Journal of Computer Information Systems

ISSN: 0887-4417 (Print) 2380-2057 (Online) Journal homepage: http://www.tandfonline.com/loi/ucis20

Information Technology and Six Sigma

Implementation

Chang-Tseh Hsieh, Binshan Lin & Bill Manduca

To cite this article: Chang-Tseh Hsieh, Binshan Lin & Bill Manduca (2007) Information Technology
and Six Sigma Implementation, Journal of Computer Information Systems, 47:4, 1-10

To link to this article: http://dx.doi.org/10.1080/08874417.2007.11645975

Published online: 05 Jan 2016.

Submit your article to this journal

Article views: 10

View related articles

Citing articles: 3 View citing articles

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=ucis20

Download by: [University of Newcastle, Australia] Date: 26 February 2017, At: 02:08
 INFORMATION TECHNOLOGY
AND SIX SIGMA IMPLEMENTATION
CHANG-TSEH HSIEH BINSHANLIN BILL MANDUCA
University of Southern Mississippi Louisiana State University Northrop Grumman Corporation
Hattiesburg, MS 39406 in Shreveport Ship Systems Sector
Shreveport, LA 71115 Pascagoula, MS 39568-0149

ABSTRACT This paper will briefly discuss the basic concepts of the Six
Sigma Process Improvement Methodology, present various
The Six Sigma methodology for quality improvement is computer applications to demonstrate the value of ISIIT in this
a system of many concepts, tools, and principles. This paper methodology, and finally, present an actual case where some
discusses the basic concepts of the Six Sigma Process Improve- of these applications were used to help a major corporation to
ment Methodology, presents various computer applications to achieve significant cost savings.
demonstrate the value of information systems and information
technology (ISlIT) in this methodology, and finally, presents an THE SIX SIGMA METHODOLOGY
actual case where some of these applications were used to help
a major corporation save over $2 million. It is hoped that the Ever since Lean and Six Sigma was first introduced to
focus on ISIIT shows how powerful Six Sigma can be when aided the business community by Motorola in 1985, this concept
by the latest computer hardware and software technology. The has primarily been applied to manufacturing. These days, the
strategic use of ISIIT within the Six Sigma methodology will go application of Six Sigma can be found in areas ranging from
a long way in making things easy for professionals to incorporate facility management and maintenance functions (12), online
the principles as part of their everyday work routine. market research (22), supply chain improvement (14), to such
Keywords: Six sigma methodology, Process improvement, non-manufacuturing areas as healthcare management (21),
Strategic use of ISIIT. managerial accounting (2), and human resources management
(25). And, the formulation and identification of useful theories
INTRODUCTION related to Six Sigma development have also been proposed (16).
The strategic value of the Six Sigma methodology has also
Six Sigma is a quality improvement management doctrine become the focus of studies by both academes and practitioners.
developed by Motorola in 1985. (3, 8, 20). It was proclaimed It has been treated as a vehicle for improving competitive
as a new approach to improving quality through statistical mea- advantage by some business executives (7), and the expanded
surements and benchmarking. In the words of Pande, Neuman, and Six Sigma system, commonly known as Keki Bhote's Proven
Cavanagh, it is because being successful and staying successful System, can even help business move from quality excellence to
in business is more challenging today than ever before (18). The total business excellence (15). It is the strategic value that makes
Six Sigma method attempts to improve such aspects within the the Six Sigma methodology even more important, especially in
operation as profit and speed increment. the global environment.
In today's economy, most people provide services rather than
making goods and products. Most of these services operate very STRATEGIC MANAGEMENT SYSTEM
inefficiently, and if the same thing happens in a factory, many
factories would produce too many defective products and they The Six Sigma approach to managing is all about identifying
would be closed down quickly. Six Sigma provides powerful what is known and not known about the various processes that
tools to improve those services to levels of accuracy and quality a company relies upon to conduct its business, and then taking
seen so far only in precision manufacturing. action in the form of problem solving teams working on projects
Six Sigma took the business community by storm in the 1990s in targeted areas, to reduce the errors and rework within these
when General Electric CEO Jack Welch incorporated it in his processes - errors that cost time, money, opportunities, and
management philosophy (9). To date, at least 25% of the Fortune customers. It must be kept in mind, however, that a successful
200 companies have a serious Six Sigma program implemented Six Sigma initiative is not just about data tools and defect
(23). Most of these implemented programs, however, focus on calculations. Pande, Neuman, and Cavanagh (18, 19) define it as
large scale improvements in defect and process measures, which a broad and comprehensive system for building and sustaining
may also involve employee motivation (17). The increasing business performance, success, and leadership. It must be viewed
popularity of the Six Sigma approach has been paralleled with as a context within which the organization will be able to integrate
the evolution of information technology, in particular those many valuable but often disconnected management "best
related to personal computers and networking technologies practices" and concepts, including systems thinking, continuous
since 1980s. Nevertheless, impacts of information systems improvement, knowledge management, mass customization, and
and information technology (ISIIT) on the Six Sigma initiative activity-based management.
has barely been mentioned, especially in the academic com- The importance of a Six Sigma initiative being part of an
munity. overall strategic vision by a business cannot be over emphasized.

