces, and maintenance and improvement of
the 5S.
This methodology develops a systematic plan for
sorting, cleaning, and ordering, enabling as a result
greater productivity, security, organizational climate,

and motivation of employees, with the consequent
improvement of organizational profitability.
A program of buildings and equipment main-
tenance, total productive maintenance (TPM), was
of fertilizer bags. Team members were encouraged
to identify the vital characteristics of the production
process based on the “consumer voice.”
Measure
To determine the defective items that occurred in each
production phase, the team was divided into small
groups in order to identify the vital procedures of each
phase of the production process. A brainstorming ses-
sion was conducted to develop the data collection plan.
The data for defective products were collected and
analyzed in order to measure the current performance
in each of the workstations. Before the data collection
it was decided to validate the measurement system,
namely studying the measurement system’s contribu-
tion to the variation in the form of repetition (the same
product measured repeatedly by the same instrument)
and reproducibility in the process (the same product
measured by different operators).
Analyze
In the analysis of causes for defective items, a classifi-
cation of defective items and the respective contribution
to the total defective items was made: the authors used
Figure 2 Six Sigma process adopted by NutriSoil
De�ne
Measure

AnalyzeImprove
Control
Develop and execute
appropriate data
collection method
Find the root
causes of the
problem
Generate
and improve
solutions
Ensuring
the results
Understand the
problem and its
�nancial impact
©
20
15
, A
SQ

Portugal
www.asq.org 53
Equation 1
% Defective bags =
Number of defective bags
× 100
Number of processed bags
An additional measure used to assess the evolution
of the production process in general was the defective
items per unit rate (DIUR) (see Equation 2):
Equation 2
DIUR =
Total number of defective items found in 20-kg bags
Number of processed bags
In this introductory phase of the Six Sigma strat-
egy, the packaging process of 20-kilogram fertilizer
bags deserved special attention, especially those sold
in malls with high-quality requirements. It was
confirmed that, to avoid noncompliance with the
specification 20 kilograms with a tolerable mini-
mum variation of 2 percent, there was a need to add
additional fertilizer to bags, to respect the weight
of the minimum specification, which constituted a
significant cost.

An ABC analysis revealed that the packing process
of 20 kilogram fertilizer bags, given that they repre-
sent 50 percent of sales and moreover intended for a
market extremely demanding in terms of quality con-
trol, deserves priority, since there lies an opportunity
for improvement that could significantly contrib-
ute to an increase of the competitive advantage of
NutriSoil and its profitability.
also implemented. The objective of this program is to
increase both production and morale and employee
satisfaction (Ismail 2013; Jain, Bhatti, and Singh
2014). Maintenance is no longer seen as a nonprofit
activity, and downtime to perform maintenance began
to be scheduled as part of the daily production. The
goal is to keep emergency and unplanned maintenance
to a minimum.
Control
Teams were formed to discuss production problems
that may cause errors, failures, and defective items.
The customer complaints analysis was also used to
identify potential problems. The creation of produc-
tion process control charts was implemented in order
to keep employees aware of their process performance
in real time:
• p chart : to control the evolution of the fraction of
defective 20-kilogram bags
• c chart: to measure the evolution of the number of
defective items produced per 20-kilogram bag
A training program was implemented in each of the
production stages with the purpose of bridging the gaps
found. To compare the performance of the production

process before and after the introduction of the Six
Sigma strategy, defective products reported by the entire
production process made use of the following bench-
mark (see Equation 1):
Figure 3 5S methodology
Source: Adapted from Coutinho (2006)
Seiri
Clearout
and classify
Store
Sort
Seiton
Con�gure
Shine
Seiso
Clean and
check
Standardize
Seiketsu
Conformity
Sustain

Shitsuke
Custom and
practice
©
20
15
, A
SQ
Portugal
54 QMJ VOL. 22, NO. 2/© 2015, ASQ
29,071.72 bags, were packed with a weight below the
minimum specification, which corresponded to a cost
of reprocessing of 58,143.44 euros.
To monitor the process progress and calculate the cp
and cpk capability index for the weight of 20-kilogram
bags, control charts for average (x-bar) and amplitude
(R) were implemented and put next to the bagging
machines. Every hour, four bags are randomly packed
and inspected and the respective weight is recorded. If
more than two bags display underweight according to
the minimum specification limit, all bags produced
since the last acceptance are retained and corrective
actions are taken if necessary.

In addition, a laborer carefully monitors the pas-
sage of the bags on the conveyor belt, and from time
to time, removes a bag that weighs less, is torn, or is
badly sealed, and puts it in a pile to be reprocessed. In
the initial stage of the study the process was consid-
ered to be stabilized, but its variability was unable to
meet the specifications.
The customer may receive more fertilizer but
never less than nominal. This excess weight is unnec-
essarily expensive. The problem was identified and
a versatile team was formed consisting of technical
services, plant operators, and quality personnel. The
investigation revealed that the filling machine needed
improvements and that it was appropriate to develop
a training program for employees. The team worked
closely with operators to achieve a reduction in pro-
cess variation with a consequent improvement.
In order to find the causes associated with the high
variation of the bag weight, the authors used a cause
and effect diagram (see Figure 5).
RESULTS
Table 1 summarizes the results of the indicators used
to assess the impact of the actions taken when imple-
menting the Six Sigma strategy.
Control charts for 20 kilogram bag weight average
(x-bar) and amplitude (R) are presented in Figure 6,
before the Six Sigma strategy. Control charts for
20- kilogram bags weight average (x-bar) and ampli-
tude (R) after the Six Sigma strategy are presented in
Packaging Process:

Before the Introduction
of the Six Sigma Process
The packaging process (see Figure 4) consists of
two bagging machines and bags 1,008,000 fertil-
izer bags annually (a daily average of 3,360 bags) in
20- kilogram plastic bags, which are sold on the
market for 15 euros. The problem detected, and the
urgent need to act immediately, is that to comply
with the minimum specification NutriSoil is filling
the bags with an average of 20.93 kilograms, which
equates to an annual waste of 699,693.67 euros; that
is, 46,646 bags are held back, and 90.74 percent of all
bags produced are overweight. However, it is generally
recognized that for the company to remain competitive
and profitable the overfilling should be minimized.
The specifications are imposed by legal rules, so the
bags must contain at least more than 19.6 kilograms
of fertilizer. Bags that do not meet the minimum speci-
fications are again inserted for reprocessing into the
production process, resulting in average unit cost of
2 euros. In addition, bags found on the market weigh-
ing less than the minimum specification are subject
to heavy fines, and 2.88 percent of the bags, that is,
Scale
Silo
Filling Machine
Figure 4 Weighing and packaging process of
20-kilogram bags
©
20

