software design for six sigma

www.it-ebooks.info
P1: OSO
fm JWBS034-El-Haik July 20, 2010 20:52 Printer Name: Yet to Come
SOFTWARE DESIGN
FOR SIX SIGMA
A Roadmap for Excellence
BASEM EL-HAIK
ADNAN SHAOUT
A JOHN WILEY & SONS, INC., PUBLICATION
www.it-ebooks.info
P1: OSO
fm JWBS034-El-Haik July 20, 2010 20:52 Printer Name: Yet to Come
SOFTWARE DESIGN
FOR SIX SIGMA
www.it-ebooks.info
P1: OSO
fm JWBS034-El-Haik July 20, 2010 20:52 Printer Name: Yet to Come
www.it-ebooks.info
P1: OSO
fm JWBS034-El-Haik July 20, 2010 20:52 Printer Name: Yet to Come
SOFTWARE DESIGN
FOR SIX SIGMA
A Roadmap for Excellence
BASEM EL-HAIK
ADNAN SHAOUT
A JOHN WILEY & SONS, INC., PUBLICATION
www.it-ebooks.info
P1: OSO
fm JWBS034-El-Haik July 20, 2010 20:52 Printer Name: Yet to Come
Copyright
C


2010 by John Wiley & Sons, Inc. All rights reserved.
Published by John Wiley & Sons, Inc., Hoboken, New Jersey.
Published simultaneously in Canada.
No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or
by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as
permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior
written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to
the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, (978) 750-8400,
fax (978) 750-4470, or on the web at www.copyright.com. Requests to the Publisher for permission
should be addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken,
NJ 07030, (201) 748-6011, fax (201) 748-6008, or online at />Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in
preparing this book, they make no representations or warranties with respect to the accuracy or
completeness of the contents of this book and specifically disclaim any implied warranties of
merchantability or fitness for a particular purpose. No warranty may be created or extended by sales
representatives or written sales materials. The advice and strategies contained herein may not be suitable
for your situation. You should consult with a professional where appropriate. Neither the publisher nor
author shall be liable for any loss of profit or any other commercial damages, including but not limited to
special, incidental, consequential, or other damages.
For general information on our other products and services or for technical support, please contact our
Customer Care Department within the United States at (800) 762-2974, outside the United States at
(317) 572-3993 or fax (317) 572-4002.
Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may
not be available in electronic format. For more information about Wiley products, visit our web site at
www.wiley.com
Library of Congress Cataloging-in-Publication Data
El-Haik, Basem.
Software design for six sigma : a roadmap for excellence / Basem S. El-Haik, Adnan Shaout.
p. cm.
ISBN 978-0-470-40546-8 (hardback)

1. Computer software–Quality control. 2. Six sigma (Quality control standard) I. Shaout,
Adnan, 1960– II. Title.
QA76.76.Q35E45 2010
005.1–dc22 2010025493
Printed in Singapore
10987654321
www.it-ebooks.info
P1: OSO
fm JWBS034-El-Haik July 20, 2010 20:52 Printer Name: Yet to Come
To our parents, families, and friends for their continuous support.
www.it-ebooks.info
P1: OSO
fm JWBS034-El-Haik July 20, 2010 20:52 Printer Name: Yet to Come
www.it-ebooks.info
P1: OSO
fm JWBS034-El-Haik July 20, 2010 20:52 Printer Name: Yet to Come
CONTENTS
PREFACE xv
ACKNOWLEDGMENTS xix
1 SOFTWARE QUALITY CONCEPTS 1
1.1 What is Quality / 1
1.2 Quality, Customer Needs, and Functions / 3
1.3 Quality, Time to Market, and Productivity / 5
1.4 Quality Standards / 6
1.5 Software Quality Assurance and Strategies / 6
1.6 Software Quality Cost / 9
1.7 Software Quality Measurement / 13
1.8 Summary / 19
References / 20
2 TRADITIONAL SOFTWARE DEVELOPMENT PROCESSES 21