Summer 2007 Journal of Computer Information Systems 1

In fact, an organization must embrace the Use of Customer Data
following themes in its strategic vision in
order to be successful (4, 5, 11):
How companies plan to use the information they have
1. A genuine focus on the customer assembled on their customers .. .
backed by an attitude that puts the 1999 2001
customers' needs first, as well as by MarKeting . 18% 52%
systems and strategies that serve to Customer Serv;ce 16% 48%
tie in the business to the "Voice of Sales 16% 34%
the Customer." P ro c e s s Improvement 2% 22%
2. Data-andfact-driven management,
F r.aud Detection ., -, 100,(Q " •. 14 %
with effective measurement sys-
tems that track both results and Product Development 4% 10%
outcomes and process, input and DaR't Use,Data 72% 0%
other predictive factors. Don't K now 0% 18%
3. Process focus, management, and
improvement as an engine for How many companies expect they will benefit from
growth and success. Processes in
their w e al t h of customer information ...
Six Sigma are documented, com-
municated, measured, and refined
1999 200 1
Increase Revenue 200/0 - .
..., 7'14%
on an ongoing basis. They are also
designed or redesigned at intervals Cut Expenses 16% 34%
to stay current with customer and No lnmpact " .- .~
72% ·" 0 %
business needs. Don't K now 0% 20%
4. Proactive management involving
habits and practices that anticipate
problems and changes, apply facts Data: Forrester Research Survey of 50 of the 1000 Largest U.S .
and data, and question assumptions Co"l>anies
about goals and "how we have
always done things."
Figure 1
5. Boundary-less collaboration, fea-
turing cooperation between internal "Consider some indirect evidence as to how well
groups and with customers, suppliers, and supply chain companies really use the inputs they receive from their
partners. customers:
6. A drive for perfection, and yet a tolerance for failure,
that gives people in a Six Sigma organization the • While point-of-sale data-gathering technologies
freedom to test new approaches even while managing (scanners, smart cash registers, credit card
risks and learning from mistakes, thereby "raising the systems, etc.) have been around for almost two
bar" of performance and customer satisfaction. decades, they've been integrated only rather
slowly into daily operations. For example, it's
It is obvious that a corporation must apply various lSI been only in the past few years that data from
IT solutions to incorporate these themes at the pace today's retail store barcode scanners have been used to
business environment demands. IS/IT, used in the context of create automatic replenishment programs that
Six Sigma, gives the manager a more objective, comprehensive, send computerized "reorder" notices to product
complete picture of his area of responsibility. However, while the manufacturers.
technology exists to get a really clear understanding of customer • Despite a lot ofpress coverage of, and investment
requirements, most companies appear not to be taking advantage in, data warehouses - vast stores of raw facts
of what is available. Consider the following excerpt from The Six collected about transactions and customer be-
Sigma Way (18,19): haviors - many companies still aren't using
these resources consistently. According to a
"It's easy to assume that most companies have a pretty survey that sampled 50 leading U.S. companies,
good handle on their customers' needs, or have people some 72 percent said that as of 1999 they were
and mechanisms in place to keep tabs on them. Certainly not making use of the customer data provided
a lot of money is spent on market research and customer by their transactional information system." See
surveys by companies of all types. . . . We would suggest, Figure I.
however, that many of the practices in use today, to keep
tabs on customers' needs, create a false sense of security. A properly applied Six Sigma initiative aligns the company's
When they are examined more closely, many companies product and process improvement projects with the overall
are likely to come to the same conclusion as that reached corporate strategic plan. There are several IS/IT software solutions
by an executive at a large insurance company: "We began to track the progress and Return on Investment (ROI) of each
to realize we didn't understand our customers as well as project. Additionally, proper use of the data mined from the use
we thought we did." of IS/IT within these projects brings opportunities for innovation