15
, A
SQ
Portugal
www.asq.org 55
DISCUSSION
The introduction of the Six Sigma strategy in
NutriSoil resulted in considerable savings, which
allowed the company to improve its competitive
advantage. The increased cpk index (from a low value
Figure 7. Figure 8 shows the p control chart, for fraction
of defective produced 20-kilogram bags, for a period of
one year of the Six Sigma strategy. Figure 9 shows the
c control chart, used to monitor the evolution of the
total number of defective items per 20-kilogram bag for
the one-year period of the Six Sigma strategy.
Table 1 Metrics employed
Key metrics employed Before the introduction of Six Sigma
After the introduction
of Six Sigma Improvement (%)
Percent of defective bags produced 4.50 2.17 51.77
Defective items per unit rate (DIUR) 4.24 2.12 50.00

cpk – Process capability 0.63 1.02 61.90
Process average (kg) 20.93 20.05 4.20
Process standard deviation (kg) 0.70 0.15 78.90
Percent of bags weighing more than 20 kg 0.907 0.6337 30.20
Percent of bags weighing below the
minimum specification – 19.60 kg 0.0288 0.0011 96.10
Bags lost per year – overweight 46,646.24 2,542.88 94.50
Cost (€) of extra weight – 20 kg bags 699,693.67 38,143.16
94.50
Bags under weight – year 29,071.72 1,142.82 96.10
Cost (€) of underweight – 19.60 kg bags 58,143.44 2,285.65
96.10
Percent of machine downtime 5% 3% 40.00
Percent of work-related accidents 1% 0.2% 80.00
Figure 5 Cause and effect diagram for the weighing and
packaging process of 20-kilogram bags
OperatorScale
Packing
machines
Work
environment
Overweight
bags
Work
conditionsWholesomeness
conditions
Cleaning

Maintenance
Training
Attitude
Precision
Dust
Vibration
Nozzles
MaintenanceVentilation
Calibration
©
20
15
, A
SQ
©
20
15
, A
SQ

Portugal
56 QMJ VOL. 22, NO. 2/© 2015, ASQ
Figure 6 Chart for 20-kilogram bag weight average and
amplitude before the Six Sigma strategy
22.50
UCL = 21.97
LCL = 19.99
X = 20.93
22.00
21.50
21.00
20.50
20.00
19.50
19.00
1 3 5 7 9 11 13 15 17 19 21 23 25
=
Sample
Control chart for average
A
ve

ra
ge
(k
g)
3.50
UCL = 3.28
LCL = 0.00
R = 1.44
3.00
2.50
2.00
1.50
1.00
0.50
0.00
1 3 5 7 9 11 13 15 17 19 21 23 25
—
Sample
Control chart for range
Ra
ng
e
(k
g)

Figure 7 Chart for average and amplitude after a year of the Six
Sigma strategy
20.40
UCL = 20.27
LCL = 19.83
X = 20.05
20.30
20.20
20.10
20.00
19.90
19.80
19.70
19.60
1 3 5 7 9 11 13 15 17 19 21 23 25
=
Sample
Control chart for average
A
ve
ra
ge
(k

g)
0.80
UCL = 0.69
LCL = 0.00
R = 0.30
0.70
0.60
0.50
0.40
0.30
0.10
0.20
0.00
1 3 5 7 9 11 13 15 17 19 21 23 25
—
Sample
Control chart for range
Ra
ng
e
(k
g)
©

20
15
, A
SQ
©
20
15
, A
SQ
Portugal
www.asq.org 57
Figure 8 p chart at the beginning and the end of a period of one
year of the Six Sigma strategy
0.090
UCL = 0.078
LCL = 0.011
p = 0.045
0.080
0.070
0.060
0.050

0.040
0.030
0.010
0.020
0.000
1 3 5 7 9 11 13 15 17 19 21 23 25
—
Sample
Initial p-chart
Fr
ac
tio
n
of
n
on
co
nf
or
m
iti
es
(p

)
0.050
UCL = 0.045
LCL = 0.00
p = 0.0217
0.045
0.040
0.035
0.030
0.025
0.010
0.005
0.015
0.020
0.000
1 3 5 7 9 11 13 15 17 19 21 23 25
—
Sample
Final p-chart
Fr
ac
tio
n

of
n
on
co
nf
or
m
iti
es
(p
)
Figure 9 c chart at the beginning and the end of a period of one
year of the Six Sigma strategy
16
UCL = 10.417
LCL = 0.000
c = 4.24
14
12
10
8
6
4
2

0
1 3 5 7 9 11 13 15 17 19 21 23 25
—
20-kg bags
Initial c-chart
N
um
be
r
of
n
on
co
nf
or
m
iti
es
p
er
b
ag

9
UCL = 6.488
LCL = 0.000
c = 2.120
8
7
6
5
4
1
2
3
0
1 3 5 7 9 11 13 15 17 19 21 23 25
—
20-kg bags
Final c-chart
N
um
be
r
of
n

on
co
nf
or
m
iti
es
pe
r
ba
g
©
20
15
, A
SQ
©
20
15
, A
SQ

Portugal
58 QMJ VOL. 22, NO. 2/© 2015, ASQ
weight lower than that mentioned in the specifica-
tions against an initial 28.84.
• The inclusion of p control charts for each stage of
the production process kept employees aware of the
performance of their process in real time and was
crucial in improving the defective bags produced
fraction of 51.34 percent.
• Using the c control chart helped in reducing the
number of defective items per bag from 4.24 defec-
tive items to 2.12, which was an improvement of
50 percent.
The 5S program of organization and standardiza-
tion tidied up the workplace, storage in particular,
resulting in an 80 percent decrease in work-related
accidents, which had as a consequence decreased
absenteeism and increased worker morale.
Implementing the Six Sigma strategy provided sev-
eral valuable lessons when promoting new projects.
First, it was necessary to educate management that
investing in quality means increasing the cost of produc-
tion. This barrier was overcome by showing the NutriSoil
owner the savings achieved. Also, management had
to be shown that 5S tidied the factory, which reduced
accidents and reduced the idle time of machines and
operators. This in turn generated savings. A training
plan that included all NutriSoil employees was necessary
to overcome active resistance to change. Employee anxi-
ety triggered by change was minimized with an effective