2.1 Introduction / 21
2.2 Why Software Developmental Processes? / 22
2.3 Software Development Processes / 23
2.4 Software Development Processes Classification / 46
2.5 Summary / 53
References / 53
vii
www.it-ebooks.info
P1: OSO
fm JWBS034-El-Haik July 20, 2010 20:52 Printer Name: Yet to Come
viii CONTENTS
3 DESIGN PROCESS OF REAL-TIME OPERATING
SYSTEMS (RTOS) 56
3.1 Introduction / 56
3.2 RTOS Hard versus Soft Real-Time Systems / 57
3.3 RTOS Design Features / 58
3.4 Task Scheduling: Scheduling Algorithms / 66
3.5 Intertask Communication and Resource Sharing / 72
3.6 Timers / 74
3.7 Conclusion / 74
References / 75
4 SOFTWARE DESIGN METHODS AND REPRESENTATIONS 77
4.1 Introduction / 77
4.2 History of Software Design Methods / 77
4.3 Software Design Methods / 79
4.4 Analysis / 85
4.5 System-Level Design Approaches / 88
4.6 Platform-Based Design / 96
4.7 Component-Based Design / 98
4.8 Conclusions / 99

References / 100
5 DESIGN FOR SIX SIGMA (DFSS) SOFTWARE
MEASUREMENT AND METRICS 103
5.1 Introduction / 103
5.2 Software Measurement Process / 105
5.3 Software Product Metrics / 106
5.4 GQM (Goal–Question–Metric) Approach / 113
5.5 Software Quality Metrics / 115
5.6 Software Development Process Metrics / 116
5.7 Software Resource Metrics / 117
5.8 Software Metric Plan / 119
References / 120
6 STATISTICAL TECHNIQUES IN SOFTWARE SIX SIGMA
AND DESIGN FOR SIX SIGMA (DFSS) 122
6.1 Introduction / 122
6.2 Common Probability Distributions / 124
6.3 Software Statistical Methods / 124
www.it-ebooks.info
P1: OSO
fm JWBS034-El-Haik July 20, 2010 20:52 Printer Name: Yet to Come
CONTENTS ix
6.4 Inferential Statistics / 134
6.5 A Note on Normal Distribution and Normality Assumption / 142
6.6 Summary / 144
References / 145
7 SIX SIGMA FUNDAMENTALS 146
7.1 Introduction / 146
7.2 Why Six Sigma? / 148
7.3 What is Six Sigma? / 149
7.4 Introduction to Six Sigma Process Modeling / 152

7.5 Introduction to Business Process Management / 154
7.6 Six Sigma Measurement Systems Analysis / 156
7.7 Process Capability and Six Sigma Process Performance / 157
7.8 Overview of Six Sigma Improvement (DMAIC) / 161
7.9 DMAIC Six Sigma Tools / 163
7.10 Software Six Sigma / 165
7.11 Six Sigma Goes Upstream—Design For Six Sigma / 168
7.12 Summary / 169
References / 170
8 INTRODUCTION TO SOFTWARE DESIGN FOR
SIX SIGMA (DFSS) 171
8.1 Introduction / 171
8.2 Why Software Design for Six Sigma? / 173
8.3 What is Software Design For Six Sigma? / 175
8.4 Software DFSS: The ICOV Process / 177
8.5 Software DFSS: The ICOV Process In Software Development / 179
8.6 DFSS versus DMAIC / 180
8.7 A Review of Sample DFSS Tools by ICOV Phase / 182
8.8 Other DFSS Approaches / 192
8.9 Summary / 193
8.A.1 Appendix 8.A (Shenvi, 2008) / 194
8.A.2 DIDOVM Phase: Define / 194
8.A.3 DIDOVM Phase: Identify / 196
8.A.4 DIDOVM Phase: Design / 199
8.A.5 DIDOVM Phase: Optimize / 203
8.A.6 DIDOVM Phase: Verify / 204
8.A.7 DIDOVM Phase: Monitor / 204
References / 205
www.it-ebooks.info
P1: OSO