2 Journal of Computer Information Systems Summer 2007

and growth to light where they might have been missed before Define - The problem is identified and customer require-
(See figure I). ments are determined;
Measure - The process and its defects are quantified;
IS/IT AND THE ANATOMY OF A PROJECT Analyze - The data is statistically reviewed and the root
causes of the problem are discovered;
The focus now shifts to the application of IS/IT to the specific Improve - The process is modified to remove the causes
parts of an improvement project. This is a data driven, process of defects;
approach commonly referred to as the DMAIC roadmap, which Control - The improved process is run in such a way that
stands for define, measure, analyze, improve, and control (24). The the defects do not recur.
term "Six Sigma" is a reference to a particular goal of reducing
defects to near zero. Sigma (a) is the Greek letter statisticians use Simply put, Six Sigma management is about tying quality
to represent the "standard deviation of a population." The sigma, improvement directly to the financial performance of the
or standard deviation, tells how much variability there is within a company. The Six Sigma goal is to link internal processes and
group of items (the "population"). The more variation there is in systems management to customer requirements. Six Sigma drives
a process, the bigger the standard deviation. the organization to a more scientific means of decision making
In statistical terms, therefore, the purpose of Six Sigma is by basing everything on measurable data. IS and IT are integral
to reduce variation to achieve very small standard deviations components of the DMAIC process. In the following sections, the
so that almost all products or services meet or exceed customer discussion will center on the elements that are critical for success
expectations (13). The ability to reduce the standard deviation in for that phase and the ISIIT tools that are usually applied.
a process (i.e. remove variation) is the key to reducing the Cost
of Poor Quality (COPQ) and positively impacting customer Define Phase
satisfaction. The smaller the value of a in a data set, the greater
is the "a capability" of the process. This may not seem intuitive In this initial phase, the tools used help to provide a top down
until one realizes that "a capability" is defined in terms of "how look at the processes from a business perspective. Everything
many a's" can be absorbed within a specified tolerance range of starts with the customer and what is important to him. Tools used
a performance measure. here help to prioritize and focus on the areas where the greatest
Sigma capability is also directly related to defect rates. A opportunities of improvement are. Each project should be very
4a capable process has 6,210 defects per million opportunities well defined before leaving this phase.
(dpmo), a 5a capability is 233 dpmo, and a eo capability is 3.4 Tasks associated with this phase include:
dpmo. In other words, if a process is said to be operating at a
Six Sigma level it means that there are only 3.4 defects per mil- • Establishing a project charter that includes the
lion opportunities in that process. Figure 2 shows the relation- business case, problem statement, project scope
ship between the number of defects, and the sigma level of a and constraints, budget, assumptions, team
process. guidelines and membership, project schedule,
Six Sigma is much more than a mathematical definition. Six and identifying significant stakeholders;
Sigma is about producing better products and services faster and • Identifying customers (external and internal)
at a lower cost. The Six Sigma methodology uses statistical tools who use the process, customer requirements
to identify the vital few factors that matter most for improving the in the form of a Voice of the Customer (VOC)
quality of processes and generating bottom-line results. It consists exercise, output requirements, and service
of the following five phases (6,10): requirements.

Sigma Convers ion Chart

700 ,000

600,000

500 ,000 -

400,000
DPMO
300 ,000 -

200 ,000

100 ,000

o
2 3 4 5 6
Sigma Level
Figure 2

Summer 2007 Journal of Computer Information Systems 3

 Each "house" matrix shows the weighted relationships

~~
between requirements down the left side and the char-
omlldon ofD88ign acteristics of each level of detail across the top. Each
Requirements relationship is rated on a scale of I (weak), 3 (moderate).
or 9 (strong). A blank represents no relationship at all (a
D_gn Requirements zero rating). The results then become the prioritized Cl'C's
that form the left side of the next matrix. Sources of data
m and information to start the QFD regarding customer
expectations are typically gathered from:
PrIo~ed Customer ReI8lIonshlp of Custom... Rilling of
Need. ~e.lgn tI:J Need. Compelltorw f4 • Targeted and multi-level interviews and surveys
In • Customer scorecards
• Data warehousing and data mining
TechnlClll
Benchm....
The cascading of customer requirements into design
151 parameters for the product, process, and control function
is the essence ofQFD. QFD is the tool used to translate the
TechnlClll EVlllullllon
Voice of the Customer into system design parameters. Most
m of the other ISIIT tools associated with this phase consist
of worksheets and other computerized templates aimed at
planning the project.
Pertonnence TargellS