communication campaign. The communication began
with an explanation of the Six Sigma strategy; why the
company adopted it; the anticipated benefits of imple-
menting Six Sigma; dissipation of fears by explaining
the development plan; how the employees would be
affected; and what training and support was available:
• Communication of the training plan, training
schedule, and reporting of training evaluation, as
well as the benefits of implementing the project,
proved to be highly motivating.
• The improvements already made were publicly rec-
ognized and celebrated in order to maintain and
strengthen the commitment to Six Sigma.
Capitalizing on this progress, the NutriSoil
Company is launching the certification process based
of 0.63 to 1.02) for the of the 20-kilogram bags was
an improvement of 60.8 percent. This demonstrates
the value of Six Sigma for SMCs. Using Six Sigma in
the process of packing 20-kilogram bags, annual sav-
ings in the order of 717,408.31 euros were achieved,
corresponding to 4.74 percent of the sales value for
this type of product. Managers accepted that the
Six Sigma strategy combined with the 5S procedure
created, for one year, above all, a change in the orga-
nizational culture of NutriSoil.
Today the authors can perceive in NutriSoil a
greater sense of participation, commitment, and
awareness of the importance of working in order
to exceed the requirements of the consumer. The
involvement and commitment of all employees,
measured by absenteeism and voluntary turnover,

facilitated the implementation of the Six Sigma strat-
egy both within the production process and in the
reduction of defective items that occur in the final
product. Specific gains of NutriSoil were:
• Reduced variation in bag weight from 0.70 kilo-
grams to 0.15 kilograms, that is, a decrease of
78.9 percent. The control charts for average (x-bar)
and amplitude (R) introduced in the packaging
process helped to determine and correct the root
causes of abnormal variation and improvement on
this process. The weighing and bagging machines,
in particular, were subjected to significant improve-
ments. TPM was determinant in increasing
productivity while increasing the morale of employ-
ees and their job satisfaction. The time machine
downtime due to breakdown decreased from 5 per-
cent to 3 percent, in particular because of properly
scheduled preventive maintenance.
• The average weight of the bags went from 20.93 to
20.05 kilograms, which equates to an annual sav-
ing of 661,550.51 euros. Currently, 2,542.88 bags
are wasted due to overfilling, compared to the
initial 46,646.44.
• There was a reduction in the number of the bags
weighing less than the minimal weight. This
resulted in savings of 55,857.79 euros in repro-
cessing costs and a decrease in the probability of
paying fines. Only 1.13 per thousand bags had a

Portugal
www.asq.org 59
Coutinho, A. 2006. Técnicas de melhoria 5S’s – O que são?
Quando se Utiliza e Principais Benefícios. São Paulo, Brazil: L
Teixeira & Melo, Ltda.
Dedhia, N. S. 2005. Six Sigma basics. Total Quali ty
Management 16, no. 5:576-574.
Evans, J., and W. M. Lindsay. 2005. An introduction to Six
Sigma & process improvement. Stamford: Cengage Learning.
Feo, J., and Z. Bar-El. 2002. Creating strategic change more
efficiently with a new design for Six Sigma process. Journal of
Change Management 3, no. 1:60-80.
Franken, B. 2007. How to explain Six Sigma by using the profit
triangle. Available at: www.iSixSigma.com/library/content/.
Harmon, P. 2010. Business process change: A guide for
business
managers and BPM and Six Sigma professionals. Burlington,
MA:
Morgan Kaufmann.
Harry, M. J. 1998. Six Sigma: A breakthrough strategy for
profit-
ability. Quality Progress (May).
The History of Six Sigma. 2007. Available at: http://europe.i-
sixsigma.com/library/content/c020815a.asp.
Ismail, N., et al. 2013. Overall equipment effectiveness
improve-

ment through total productive maintenance in assembly cell.
International Journal of Customer Relationship Marketing and
Management (IJCRMM) 4, no. 1:13-27.
Jain, A., R. Bhatti, and H. Singh. 2014. Total productive main-
tenance (TPM) implementation practice: A literature review
and directions. International Journal of Lean Six Sigma 5, no.
3:293-323.
Kumar, P. 2002. Six Sigma in manufacturing. Productivity 43,
no. 2 (July-September):196-202.
Lucier, G., and S. Sheshadri. 2001. GE takes Six Sigma beyond
the bottom line. Strategic Finance 82, no. 11:101-104.
Magnusson, K., D. Kroslid, and B. Bergman. 2003. Six Sigma –
The pragmatic approach. Scandinavia: Lund Studenlitteratur.
McClusky, R. 2000. The rise, fall, and revival of Six Sigma.
Measuring Business Excellence 4, no. 2:6-17.
Montgomery, D. 2009. Introduction to statistical quality
control.
New York: John Wiley & Sons (Asia), Pte, Ltd.
Nanda, V., and J. Robinson. 2011. Six Sigma. New York:
McGraw Hill Companies.
Pande, P., R. Neuman, and R. Cavanagh. 2003. The Six Sigma
way – How GE, Motorola and other top companies are honing
their performance. New Delhi: Tata McGraw-Hill.
Ray, S., and P. Das. 2009. Improving efficiency and
effectiveness
of APQP process by using DFSS tolls. International Journal of
Six

Sigma and Competitive Advantage 5, no. 3:222-236.
on ISO 9001:2008, and living up to its respect for the
environment; they would like to obtain environmental
certification ISO 14001 – Environmental Certification,
in the short to medium term.
CONCLUSIONS
The Six Sigma program combined with 5S proved
to be an effective, simple way to design and improve
a productive process. By following these systematic
lean approaches, and applying them to Nutrisoil, the
authors achieved not only a substantial reduction in
costs and in the work-related accident rate, but also a
decrease in turnover and absenteeism in a relatively
short period of time.
Further research is needed to evaluate the effec-
tiveness of these approaches in other SMEs in various
settings. The main difficulty encountered was the
owners’ resistance to sharing decision-making power.
This was overcome by demonstrating how Six Sigma
generated money for the company. The Nutrisoil
owner said after implementation that “a change is
visible in the organizational culture of NutriSoil,
based on the ‘consumer voice’ and the adoption of
the best practices, allowing the company to face the
future with more optimism.”
REFERENCES
Antony, J. 2011. Assessing the status of Six Sigma implementa-
tion in the UK manufacturing small and medium-sized
enterprises.
Available at: http://www.gcal.uc.uk/.