fm JWBS034-El-Haik July 20, 2010 20:52 Printer Name: Yet to Come
x CONTENTS
9 SOFTWARE DESIGN FOR SIX SIGMA (DFSS):
A PRACTICAL GUIDE FOR SUCCESSFUL DEPLOYMENT 207
9.1 Introduction / 207
9.2 Software Six Sigma Deployment / 208
9.3 Software DFSS Deployment Phases / 208
9.4 Black Belt and DFSS Team: Cultural Change / 234
References / 238
10 DESIGN FOR SIX SIGMA (DFSS) TEAM AND TEAM
SOFTWARE PROCESS (TSP) 239
10.1 Introduction / 239
10.2 The Personal Software Process (PSP) / 240
10.3 The Team Software Process (TSP) / 243
10.4 PSP and TSP Deployment Example / 245
10.5 The Relation of Six Sigma to CMMI/PSP/TSP
for Software / 269
References / 294
11 SOFTWARE DESIGN FOR SIX SIGMA (DFSS) PROJECT
ROAD MAP 295
11.1 Introduction / 295
11.2 Software Design For Six Sigma Team / 297
11.3 Software Design For Six Sigma Road Map / 300
11.4 Summary / 310
12 SOFTWARE QUALITY FUNCTION DEPLOYMENT 311
12.1 Introduction / 311
12.2 History of QFD / 313
12.3 QFD Overview / 314
12.4 QFD Methodology / 314
12.5 HOQ Evaluation / 318

12.6 HOQ 1: The Customer’s House / 318
12.7 Kano Model / 319
12.8 QFD HOQ 2: Translation House / 321
12.9 QFD HOQ3—Design House / 324
12.10 QFD HOQ4—Process House / 324
12.11 Summary / 325
References / 325
www.it-ebooks.info
P1: OSO
fm JWBS034-El-Haik July 20, 2010 20:52 Printer Name: Yet to Come
CONTENTS xi
13 AXIOMATIC DESIGN IN SOFTWARE DESIGN FOR
SIX SIGMA (DFSS) 327
13.1 Introduction / 327
13.2 Axiomatic Design in Product DFSS:
An Introduction / 328
13.3 Axiom 1 in Software DFSS / 338
13.4 Coupling Measures / 349
13.5 Axiom 2 in Software DFSS / 352
References / 354
Bibliography / 355
14 SOFTWARE DESIGN FOR X 356
14.1 Introduction / 356
14.2 Software Reliability and Design For Reliability / 357
14.3 Software Availability / 379
14.4 Software Design for Testability / 380
14.5 Design for Reusability / 381
14.6 Design for Maintainability / 382
References / 386
Appendix References / 387

Bibliography / 387
15 SOFTWARE DESIGN FOR SIX SIGMA (DFSS) RISK
MANAGEMENT PROCESS 388
15.1 Introduction / 388
15.2 Planning for Risk Management Activities in Design and
Development / 393
15.3 Software Risk Assessment Techniques / 394
15.4 Risk Evaluation / 400
15.5 Risk Control / 403
15.6 Postrelease Control / 404
15.7 Software Risk Management Roles and
Responsibilities / 404
15.8 Conclusion / 404
References / 407
16 SOFTWARE FAILURE MODE AND EFFECT
ANALYSIS (SFMEA) 409
16.1 Introduction / 409
16.2 FMEA: A Historical Sketch / 412
www.it-ebooks.info
P1: OSO
fm JWBS034-El-Haik July 20, 2010 20:52 Printer Name: Yet to Come
xii CONTENTS
16.3 SFMEA Fundamentals / 420
16.4 Software Quality Control and Quality Assurance / 431
16.5 Summary / 434
References / 434
17 SOFTWARE OPTIMIZATION TECHNIQUES 436
17.1 Introduction / 436
17.2 Optimization Metrics / 437
17.3 Comparing Software Optimization Metrics / 442