I8l Measure Phase

Figure 3
I HouaeofQuaUty I The objective of the Measure Phase is to acquire good
data that is an accurate record of the performance of the
The one ISIIT application that incorporates all these tasks process. This is usually completed using the following five-step
into one comprehensive study is known as Quality Function implementation process:
Deployment (QFD) or House of Quality. QFD is an iterative
process for continually refining customer requirements to ever- I. Select what to measure
increasing levels of detail and specificity. The QFDlHouse of 2. Develop operational definitions
Quality is shown schematically in Figure 3. 3. Identify data source(s)
QFD is extremely useful when determining what the customer 4. Prepare collection and sampling plan
requirements and priorities are and then flowing these requirements 5. Implement and refine measurement
down to critical product, process, and control design parameters.
The following set of cascading matrixes (Figure 4), or Houses Before the process can be measured, the measuring system
of Quality demonstrate the flow from customer expectations to must be validated much like a pressure gauge is calibrated for use
supplying a finished product or service: before installing. for example, into a hydraulic system. This action
identifies and quantifies the
different sources of variation
that affect a measurement
system. The purpose for
performing a Measurement
System Analysis (MSA) is
to assess how much variation
is associated with the mea-
surement system and to com-
pare it to the total process vari-
ation or tolerances. Variation
in measurements can be at-
tributed to variation in the
product itself or to variation in
the measurement system. The
Product o..lgn variation in the measurement
CTC,"
system itself is sometimes
called measurement error.
Typical ISIIT tools used
in this phase include deter-
Figure 4
mining the data sampling
plan, collecting data from the
House of Quality Matrix Iteration Flow .. 'g . Proc...
Control process for future statistical
analysis such as Sigma Level

4 Journal of Computer Information Systems Summer 2007

and Process Capability, measuring defects and process yields, • Allows for the performances of dissimilar processes to
measuring/calculating Cost Of Poor Quality (COPQ), Pareto be compared.
charts, run charts, histograms (frequency plots), process flow
diagrams, and control charts. Before moving to the Analyze phase, Analyze Phase
the problems occurring within the process need to be precisely
defined and under what conditions they are likely to appear must The goal of the Analyze Phase is to use data analysis tools
be documented. Data needs to be collected to demonstrate what and process analysis techniques to identify and verify root causes
specifically is the main problem(s), how critical input, process and of the problem. In data analysis, the data that has been collected
output measures were prioritized and selected, show the results about the process is used to find patterns, trends, and other
of the measurement system analysis, determine what patterns differences that can suggest, support, or reject theories about
exhibited in the data, and determine the process capability. the causes of defects. With process analysis, a detailed look at
One of the most useful tools is the Capability Analysis. Process the existing key processes that supply customer requirements is
capability measures provide an index regarding how "good" the conducted in order to identify cycle time, rework, downtime, and
process is. More specifically, process capability is a measure other steps that don't add value for the customer.
of the performance of a process relative to the specifications of Figure 5 shows the three phases of root cause analysis typically
the process output. Determining the capability of a process is performed in a Six Sigma project during the Analyze Phase:
sometimes referred to as comparing the "voice of the process" to
the "voice of the customer." Process capability gives a manager One way to help validate or refute these theories is
some useful information: to perform hypothesis testing, or tests of statistical signi-
ficance. They are some of the most important techniques
• Can be applied to administrative processes or production- used by statisticians to look for patterns or test their sus-
oriented processes. picions about what is really happening in a process and
• Predicts the defect rate of a process. Once the process have a number of applications in Six Sigma projects.
capability is known, the proportion of the output that Figure 6 is a hypothesis testing decision tree which helps
will be defective can be estimated. to determine the correct statistical test to use based on
• Allows process improvements to be quantified (i.e. before the type of data (continuous vs. discreet), the type of data
and after the improvement) with capability indices. distribution (normal vs, non-normal) and the sample size.

Figure 5

Summer 2007 Journal of Computer Information Systems 5

 Reglllaeion Ana Iytia Chi-$qua III Teat

Nonnal No Two ormOlll

Distribution? sample.. One
factor

Two orMolll Two orMolll

Two Sample..
On. Sampl. Two Samplea Sample.. One Sample.. Two
Palllld Data
Factor Factol'l

On....mpl. t-teat Two... mpl. tot.at Pallll t-teat One·Way AWJVA Two-Way AWJVA
Figure 6