Antony, J., and R. Banuelas. 2002. Key ingredients for the
effec-
tive implementation of Six Sigma program. Measuring Business
Excellence 6, no. 4:20-27.
Antony, J., M. Kumar, and C. N. Madu. 2005. Six Sigma in
small and medium sized UK manufacturing enterprises: Some
empirical observations. International Journal of Quality and
Reliability Management 22, no. 8:860-874.
Brewer, P., and J. Eighme. 2005. Using Six Sigma to improve
the
finance function. The Journal of Finance 87, no. 7:27-33.
Buss, P., and N. Ivey. 2001. Dow Chemical design for Six
Sigma rail delivery project. In Proceedings of the 2001 Winter
Simulation Conference, 1248-1251.
Calcutti, T. 2001. Why is Six Sigma so successful? Journal of
Applied Statistics 28, no. 3/4:301-306.
Portugal
60 QMJ VOL. 22, NO. 2/© 2015, ASQ
Ray, S., and P. Das. 2010. Six Sigma project selection meth-
odology. International Journal of Six Sigma and Competitive
Advantage 1, no. 4:293-309.
Rodriguez, R. N. 1992. Recent developments in the process
capa-
bility analysis. Journal of Quality Technology 24, no. 4:176-

187.
Rudissil, F., and D. Clary. 2004. The management accountant’s
role in Six Sigma. Strategic Finance 85, no. 5:35-39.
Toussaint, J., and R. Gerard. 2010. On the mend. Cambridge
MA: Lean Enterprise Institute.
Weiner, M. 2004. Six Sigma. Communications World 21, no.
1:26-29.
Wessel, G., and P. Burcher. 2004. Six Sigma for small and
medium-sized enterprises. The TQM Magazine. Bedford, UK.
16(4), 264-272.
Womack, J. 2011. Gemba walks. Cambridge, MA: Lean
Enterprise Institute, Inc.
Wright, J. N., and R. Basu. 2008. Project management and Six
Sigma: Obtaining a fit. International Journal of Six Sigma and
Competitive Advantage 4, no. 1:81-94.
Wu, C. W., W. L. Pearn, and S. Kotz. 2009. An overview of
theory and practice on process capability indices for qual-
ity assurance. International Journal of Production Economics
117:338-359.
Wu, C. W. 2008. Assessing process capability based on
Bayesian approach with subsamples. European Journal of
Operations Research 184:207-228.
Zucker, D. 2007. Integrating project management into a Six
Sigma System, August 2011.
BIOGRAPHY

Amândio Pereira Baía is a professor in the management depart-
ment at Guarda Polytechnic. He has a master’s degree in
industrial management from Clemson University, and a
doctorate
in statistics and operational research from Coventry University
in
England. He can be reached by email at [email protected]
mailto:[email protected]
Urinary Obstruction
Case Studies
The 57-year-old patient noted urinary hesitancy and a decrease
in the force of his urinary
stream for several months. Both had progressively become
worse. His physical examination
was essentially negative except for an enlarged prostate, which
was bulky and soft.
Studies Results
Routine laboratory studies Within normal limits (WNL)
Intravenous pyelogram (IVP) Mild indentation of the interior
aspect of the bladder,

indicating an enlarged prostate
Uroflowmetry with total voided
flow of 225 mL
8 mL/sec (normal: >12 mL/sec)
Cystometry Resting bladder pressure: 35 cm H2O (normal: <40
cm H2O)
Peak bladder pressure: 50 cm H2O (normal: 40-90 cm H2O)
Electromyography of the pelvic
sphincter muscle
Normal resting bladder with a positive tonus limb
Cystoscopy Benign prostatic hypertrophy (BPH)
Prostatic acid phosphatase
(PAP)
0.5 units/L (normal: 0.11-0.60 units/L)
Prostate specific antigen (PSA) 1.0 ng/mL (normal: <4 ng/mL)
Prostate ultrasound Diffusely enlarged prostate; no localized
tumor
Diagnostic Analysis

Because of the patient’s symptoms, bladder outlet obstruction
was highly suspected. Physical
examination indicated an enlarged prostate. IVP studies
corroborated that finding. The
reduced urine flow rate indicated an obstruction distal to the
urinary bladder. Because the
patient was found to have a normal total voided volume, one
could not say that the reduced
flow rate was the result of an inadequately distended bladder.
Rather, the bladder was
appropriately distended, yet the flow rate was decreased. This
indicated outlet obstruction.
The cystogram indicated that the bladder was capable of
mounting an effective pressure and
was not an atonic bladder compatible with neurologic disease.
The tonus limb again
indicated the bladder was able to contract. The peak bladder
pressure of 50 cm H2O was
normal, again indicating appropriate muscular function of the
bladder. Based on these
studies, the patient was diagnosed with a urinary outlet
obstruction. The PAP and PSA
indicated benign prostatic hypertrophy (BPH). The ultrasound
supported that diagnosis.

Cystoscopy documented that finding, and the patient was
appropriately treated by
transurethral resection of the prostate (TURP). This patient did
well postoperatively and had
no major problems.
Critical Thinking Questions
1. Does BPH predispose this patient to cancer?
2. Why are patients with BPH at increased risk for urinary tract
infections?
3. What would you expect the patient’s PSA level to be after
surgery?
4. What is the recommended screening guidelines and treatment
for BPH?
5. What are some alternative treatments / natural homeopathic
options for treatment?
Inflammatory Bowel Disease
Case Study
The patient is an 11-year-old girl who has been complaining of
intermittent right lower
quadrant pain and diarrhea for the past year. She is small for her

age. Her physical
examination indicates some mild right lower quadrant
tenderness and fullness.
Studies Results
Hemoglobin (Hgb), 8.6 g/dL (normal: >12 g/dL)
Hematocrit (Hct), 28% (normal: 31%-43%)
Vitamin B12 level, 68 pg/mL (normal: 100-700 pg/mL)
Meckel scan, No evidence of Meckel diverticulum
D-Xylose absorption, 60 min: 8 mg/dL (normal: >15-20 mg/dL)
120 min: 6 mg/dL (normal: >20 mg/dL)
Lactose tolerance, No change in glucose level (normal: >20
mg/dL rise in
glucose)
Small bowel series, Constriction of multiple segments of the
small intestine
Diagnostic Analysis
The child's small bowel series is compatible with Crohn disease
of the small intestine.
Intestinal absorption is diminished, as indicated by the

abnormal D-xylose and lactose
tolerance tests. Absorption is so bad that she cannot absorb
vitamin B12. As a result, she has
vitamin B12 deficiency anemia. She was placed on an
aggressive immunosuppressive
regimen, and her condition improved significantly.
Unfortunately, 2 years later she
experienced unremitting obstructive symptoms and required
surgery. One year after surgery,
her gastrointestinal function was normal, and her anemia had
resolved. Her growth status
matched her age group. Her absorption tests were normal, as
were her B12 levels. Her
immunosuppressive drugs were discontinued, and she is doing
well.
Critical Thinking Questions
1. Why was this patient placed on immunosuppressive therapy?
2. Why was the Meckel scan ordered for this patient?
3. What are the clinical differences and treatment options for
Ulcerative Colitis and Crohn’s
Disease? (always on boards)
4. What is prognosis for patients with IBD and what are the
follow up recommendations for

managing disease?
Making the Case for Quality
Process Management Approach
Reduces Scrap, Saves Alcoa Millions
• A team at Alcoa Power
and Propulsion sought
to improve product
quality, reduce waste and
inefficiency, and cut costs.
• To achieve these goals,
the team developed
and implemented a
structured process
management approach
across the business unit.
• By defining and then
standardizing key
processes, the unit
minimized variation,
shared best practices, and
sustained improvements.
• The project improved
customer satisfaction,
reduced costs of poor
quality, and saved
millions of dollars.