17.4 Performance Analysis / 453
17.5 Synchronization and Deadlock Handling / 455
17.6 Performance Optimization / 457
17.7 Compiler Optimization Tools / 458
17.8 Conclusion / 464
References / 464
18 ROBUST DESIGN FOR SOFTWARE DEVELOPMENT 466
18.1 Introduction / 466
18.2 Robust Design Overview / 468
18.3 Robust Design Concept #1: Output Classification / 471
18.4 Robust Design Concept #2: Quality Loss Function / 472
18.5 Robust Design Concept #3: Signal, Noise, and
Control Factors / 475
18.6 Robustness Concept #4: Signal–to-Noise Ratios / 479
18.7 Robustness Concept #5: Orthogonal Arrays / 480
18.8 Robustness Concept #6: Parameter Design Analysis / 483
18.9 Robust Design Case Study No. 1: Streamlining of Debugging
Software Using an Orthogonal Array / 485
18.10 Summary / 491
18.A.1 ANOVA Steps For Two Factors Completely Randomized
Experiment / 492
References / 496
19 SOFTWARE DESIGN VERIFICATION AND VALIDATION 498
19.1 Introduction / 498
19.2 The State of V&V Tools for Software DFSS Process / 500
19.3 Integrating Design Process with Validation/Verification
Process / 502
19.4 Validation and Verification Methods / 504
www.it-ebooks.info
P1: OSO

fm JWBS034-El-Haik July 20, 2010 20:52 Printer Name: Yet to Come
CONTENTS xiii
19.5 Basic Functional Verification Strategy / 515
19.6 Comparison of Commercially Available Verification and
Validation Tools / 517
19.7 Software Testing Strategies / 520
19.8 Software Design Standards / 523
19.9 Conclusion / 525
References / 525
INDEX 527
www.it-ebooks.info
P1: OSO
fm JWBS034-El-Haik July 20, 2010 20:52 Printer Name: Yet to Come
www.it-ebooks.info
P1: OSO
fm JWBS034-El-Haik July 20, 2010 20:52 Printer Name: Yet to Come
PREFACE
Information technology (IT) quality engineering and quality improvement methods
are constantly getting more attention from world corporate leaders, all levels of
management, design engineers, and academia. This trend can be seen easily by the
widespread of “Six Sigma” initiatives in many Fortune IT 500 companies. For a
Six Sigma initiative in IT, software design activity is the most important to achieve
significant quality and reliability results. Because design activities carry a big portion
of software development impact, quality improvements done in design stages often
will bring the most impressive results. Patching up quality problems in post-design
phases usually is inefficient and very costly.
During the last 20 years, there have been significant enhancements in software
development methodologies for quality improvement in software design; those meth-
ods include the Waterfall Model, Personal Software Process (PSP), Team Software
Process (TSP), Capability Maturity Model (CMM), Software Process Improvement

Capability Determination (SPICE), Linear Sequential Model, Prototyping Model,
RAD Model, and Incremental Model, among others.
1
The historical evolution of
these methods and processes, although indicating improvement trends, indicates gaps
that each method tried to pick up where its predecessors left off while filling the gaps
missed in their application.
Six Sigma is a methodology to manage process variations that use data and
statistical analysis to measure and improve a company’s operational performance. It
works by identifying and eliminating defects in manufacturing and service-related
processes. The maximum permissible defects are 3.4 per one million opportunities.
2
1
See Chapters 2 and 4.
2
See Chapter 6.
xv
www.it-ebooks.info
P1: OSO
fm JWBS034-El-Haik July 20, 2010 20:52 Printer Name: Yet to Come
xvi PREFACE
Although Six Sigma is manufacturing-oriented, its application to software problem
solving is undisputable because as you may imagine, there are problems that need to
be solved in software and IT domains. However, the real value is in prevention rather
than in problem solving, hence, software Design For Six Sigma (DFSS).
DFSS is very vital to software design activities that decide quality, cost, and
cycle time of the software and can be improved greatly if the right strategy and
methodologies are used. Major IT corporations are training many software design
engineers and project leaders to become Six Sigma Black Belts, or Master Black
Belts, enabling them to play the leader role in corporate excellence.