There are various stand alone, web-based, or MS Excel add- 11. Determine the action(s) to be taken.
on software packages that are available to perform hypothesis 12. Re-calculate the RPNs based on the action plans.
testing. One other advanced statistical method that is beyond the
scope of this paper but bears mentioning as part of the ISIIT that is In this phase, the tools presented help project participants
available for use on Six Sigma projects is Design of Experiments discover the root cause(s) of the problem and the relationship
(DOE). DOE is based on simultaneously testing multiple factors between variables and what some of the key process indicators
that affect a process, service, or product. It avoids the pitfalls are. The result of the Analyze Phase should leave us with several
normally associated with testing many factors at once by using opportunities for improvement to work with in the Improve
very specific combinations of factors in very specific patterns Phase.
(I).
Finally, one other computer-based decision tool that is used Improve Phase
extensively in the Analyze Phase is called a Failure Mode Effect
Analysis (FMEA). This quality-planning tool is used to identify Potential problem solutions are tested in the Improve Phase.
and eliminate potential product or process failures or defects. The task here is to select the best possible solution and then predict
It becomes a very valuable decision tool as it helps a manager the new capability of the process and/or value stream. One of the
systematically identify, analyze, prioritize and document best ways this is done is through a planned Design of Experiment
potential failure modes, their consequences on a system, product as mentioned previously. Essentially the Measurement and
or process, and prevent future failures from occurring. The twelve Analyze Phases are repeated with the new process in place and
steps to conduct an FMEA are as follows (1): the results compared to the previous process. These results are
typically presented to decision makers and the new process is put
l. Study the system, product or process to be analyzed. into place permanently.
2. Brainstorm, over multiple brainstorming sessions, the
possible range of failure modes. Control Phase
3. List the potential consequences of each failure mode.
4. Assign severity (SEV) scores for each failure mode. The objective in the Control Phase is to implement the
5. Identify the cause(s) of each failure mode. accepted solution and ensure that the gains derived from the
6. Assign occurrence (OCC) scores for each cause. improvement are maintained. The two basic tools here that
7. Identify current controls to detect the failure modes. lend themselves very well to ISIIT application are run charts
8. Assign an escaped detection (DET) score for each and Statistical Process Control (SPC) charts. Control charts are
cause and control. used to track a performance measure or key process indicator
9. Calculate the Risk Priority Number (RPN) for each variable over time. This is sometimes referred to as "listening to
line in the FMEA. the voice of the process." When a process measure is in control,
10. Prioritize the failure modes and causes based on the its behavior on a control chart exhibits random variation when
RPN. its distribution is stable. A distribution is said to be stable when

6 Journal of Computer Informatiou Systems Summer 2007

 , ,

Control Chart Feafures

'-
Pmca!llllTmnd

r
A /
~.
e
:::I
,
¥/ <, I
UCL'"
;
.i l/\rI\ /\ r: r"M , rr- I
1"1
Control Region ~ptures
Variation Natural to the
>-
=
"=
V\/ 1/

\\ \A /\ 1\
A"
;\ 1\ A I Original Process

:::I , v 'JyllV \/\/\/

0 \.. ~
\ "-
Let ...

" \
\
Control Lll1'lltsEstablished
_. ,
... , .ti

\: _u_
T: __

UCL • UpperCoritrQl Limit CL• center Line (mean) LCL~(owerCoritrQl Limit

Figure 7

its average and standard deviation is not shifting. A control chart design engineering transactional process. It is presented following
is used to distinguish random (or common cause) variation from the Six Sigma DMAIC format,
special cause variation and thus will indicate when action is to be The primary motivating factors in initiating this Lean/Six
taken and when action is not to be taken. They are designed to Sigma project were excessive time in getting a Technical In-
detect outliers, shifts, and trends. It is a run chart with statistically struction (TI) package approved (placed into Configuration
generated control limits. Management where Procurement can issue a purchase order)
Twenty data points are typically considered to be a sufficient and defects in TI package preparation. The purpose of this re-
number to establish control limits. Control limits are calculated port is to compile the results of this project into one concise
on the basis of the variation within the process and are typically document to be used as a reference for future Lean/Six Sigma
placed three standard deviations above and below the mean of activities.
the data. This placement marks the bounds of common cause or
random variation. In contrast, specifications limits represent the IMPROVEMENT GOALS
Voice of the Customer and are placed on capability charts but
not on control charts. Control charts represent the Voice of the • Improve quality and eliminate rework by standardizing
Process (I). Many manufacturing processes, such as chemical TI package development.
plating baths, use real-time control charts with touch screen • Solidify TI development/release schedule utilizing the
controls to give plating line operators up-to-the-minute status program's Integrated Master Schedule (lMS).
of the process. Alarms sound when an out of control situation • Reduce Engineering Development to Vault average
occurs so the operators can act quickly before there is a really cycle time from -40 days to <7 days by August 30, 2003
significant problem that interrupts production. Figure 7 shows the «70 days overall with Procurement added.).
components of a basic control chart. • Improve the program's overall funding expenditure
Once improvements to the process become permanent and profile by providing Tis to Procurement in a timely
stable, it can now be monitored closely to control the gains. manner.
Metrics can now be reported at the desired frequency and a
database can be built to use the data for future reference. The critical questions to discuss during each stage in the DMAIC
The following section presents highlights from a real world model are summarized in Figure 8 (next page).
completed Six Sigma project. Excerpts from each phase of the
DMAIC are shown with examples of ISIIT tools in action. Calculated Process Capability
and Baseline Sigma Level
ISIIT DEMONSTRATED
After collecting the initial process information, the next
This section will present examples of how IS/IT was used in step in the Measure Phase was to calculate the baseline process
a Six Sigma project at a Fortune 500 company. It focuses on a capability and sigma level. Both of these measures indicate how