• The team was named
a finalist in ASQ’s
2015 International
Team Excellence
Awards competition.
At a Glance . . .
As Alcoa Power and Propulsion sought to minimize
manufacturing process waste,
inefficiency, and related expenses, two key findings caught the
attention of the
business unit’s executive leadership team. Customer satisfaction
surveys revealed
60 percent of respondents held an unfavorable or neutral
opinion of the unit, while
at the same time, internal quality measures revealed the
business unit recorded the
highest levels of scrap in the organization for the year. The
findings represented
a significant opportunity, but improvement would require rapid
deployment in a
methodical, deliberate, and sustainable manner.
About Alcoa Power and Propulsion
Alcoa Power and Propulsion (APP), a unit of New York City-
based Alcoa Inc., is divided into three
segments: structural castings and special products, industrial

gas turbine airfoils, and aerospace airfoils.
This business unit serves the defense, energy, aerospace, and
industrial markets, where its products
are used in military and commercial aircraft engines as well as
power-generation turbines. Additional
products and services include molds, hot isostatic pressings,
specialty coatings and tools, as well as
machinery. APP operates 25 production facilities around the
world, employing more than 9,000 people.
Working to Improve Processes
The APP leadership team discovered scrap and rework issues
were resulting in high customer claims
and delivery issues, therefore they sought solutions for
significant reductions to scrap levels to improve
delivery, rework, and customer satisfaction. In the past,
substantial scrap level decreases were thought
to be unattainable because casting is a particularly complex
process (see the sidebar, The Investment
Casting Process, for more details) with many sources of
variation.
In late 2011, APP’s quality focus shifted toward sustainable and
continuous process improvement,
particularly at nine of the unit’s super alloy foundries (five in
the United States, three in Western
Europe, and one in Japan). While these foundries were the
starting point for the initiative because
they recorded the highest levels of scrap, the ultimate goal for
APP was to create a sustainable solu-
tion for deployment at the plant level by mid-2012. The right
solution would need to capture and
by Janet Jacobsen

May 2016
ASQ www.asq.org Page 1 of 5
communicate tribal knowledge as well as monitor capable
processes. Creating
a new approach could directly impact financial performance by
reducing scrap,
rework, and returns.
Identifying Stakeholders and Building the Team
One of the first steps in forming a project improvement team
was to identify
stakeholder groups from which team members would be
selected. The following
stakeholders were included:
• Internal groups:
engineering
process owners,
business unit and
site managers,
operations
personnel, and
members of the
quality staff.
• External groups:
Alcoa and the
business unit’s
customers.

Once these groups
were determined and
the required knowledge
and skills for each group identified, skill and knowledge gaps
were identified. For
example, the engineering process owners needed training on the
DMAIC method,
while the quality group required additional training on the audit
process. Corporate
quality staff provided introductory training on the
organization’s process manage-
ment methodology, and APP created a Six Sigma training class
for its engineers.
It made sense to build upon already-existing team structures and
working relation-
ships at APP where there are business unit process owners and
corresponding plant
process owners. The improvement project’s deployment team
included members
from these process owner groups.
The project goal was to develop a sustainable methodology to
reduce the APP scrap
rate by 10 percent year over year by June 2012, as depicted in
Figure 1. The team
was to develop manufacturing process models for key areas,
validate the methodol-
ogy at pilot locations, and then prepare a full implementation
plan.
Analyzing the Opportunity
When analyzing the improvement opportunity at hand, the team
quickly realized the

part-focused scrap reduction methodology, which had been used
for years, reached
the point of diminishing returns. To achieve the additional scrap
reductions required
to meet the stated goal, process stability became the new focus.
Since the investment casting process includes several complex
steps, the team cre-
ated process maps to identify, understand, and analyze each step
for its savings
potential. Pareto charts helped illustrate the contribution of
each step to the overall
scrap rate. Brainstorming, scrap rates, and defect data were used
to identify potential
The majority of APP’s business is related to
its investment casting process, which includes
the following steps:
• APP receives a design from one
of its customers, and through a
collaborative process, the proper
shape of the casting is determined.
• A die is constructed, usually in
wax, to produce the pattern.
• This wax die is then assembled
into a cluster to produce multiple
castings economically through a
single pour of molten alloy.
• The wax cluster process continues as
a ceramic slurry is applied, followed
by ceramic sand, and then into
the controlled drying step until the

desired shell thickness is achieved.
• The wax is then melted out and the
resulting shell is fired to create a hard
ceramic ready for casting. Selective
insulation that eliminates casting
defects, such as micro-shrinking, may
be applied at this point before the
shell proceeds to the casting stage.
• During the casting step, metal is melted
under vacuum to maintain purity.
Once melted, the mold is loaded
through a vacuum-lock system.
• After the mold is cast, the gating
is removed and the castings move
through the post-cast operations
and quality inspection.
• Once approved, the certified casting
is shipped to the customer.
THE INVESTMENT
CASTING PROCESS
This is the casting portion of
the investment casting process.
Pe
rc
en
t
Year

2011 2012 2013
Figure 1: Desired Scrap Trend Across APP
improvement opportunities within the
process. In addition, the team utilized
scrap rates at each APP location to
search for potential savings.
Scrap by point-of-cause data was analyzed
within each process area to determine
which processes held the greatest savings
potential. In addition to pinpointing spe-
cific processes for improvement, quality
tools such as Pareto analysis were also
useful in determining which plants to tar-
get for early implementation.
At the outset, the project scope was
quite broad—global business unit scrap
reduction—a hefty challenge. After
reviewing the data and securing buy-in
from leadership, the improvement team
narrowed the scope to focus on the great-
est opportunities—the wax, shell, and
cast manufacturing operations of nine
selected locations. Although the scope
was now more narrowly focused, the
challenge was still immense, requiring
the engagement of the entire operation.
At this point, team membership grew

to include additional process owners,
engineers, quality staff, operations,
and leadership personnel from plants
involved in the project.
Developing
Solution
s
Because the nine APP locations for this
improvement initiative were located
around the world, it was imperative to
create a cohesive language that could tie
all the foundry processes together. This
common language was necessary for not
only deploying the methods and tools,
but also identifying the key operations
and sub-processes at various plants for
comparative and benchmarking purposes.
Process-mapping exercises, shop-floor
interviews, brainstorming sessions, and
benchmarking studies generated data
to help develop potential solutions.