Our book, Software Design For Six Sigma: A Roadmap for Excellence, constitutes
an algorithm of software design
3
using the design for Six Sigma thinking, tools, and
philosophy to software design. The algorithm also will include conceptual design
frameworks, mathematical derivation for Six Sigma capability upfront to enable
design teams to disregard concepts that are not capable upfront . . . learning the
software development cycle and saving developmental costs.
DFSS offers engineers powerful opportunities to develop more successful systems,
software, hardware, and processes. In applying Design for Six Sigma to software
systems, two leading experts offer a realistic, step-by-step process for succeeding with
DFSS. Their clear, start-to-finish road map is designed for successfully developing
complex high-technology products and systems.
Drawing on their unsurpassed experience leading DFSS and Six Sigma in de-
ployment in Fortune 100 companies, the authors cover the entire software DFSS
project life cycle, from business case through scheduling, customer-driven require-
ments gathering through execution. They provide real-world experience for applying
their techniques to software alone, hardware alone, and systems composed of both.
Product developers will find proven job aids and specific guidance about what teams
and team members need to do at every stage. Using this book’s integrated, systems
approach, marketers and software professionals can converge all their efforts on what
really matters: addressing the customer’s true needs.
The uniqueness of this book is bringing all those methodologies under the umbrella
of design and giving a detailed description about how those methods, QFD,
4
robust
design methods,
5
software failure mode and effect analysis (SFMEA),
6

Design for
X,
7
and axiomatic design
8
can be used to help quality improvements in software
development, what kinds of different roles those methods play in various stages of
design, and how to combine those methods to form a comprehensive strategy, a design
algorithm, to tackle any quality issues during the design stage.
This book is not only helpful for software quality assurance professionals, but
also it will help design engineers, project engineers, and mid-level management to
3
See Chapter 11.
4
Chapter 12.
5
Chapter 18.
6
Chapter 16.
7
Design for X-ability includes reliability, testability, reusability, availability, etc. See Chapter 14 for more
details.
8
Chapter 13.
www.it-ebooks.info
P1: OSO
fm JWBS034-El-Haik July 20, 2010 20:52 Printer Name: Yet to Come
PREFACE xvii
gain fundamental knowledge about software Design for Six Sigma. After reading this
book, the reader could gain the entire body knowledge for software DFSS. So this

book also can be used as a reference book for all software Design for Six Sigma-
related people, as well as training material for a DFSS Green Belt, Black Belt, or
Master Black Belt.
We believe that this book is coming at the right time because more and more IT
companies are starting DFSS initiatives to improve their design quality.
Your comments and suggestions to this book are greatly appreciated. We will give
serious consideration to your suggestions for future editions. Also, we are conducting
public and in-house Six Sigma and DFSS workshops and provide consulting services.
Dr. Basem El-Haik can be reached via e-mail:

Dr. Adnan Shaout can be reached via e-mail:

www.it-ebooks.info
P1: OSO
fm JWBS034-El-Haik July 20, 2010 20:52 Printer Name: Yet to Come
www.it-ebooks.info
P1: OSO
fm JWBS034-El-Haik July 20, 2010 20:52 Printer Name: Yet to Come
ACKNOWLEDGMENTS
In preparing this book we received advice and encouragement from several people.
For this we are thankful to Dr. Sung-Hee Do of ADSI for his case study contribution
in Chapter 13 and to the editing staff of John Wiley & Sons, Inc.
xix
www.it-ebooks.info
P1: OSO
fm JWBS034-El-Haik July 20, 2010 20:52 Printer Name: Yet to Come
www.it-ebooks.info
P1: JYS
c01 JWBS034-El-Haik July 20, 2010 14:44 Printer Name: Yet to Come
CHAPTER 1

SOFTWARE QUALITY CONCEPTS
1.1 WHAT IS QUALITY
The American Heritage Dictionary defines quality as “a characteristic or attribute of
something.” Quality is defined in the International Organization for Standardization
(ISO) publications as the totality of characteristics of an entity that bear on its ability
to satisfy stated and implied needs.
Quality is a more intriguing concept than it seems to be. The meaning of the
term “Quality” has evolved over time as many concepts were developed to improve
product or service quality, including total quality management (TQM), Malcolm
Baldrige National Quality Award, Six Sigma, quality circles, theory of constraints
(TOC),Quality Management Systems (ISO 9000 and ISO 13485), axiomatic quality
(El-Haik, 2005), and continuous improvement. The following list represents the
various interpretations of the meaning of quality:
r
“Quality: an inherent or distinguishing characteristic, a degree or grade of ex-
cellence” (American Heritage Dictionary, 1996).
r
“Conformance to requirements” (Crosby, 1979).
r
“Fitness for use” (Juran & Gryna, 1988).
r
“Degree to which a set of inherent characteristic fulfills requirements”
ISO 9000.
Software Design for Six Sigma: A Roadmap for Excellence, By Basem El-Haik and Adnan Shaout
Copyright
C