Summer 2007 Journal of Computer Information Systems 7

 14 Questions for the DMAIC Model
w 1. Which value stream are you supporting? Who is the customer? (Intemal and extemal) Who is the value
Z stream owner? Who are the stakeholders?
u:::
W 2. Within the value stream, which process or processes have the highest priority for improvement?
C
3. How is the process perfonned? How does the value flow? What acti",ty is value-added and what is non
W value-added?
a:::
::)
4. What are the process perfonnance measures, (i.e., How will process improvement be gauged?) How
(I) accurate/precise is the measurement system?
<C 5. What are the customer-drisen requirements or specifications for all of the perfonnance measures? Are the
W
:IE process perfonnance measures in control? How capable is the process? What are the improvement goals
for the value stream or process perfonnance measures?

6. What kinds of waste and cost of poor quality exist in the value stream or process? What is the financial
and/or customer impact?
7. What are all the sources of variability in the value stream or process? Which of these are controlled by
W the stakeholders? How do we control them? What is the method of documenting and maintaining this
~
..oJ
control?
8. Are any sources of waste or variability supplier-dependent? If so, what are they, who are the suppliers,
<C and what effort is being made to eliminate waste and variability?
Z
<C
9. What are the key input variables that affect the average and de",ation of the measures of perfonnance?

10. What are the relationships between the measures of performance and the key input variables? Do any 0
the key input variables interact?
W
>
0
11. What settings or values for the key input Variables will optimize the measures of perfonnance?

a:::
a. 12. For the optimal settings of the key input variables, what kind of variability still exists in the perfonnance

-
:IE
..oJ
measures?

13. Has a process flow and control system been implemented to sustain the gains and continuously improve
0
a::: the process?
t-
Z 14. How much improvement has the value stream or process shown in the past six months? How much
0 time and/or money have our efforts sawd the company?
U
Figure 8

well the process is currently behaving with respect to the output fall completely within the specification limits. A value of 1.33 or
specifications. Process capability is usually expressed with two greater is usually desired.
indices, namely, Cpand Cpk. The sigma level is a commonly used measure of process
Cp is a capability index which shows the process capability capability that represents the number of standard deviations
potential but does not consider how centered (on the mean) the between the center of a process and the closest specification
process is. Cp may range in value from 0 to infinity, with a large limit. In other words, it compares the Voice of the Process
value indicating greater potential capability. A value of 1.33 or with the Voice of the Customer. It is correlated to the defect
greater is usually desired. rate and the complexity of the process. If a process is said to be
C pk is an index used to compare the natural tolerance of a Six Sigma Capable, our capability indices should read as fol-
process with the specification limits. Cpk has a value equal to Cp lows:
if the process is centered on the nominal; if Cpk is negative, the
process mean is outside the specification limits; if Cpk is between Cp 2: 2.0 (the higher the better)
o and I, then the natural tolerances of the process fall outside the Cpk 2: 1.5 (the higher the better)
specification limits. If Cpk is larger than I, the natural tolerances Defects Per Million (dpm) s 3.4 (the lower the better)