The data was first used for comparing
Alcoa plants with similar processes and
improvement requirements. The com-
pany saved $20 million at seven Alcoa
plants that had implemented process
management initiatives. These results
inspired the improvement team to pursue
a similar approach within APP.
The left portion of Figure 2 depicts a
typical plant-based approach to process
improvement where efforts are confined.
In contrast, the right side of Figure
2 shows how an integrated approach
involving plants throughout the BU
enables the sharing of approaches with
common templates and coordinated
plant projects. The new approach would
Figure 2: Potential Implementation Approaches for Process
Management Directive
Plant 3
Plant 2

Plant 1
Coordinated pilot plant projects
Pilot efforts well-supported
Ef
cient local leveraging
Common process map template
Integrated BU-wide approach
Plant 3
Plant 2
Plant 1
Independent plant projects
Wasteful duplicated efforts
Poor leveraging
Independent process maps

Traditional plant-based approach
leverage one plant’s independent ideas by sharing its
solutions across the entire business unit to spur global
improvement. This business-unit approach also enabled
process owners to transfer technical details through a
common language that was lacking in the past, and to
build a collaborative culture.
The team developed a three-stage plan to implement
its process management approach. The first stage
involved pilot plants; the second fast-follower plants;
before finally, implementation to a broad business-unit
approach. This three-stage rollout provided data that team
members used to validate the efficiency and speed of the
shared and integrated approaches.
The key to incorporating solutions throughout the busi-
ness unit was a structured, common process model, which
incorporates the tools recognized as the best-suited means

to get processes in control, and to identify key operations
and sub-processes at multiple plant locations. Utilizing
this process management approach, as illustrated in
Figure 3, would increase sustainability through manage-
ment review, quality audits, rigorous control plans, and a
focus on the correct variables.
The process management solution was validated through
the pilot and fast-follower stages. For example, one of
the pilot-stage plants deployed a new casting/mold wrap
process and demonstrated a 30 percent reduction in
related scrap in just three months. Then a fast-follower
plant used the same new process as the pilot plant and
achieved similar scrap reduction results, but this time
within just 30 days. This demonstrated that lessons
learned at one plant advance the process management
efforts at other locations when similar process improve-
ments are implemented.
Initial
assessment
Identi�cation of
key processes and
process managers

Audit and
management
review
Sustain
the gains
Process
improvement
In control,
capable, and
sustainable
Yes
No
Figure 3: Alcoa Process Management Approach
Overcoming Resistance
During the team’s initial analysis members discovered a few

areas of
resistance, including:
• Concerns regarding the availability of resources, particularly
the ability for employees to handle the new initiative while still
meeting production goals.
• Stakeholders at various levels lacked a sense of urgency and
showed some hesitancy about making changes.
• Because the process management approach was a
nontraditional
method for reducing scrap, a change in culture would be
essential
to inspire employees to embrace the new process-focused
method.
The team understood providing information and education was
vital to
achieve positive change, so the process management model
includes
routine communication such as holding team-building meetings
and shar-
ing monthly process audit databases. Resistance was also
mitigated by

engaging the entire workforce in the new process management
approach.
Examples included employees who helped by defining
improvement
opportunities, planning and prioritization activities,
participating in kai-
zen events, as well as establishing and executing control plans.
Exceeding Goals and Realizing Additional Benefits
Once the pilot and fast-follower stages were completed, APP
imple-
mented the process management model throughout the business
unit.
While the initial project goal was to reduce scrap by 10 percent
year
over year, those expectations were exceeded by more than $20
mil-
lion in improvements from the three key areas: wax, shell, and
cast.
Tyrus Hansen, APP team lead, said reductions in scrap and
other
costs of poor quality continue year over year, as highlighted in
APP’s
Quality Index, shown in Figure 4.

Specific examples of dramatic cost savings include:
• A 77 percent improvement in wax expenses in one department
at
one plant, which resulted in a $38,000 annual savings.
• A sub-process involving shell weight control to reduce
material
costs at one production facility resulted in a 21 percent
improvement and an annual savings of $400,000.
• A sub-process of the cast/wrap step reduced scrap and rework
expenses by more than 30 percent, approximately $750,000, at
one location.
Not only did the new approach lower scrap rates, but it also
reduced
rework activities and customer returns, leading to more on-time
deliv-
ery and higher levels of customer satisfaction. At the outset of
this
project, all nine plants involved in the pilot were rated in the
poor
performance category, but since that time, all of the

manufacturing
locations have fully integrated the process management
approach.
An unexpected benefit of the work was an improved workforce
devel-
opment. Many team members completed training and then
gained
valuable experience with basic problem-solving tools—new
skills that enabled employ-
ees to operate in a more proactive, process-focused manner.
Members of the improvement team learned valuable lessons
throughout the project, but
Hansen said the biggest lesson centered on discipline as team
members realized, “If we
maintain the discipline, quality will improve,” he said. “The
proactive work minimizes
the time spent managing problems and fighting fires.”

Sustaining the Momentum
The APP process management models for the wax, shell, and
cast operations were
revised to reflect the new approach. These models were
deployed through the process
owner group at each APP location. Results are measured and
reported at the process
and sub-process levels with metrics directly linked to each
location’s business case.
Reports and feedback soon became more proactive and process-
focused rather than
reactive and problem-focused as was the case before this
improvement initiative.
To ensure the new process management approach remains
sustainable and continues to
perform to expectations, APP implemented a number of
controls, including:
• The business case must drive all activities associated with this
approach and the
activities are measured for impact.
• A reporting structure was created to communicate specific

details on a regular basis.
• All efforts are recorded and included in the business unit
manuals and audit tools.
As APP continued to implement the process management
approach, the business unit
realized further benefits, such as:
• By emphasizing control of the entire process, additional
operators and supervisors
are engaged in more proactive roles.
• The workforce is more satisfied and engaged due to greater
stability, education,
and structure.
• With fewer daily “firefighting” activities, time to focus on key
process metrics for
employees increases, triggering reactions before issues become
scrap problems.
GOAL
reduction
from 2013
6.59%