2010 John Wiley & Sons, Inc.
1
www.it-ebooks.info

P1: JYS
c01 JWBS034-El-Haik July 20, 2010 14:44 Printer Name: Yet to Come
2 SOFTWARE QUALITY CONCEPTS
r
“Value to some person” (Weinberg).
r
“The loss a product imposes on society after it is shipped” (Taguchi).
r
“The degree to which the design vulnerabilities do not adversely affect product
performance” (El-Haik, 2005).
Quality is a characteristic that a product or service must have. It refers to the
perception of the degree to which the product or service meets the customer’s ex-
pectations. Quality has no specific meaning unless related to a specific function or
measurable characteristic. The dimensions of quality refer to the measurable char-
acteristics that the quality achieves. For example, in design and development of a
medical device:
r
Quality supports safety and performance.
r
Safety and performance supports durability.
r
Durability supports flexibility.
r
Flexibility supports speed.
r
Speed supports cost.
You can easily build theinterrelationship between quality andallaspects of product
characteristics, as these characteristics act as the qualities of the product. However,
not all qualities are equal. Some are more important than others. The most important
qualities are the ones that customers want most. These are the qualities that products

and services must have. So providing quality products and services is all about
meeting customer requirements. It is all about meeting the needs and expectations of
customers.
When the word “quality” is used, we usually think in terms of an excellent design
or service that fulfil’s or exceeds our expectations. When a product design surpasses
our expectations, we consider that its quality is good. Thus, quality is related to
perception. Conceptually, quality can be quantified as follows (El-Haik & Roy, 2005):
Q =

P

E
(1.1)
where Q is quality, P is performance, and E is an expectation.
In a traditional manufacturing environment, conformance to specification and
delivery are the common quality items that are measured and tracked. Often, lots are
rejected because they do not have the correct documentation supporting them. Quality
in manufacturing then is conforming product, delivered on time, and having all of the
supporting documentation. In design, quality is measured as consistent conformance
to customer expectations.
www.it-ebooks.info
P1: JYS
c01 JWBS034-El-Haik July 20, 2010 14:44 Printer Name: Yet to Come
QUALITY, CUSTOMER NEEDS, AND FUNCTIONS 3
X
µ(X)
1
0
K
FIGURE 1.1 A membership function for an affordable software.

1
In general, quality
2
is a fuzzy linguistic variable because quality can be very
subjective. What is of a high quality to someone might not be a high quality to
another. It can be defined with respect to attributes such as cost or reliability. It is a
degree of membership of an attribute or a characteristic that a product or software
can or should have. For example, a product should be reliable, or a product should
be both reliable and usable, or a product should be reliable or repairable. Similarly,
software should be affordable, efficient, and effective. These are some characteristics
that a good quality product or software must have. In brief, quality is a desirable
characteristic that is subjective. The desired qualities are the ones that satisfy the
functional and nonfunctional requirements of a project. Figure 1.1 shows a possible
membership function, µ(X), for the affordable software with respect to the cost (X).
When the word “quality” is used in describing a software application or any
product, it implies a product or software program that you might have to pay more
for or spend more time searching to find.
1.2 QUALITY, CUSTOMER NEEDS, AND FUNCTIONS
The quality of a software product for a customer is a product that meets or exceeds
requirements or expectations. Quality can be achieved through many levels (Braude,
1
where Kis themax costvalueof thesoftware afterwhich thesoftware will benot beaffordable (µ(K) = 0).
2
J. M. Juran (1988) defined quality as “fitness for use.” However, other definitions are widely discussed.
Quality as “conformance to specifications” is a position that people in the manufacturing industry often
promote. Others promote wider views that include the expectations that the product or service being deliv-
ered 1) meets customer standards, 2) meets and fulfills customer needs, 3) meets customer expectations,
and 4) will meet unanticipated future needs and aspirations.
www.it-ebooks.info