8 Journal of Computer Information Systems Summer 2007

Initial defect and yield rates were calculated with Cycle Time as more data to confirm this hypothesis. As a group we decided that
being the only defect measured (A TI was considered defective due to the complexity of the BAR process, working to improve
if the time from Engineering Release to Vaulting exceeded 14 it was an effort beyond the scope of this project and would be
days). These rates were calculated as follows: an excellent basis for a separate Lean/Six Sigma project. This
left the role of the Program Office Project Engineer as a place of
Proportion of Defective TI Packages focus.
One other observation we noticed was that many of the sub-
Number of Defective Tis processes were being performed in series. That is, the TI package
% Defective = No.ofTo~ITIPackages would go to one person for signature and then to the next person
and so on. Our second hypothesis was that the process would go
= 73/98 X 100 faster if we could perform several steps in parallel (at the same
time).
As the team looked at the data, we also began to brainstorm
= 74.5%
ideas on some of the possible causes of inefficiency and waste
within the process. The results of this effort could be shown in the
The process yield rate was therefore, I - the defect rate, or form of an Ishikawa (Fishbone) Diagram. It helped us to organize
25.5%. Figure 9 is a graphical representation of the capability our thoughts and focus our efforts on only those things we knew
of the TI Development Process as of the scheduled assessment we could control. Most of the factors that we could control fell
date. All capability indices were very unsatisfactory and a picture into the area of developing and following procedures and ensuring
based on actual data began to form and validate something we all personnel involved were trained to perform the tasks within
already intuitively knew was true. A complete analysis of the the process.
data with conclusions is provided in the Analysis Phase of this With this data analysis and hypotheses in hand, we moved
report. into the Improve Phase where we began to identify solutions to

11Development Process
Current State capability Measurement

Mean - 39.224
StdDev - 40 .803 . In spec
USL - 14 • Out spec left
LSL - 7
Sigma Capability - .C • Out spec righ t
Cpk - -.2061
. LSL
Cp - .0 28 6
DPM - 946,613 . USL
N - 98

g N
~
,
lO lO
...... r-.:
N
N
o
N
m
......
,
~
N ~ l"l
~ ~

Figure 9

In this case, a negative Cpt indicates that the majority of the improve the process. This proved to be the solution that worked
sample data was outside the specification limits (7-14 days) best. Advantages of this approach were as follows:
set up by the customer (red areas under the curve in Figure 2).
The blue area under the curve indicates the small proportion of • Approval of the various elements of a TI package
sample data that was actually within the customer's specification was done at the same time the package was being
range. The poor performance of this process is also confirmed written by the engineer. This helped us achieve our
by the low Sigma Capability and C p Another confirmation is goal of working more steps in parallel rather than in
the extremely high defect rate of 946,613 DPM (Defects Per series;
Million).
• In addition to reducing cycle time, the new process also
HYPOTHESES AND CONCLUSIONS gave more flexibility to the personnel performing the
tasks. TIs were worked as they were received rather
This information confirmed what we suspected were the than allowing them to pile up into batches. Continuous
best areas for improvement: I) the Budget Adjustment Request flow of work was achieved.
(BAR) process handled by the CAM and Business Management
Department and 2) the role of the Program Office Project Engineer • Quality metrics were developed and therefore we saw
in obtaining approvals from the various directors. These two areas a dramatic increase in TI package quality. The rework
appeared to be bottlenecks in the process, but we felt we needed virtually disappeared.