Reduction
from 2013
9.23%
GOAL
reduction
from 2012
7.74%
Reduction
from 2012
17.24%
GOAL
reduction
from 2014
19.22%
Reduction
from 2014
2.11%
Represents millions
of dollars of opportunity

Actual
2012
Projected
2013
Actual
2013
Projected
2014
Actual
2014
Projected
2015
Actual
2015
Figure 4: APP Quality Index Trend from 2012 to 2015
Sharing the Team’s Story

The team has devoted a great deal of
time in sharing its success story with
others. Within Alcoa, the model used in
this project now serves as the corporate
standard and is being applied at various
locations and business units around the
globe with similar results. The team’s
effort was also highlighted at the 2014
Alcoa Corporate Process Management
Network Conference and was a can-
didate in the 2013 Alcoa Corporate
Impact Award Competition.
At the prompting of Alcoa’s cor-
porate quality department, the team
entered its project into the ASQ 2015
International Team Excellence Award
(ITEA) competition. The project was
named a finalist, and this recognition
enabled members of the team to share
APP’s proven process management
approach during the World Conference
on Quality and Improvement.
For More Information

• For additional information on
Alcoa Power and Propulsion,
visit the organization’s website at
www.alcoa.com.
• To learn more about this project
from a team interview, watch this
video http://videos.asq.org/alcoa-
six-sigma-and-teams.
• To view more examples of
successful improvement projects in
quality, visit the ASQ Knowledge
Center at asq.org/knowledge-center/
case-studies.
• Learn more about the International
Team Excellence Award Process
by visiting the official website at
asq.org/wcqi/team-award/index.aspx.
About the Author
Janet Jacobsen is a freelance writer

specializing in quality and compliance
topics. A graduate of Drake University,
she resides in Cedar Rapids, IA.
www.alcoa.com
http://videos.asq.org/alcoa-six-sigma-and-teams
http://videos.asq.org/alcoa-six-sigma-and-teams
http://asq.org/knowledge-center/case-studies
http://asq.org/knowledge-center/case-studies
http://asq.org/wcqi/team-award/index.aspx
Making the Case for Quality
February 2017
by Sivaram Pandravada and Thimmiah Gurunatha
Using Exploratory Data Analysis to
Improve the Fresh Foods Ordering
Process in Retail Stores

With the abundance of data now available in the information
era, data science offers significant
opportunities to complement quality approaches to problem
solving and continuous improvement.
Unfortunately, as Roger D. Peng and Elizabeth Matsui point out
in their book, The Art of Data
Science, “Data analysis is hard, and part of the problem is that
few people can explain how to do
it.”1 Quality professionals may find that applying the principles
of data science is not always as
straightforward and formulaic as they would like.
In illustrating how explorative data analysis and basic statistics
helped a grocery chain reduce
inefficiencies in its retail inventory and ordering process, this
case study presents a real-world
example of how the thought processes of data scientists can
contribute to quality practice. Once
the chain saw the difference that the data-based approach made
in targeting waste and ineffi-
ciency in one store, it successfully replicated its process
improvements to increase profitability
throughout the organization. The project approach and lessons
learned can also be applied in other
retail settings.

A data science approach to assessing the problem
The short shelf lives of fresh foods along with fluctuating
consumer demand had been causing a
European retail chain’s stores to hold clearance sales with zero
or negative margin or to write off
some inventory as “shrinkage.” Annually, the problem of
shrinkage accounted for revenue losses of
up to 20 percent.
Although the ordering process was automated, the algorithm
was better suited to articles with lon-
ger shelf life (>90 days); hence, department managers would
often overwrite the automatic system
and place orders manually. There were thousands of stock
keeping units (SKUs) and no defined
tolerance limits to manage stock and shrinkage across various
stores. The organization needed an
approach for monitoring and controlling this waste.
Granular data is key to accurate and productive problem
assessment. When it is not available,
significant time goes into defining metrics and capturing
measurements before analysis can begin.

With the prevalence of sophisticated database management
systems, the trick becomes extract-
ing information based on the questions that need to be answered
to gain a better understanding of
any problem.
For the problem of shrinkage, using data science and lean Six
Sigma started with two questions:
• The short shelf lives of fresh
foods along with fluctuating
consumer demand meant
that a European retail
chain’s stores often had to
hold clearance sales with
zero or negative margins
or write off inventory.
• Explorative data analysis
and basic statistics helped
the chain identify and
reduce inefficiencies in its
inventory and ordering
process, minimizing the
gap between quantities sold

and quantities ordered.
• Rainbow statistical process
control (SPC) charts, a
variation on traditional
SPC, helped ensure
ongoing monitoring of
the stock-to-sales ratio
and triggered corrective
action in real time,
bringing sustainable results
within three months.
At a Glance . . .
http://www.asq.org
Question 1: What are we trying to optimize in this project?
Answer: The team seeks to decrease shrinkage in fresh food
divisions and also decrease the use of discount sales that lead to
reduced margins.

Question 2: What factors have the most effect on shrinkage
and discount sales?
Answer: Shrinkage and discount sales occur as a result of
excess inventory in stores because of improper ordering pro-
cesses. Key factors to study include the following:
• Identify the biggest contributing categories and departments
leading to high shrinkage and discount sales
• Quantify the demand or conduct a sales study of these
articles by number of SKUs
• Quantify the supply or purchase of these articles by number
of SKUs
• Review gaps in the current ordering process leading to
supply versus demand mismatch
The Japanese concept of muda, which translates to “waste” or
“any activity that consumes resources but creates no value for
the customer,” provided direction for the team’s next task of
identifying and reducing waste to improve process efficiency.
Pareto analysis helped the team understand which categories

and SKUs were contributing to 80 percent of shrinkage and dis-
count sales and prioritize them for improvement.
Exploratory data analysis
The team used data analytics and lean Six Sigma tools to iden-
tify and implement corrective action that would reduce waste
and improve the ordering process. Targeting the SKUs that
were contributing significantly to high shrinkage and discount
sales, the team started with analysis of the daily unit sales of
one fresh food SKU with a shelf life of five days and the high-
est shrinkage component within the fresh food category.
The histogram data in Figure 1 show daily sales in one store
where there was maximum shrinkage due to this fresh food
SKU. One observation that immediately stands out is that
the sales data appear to be normally distributed if three data
points for times when demand was an outlier are omitted from
consideration. The distribution is good news from a modeling
perspective and in terms of predictability.
In Figure 1, it is also easy to see that out of 225 days, the store
sold its highest quantities on only a small percentage of days:

• ≥ 280 sold on only 19 days
• ≥ 320 sold on only nine days
• ≥ 400 sold on only two days
Figure 2 presents a histogram for quantities ordered for this
same SKU. Reviewing alongside the data on quantities sold
leads to a few notable takeaways:
• The ordering quantity did not follow a normal
distribution, even though the process is within the control
of the stores
• Out of 142 orders placed for the same time period of 225
days, approximately 80 were for quantities ≥ 300
• Department managers were particularly surprised by the
fact that approximately 56 percent of the time, an order
was placed for a quantity that sold only 4 percent of
the time
A simple comparison of the sales and order quantities
summarized by weekday should give any process owner
many subsequent questions to ask the department manager
56

50
31
10
6
1 1 0 0 0 0 1
40
21
8
80
30
20
10
0

40
50
60
160 240 320 400 480 560
Fr
eq
ue
nc
y
Quantity sold
36
51
7

2 0 1
11
0 0
3
0
21
0 02
3
5
00 01
0
30
20
10

0
40
50
75 150 225 300 375 450
Fr
eq
ue
nc
y
Quantity ordered
Figure 1: Histogram of quantity sold and frequency
of occurrence
Figure 2: Histogram of quantity ordered and frequency
of occurrence
http://www.asq.org

http://asq.org/learn-about-quality/cause-analysis-
tools/overview/pareto.html
responsible for ordering. In particular,
why is there so much variation between
the ordering process and the quantity
sold when the article can be ordered six
days in a week and has only five days of
shelf life?
Table 1 focuses on a specific category/
SKU and iterates the gaps in quantity of
goods sold and goods received/ordered.
Model building and interpretations
Model building is part of the improve
phase of a typical Six Sigma define,
measure, analyze, improve, control
(DMAIC) project. Having contributed
to defining and understanding the prob-
lem and analyzing opportunities for

improvement, data still have a role to
play in constructing and implementing
solutions for sustained results.
For this project, a simple model might
be a rule-of-thumb estimation or an
empirical equation that can predict the
ordering quantity and the quantity to be
maintained as minimum inventory. It
involves iterations of steps and assump-
tions to arrive at an equation that best
fits the demand of the store for the item.
One has to make assumptions for the
model and conduct backtesting based
on quantities demanded to ascertain
that there are no instances of lost sales.
Comparing the quantities ordered
against actual lost sales incidents plus
shrinkage incidents and discount sale
incidents leads to a calculation of the
benefits for the project. In the spirit of
lean, a 10 percent reduction in shrink-
age within three months was selected as

the improvement target. The basis of the
model is to minimize the gap between
the quantity sold, represented by the
blue line in Figure 3, and the quantity
ordered, shown in red.
With the model and goals established,
interpretation and communication then
become project-specific and store-
specific based on assortment demands
within different geographies.
This approach is complementary to Six
Sigma DMAIC, but instead of emphasiz-
ing the conventional defects per million
opportunities (DPMO), waste reduction
is measured in terms of money written
off as shrinkage, a metric to which store
and department managers can more eas-
ily relate. As a result, the control charts
were designed around shrinkage.
Turning data into action
In order to reduce excess ordering, the

warehouse used color coding to moni-
tor and track inventory. The colors of
tags on the shelves helped ensure that
department managers would see indica-
tions of stock levels during daily physical
inspections. Red tags indicated SKUs
with high shrinkage. Blue tags signaled
SKUs with high stock levels.
Ordering proceeded according to stock
shelf life and quantity sold. The store
manager would only have to look at the
tags on the shelf to understand the need
to investigate any SKUs that might lead
to shrinkage events and require preven-
tive action.
A rainbow SPC chart (see Tools Used
in This Case Study on page 5) was
Table 1: Comparison of quantities sold versus quantities
ordered by day of
the week

Number
days
ordered
Quantity
received
Average
of quantity
received
Number
days sold
Quantity
sold
Average
of quantity
sold
Sunday 32 3,186 100
Monday 22 5,612 255 32 5,951 186

Tuesday 28 7,099 254 32 5,941 186
Wednesday 16 4,776 298 32 5,731 179
Thursday 30 7,827 261 32 5,994 187
Friday 18 5,169 287 32 6,536 204
Saturday 29 8,650 298 33 5,903 179
Total 143 39,133 274 225 39,242 174
0
20
40
60
80
100
Pe

rc
en
t
0 100 200 300 400 500 600
39.6
34
9.
4
80
25
0
85.6
23
4.
2

Variable
Quantity sold
Quantity received
Data
Normal
StDev
71.06
90.01
Mean
174.4
273.7
N
225
143
Figure 3: Cumulative distribution of quantities sold and
ordered
http://www.asq.org

deployed for real-time identification
of spikes in the stock-to-sales ratio, as
represented in Figure 4. The escalation
matrix that was established required
root cause analysis for any rise or fall
out of the green zone. Entry of the
ratio into the red zone would need
intervention of the store manager,
while entry into the orange/pink zone
would prompt the intervention of the
floor manager. Entry of the ratio into
the yellow zone would need the atten-
tion of the department manager who
is responsible for placing the order.
Acting on the immediate feedback
provided in rainbow SPC monitoring
helped reduce shrinkage by 20 percent
over a period of one month.
Figure 4 presents the rainbow control
chart plotting the stock-to-sales ratio

based on quantity. At various points, the
ratio enters the yellow, orange, and red
zones, requiring escalation by one level
up at each stage and root cause analysis
by different levels of management.
The approach ensured real-time track-
ing of variation and brought sustainable
results within three months. The team
monitored process data for the 10 top-
selling articles for each department
using a simple Excel template with a
rainbow SPC macro and subsequently
built a dynamic reporting and chart-
ing tool using other data visualization
software. Internal benchmarking then
allowed the organization to continuously
improve the ordering process across
store outlets.
Pre-control inventory stock management
Product line/type of service Fresh chicken Country Xyz
Quality characteristics Stock + ordering to sales ratio

Department Fresh food
Date(s) 1/10/2015 to 30/04/2016 Ordering team/source# Fresh
food
Purpose of data collection Monitoring ordering process to
improve performance Station number/department EBITDA

Achieving Customer Specifications Through Process Improv.docx

Recommended

Recommended

More Related Content

Similar to Achieving Customer Specifications Through Process Improv.docx

Similar to Achieving Customer Specifications Through Process Improv.docx (20)

More from bartholomeocoombs

More from bartholomeocoombs (20)

Recently uploaded

Recently uploaded (20)

Achieving Customer Specifications Through Process Improv.docx