Summer 2007 Journal of Computer Information Systems 9

 Throughout the entire processes of conducting quality 10. Eckes, G. (2001) The Six Sigma Revolution, John Wiley and
improvement initiatives using Six Sigma methodologies, IS/ Sons, New York, NY.
IT have been heavily employed. The company deployed the 11. Harry, M. and Schroeder, R. (2000) Six Sigma: The
state-of-the-art information system infrastructure to support the Breakthrough Management Strategy Revolutionizing the
implementation ofthe Six-Sigma based models, and to disseminate World's Top Corporations, Doubleday, New York, NY.
results to all participating decision makers. Without the use of 12. Holtz, R. and Campbell, P. (2004) "Six Sigma: Its
such IS/IT, it is impossible to integrate so many complex tasks implementation in Ford's facility management and
in a seamless fashion and derive feasible solutions for quality maintenance functions," Journal ofFacilities Management,
improvement within such short period of time. March, Vol. 2 Issue 4, pp. 320-329.
13. Kiemele, M., Schmidt, S. and Berdine, R. (1997) Basic
CONCLUSION Statistics: Tools for Continuous Improvement, 4th Edition,
Air Academy Press, LLC, Colorado Springs, Colorado.
The Six Sigma methodology for quality improvement is 14. Knowles, G., Whicker, L., Femat, J., Canales, F. (2005) "A
a system of many concepts, tools, and principles. This paper conceptual model for the application of Six Sigma meth-
could not cover the entire breadth and depth of this subject. The odologies to supply chain improvement," International Jour-
enclosed case, nevertheless, clearly demonstrated the significant nal of Logistics: Research & Applications, March, Vol. 8
impact of ISIIT on the implementation of the Six Sigma approach. Issue 1, pp. 51-65.
The advent of personal computing technologies coupled with the 15. Lawton, L. (2004) "The Power of Ultimate Six Sigma:
advancement of networking systems and Internet infrastructure Keki Bhote's Proven System for Moving Beyond Quality
as well as the sophisticated software tools have contributed to Excellence to Total Business Excellence," Journal of
the success of implementing a Six Sigma quality control system, Organizational Excellence, Summer, Vol. 23 Issue 3, p108.
especially in small and medium businesses. 16. Linderman, K., Schroeder, R., Zaheer, S., Choo, A. (2003)
It is hoped that the focus on ISIIT shows how powerful Six "Six Sigma: a goal-theoretic perspective," Journal of
Sigma can be when aided by the latest computer hardware and Operations Management, March, Vol. 21 Issue 2, pp. 193-
software technology. The strategic use of IS/IT within the Six 203.
Sigma methodology will go a long way toward making things 17. Mcadam, R. and Lafferty, B. (2004) "A multilevel case study
easy for professionals to incorporate the principles as part of their critique of six sigma: statistical control or strategic change?"
everyday work routine. International Journal of Operations & Production Man-
agement, Vol. 24 Issue 5, pp. 530-549.
REFERENCES 18. Pande, P., Neuman, R., Cavanagh, R. (2000a) The Six Sigma
Way: How GE, Motorola, and Other Top Companies are
1. Adams, M., Kiernele, M., Pollack, L., and Quan, T. (2003) Honing Their Performance, McGraw Hill, New York.
Lean Six Sigma: A Tools Guide, Air Academy Associates, 19. Pande, P., Neuman, R., Cavanagh, R. (2000b) The Six Sigma
Colorado Springs, Colorado. Way Team Field Book, McGraw Hill, New York.
2. Albright, T. and Lam, M. (2006) "Managerial Accounting 20. Raisinghani, M., Ette, H., Pierce, R., Cannon, G., Daripaly,
and Continuous Improvement Initiatives: A Retrospective P. (2005) "Six Sigma: concepts, tools, and applications," In-
and Framework," Journal of Managerial Issues, Summer, dustrial Management & Data Systems, Vol. 105 Issue 4, pp.
Vol. 18 Issue 2, pp. 157-174. 491-505.
3. Box, T. (2006) "Six Sigma Quality: Experiential Learning," 21. Revere, L. and Black, K. (2003) "Integrating Six Sigma with
SAM Advanced Management Journal, Winter, Vol. 71 Issue total Quality Management: A Case Example for Measuring
1, pp. 20-23. Medication Errors," Journal of Healthcare Management;
4. Brue, G. (2002) Six Sigma for Managers, McGraw Hill, November/December, Vol. 48 Issue 6, pp. 377-391.
New York, New York. 22. Rylander, D. and Provost, T. (2006) "Improving the Odds:
5. Chowdhury, S. (2002) Design For Six Sigma, Dearborn Combining Six Sigma and Online Market Research for Better
Trade Publishing Co., Chicago, Illinois. Customer Service," SAM Advanced Management Journal,
6. Creveling, c., Slutsky, J., Antis, Jr., D. (2003) Design for Six Winter, Vol. 71 Issue 1, pp. 13-19.
Sigma in Technology and Product Development, Prentiss 23. Thawani, S. (2004) "Six Sigma--Strategy for Organizational
Hall, Upper Saddle River, New Jersey. Excellence," Total Quality Management & Business Excel-
7. de Mast, J. (2006) "Six Sigma and competitive advantage," lence, July/August, Vol. 15 Issue 5/6, pp. 655-664.
Total Quality Management & Business Excellence, May, 24. Voelkel, J. (2005) "What Makes a Six Sigma Project
Vol. 17 Issue 4, pp. 455-464. Successful?" Quality Progress, May, Vol. 38 Issue 5, pp. 66-
8. Dedhia, N. (2005) "Six sigma basics," Total Quality 68.
Management & Business Excellence, July, Vol. 16 Issue 5, 25. Wyper, B. and Harrison, A. (2000) "Deployment of Six Sigma
pp.567-574. methodology in Human Resource function: a case study,"
9. Donnelly, S. (2006) "Lean and Mean", Time, July 10,2006, Total Quality Management, July, Vol. 11 Issue 4/5/6, pp.
pp. A1-A4. S720-5727.

10 Journal of Computer Information Systems Summer 2007