Upload
others
View
3
Download
0
Embed Size (px)
Citation preview
University of Central Florida University of Central Florida
STARS STARS
Electronic Theses and Dissertations, 2004-2019
2016
A New Six Sigma Implementation Approach For Power Generation A New Six Sigma Implementation Approach For Power Generation
Gas Turbines Repair Process Development Gas Turbines Repair Process Development
Somesh Ghunakikar University of Central Florida
Part of the Industrial Engineering Commons
Find similar works at: https://stars.library.ucf.edu/etd
University of Central Florida Libraries http://library.ucf.edu
This Doctoral Dissertation (Open Access) is brought to you for free and open access by STARS. It has been accepted
for inclusion in Electronic Theses and Dissertations, 2004-2019 by an authorized administrator of STARS. For more
information, please contact [email protected].
STARS Citation STARS Citation Ghunakikar, Somesh, "A New Six Sigma Implementation Approach For Power Generation Gas Turbines Repair Process Development" (2016). Electronic Theses and Dissertations, 2004-2019. 4968. https://stars.library.ucf.edu/etd/4968
A NEW SIX SIGMA IMPLEMENTATION APPROACH FOR POWER
GENERATION GAS TURBINES REPAIR PROCESS DEVELOPMENT
by
SOMESH JANARDAN GHUNAKIKAR
B.E. Mechanical Engineering, Shivaji University, India, 1998
M.E. Mechanical Engineering, Shivaji University, India, 2000
M.S. Aeronautics and Astronautics, Purdue University, USA, 2013
A dissertation submitted in partial fulfillment of the requirements
for the degree of Doctor of Philosophy
in the Department of Industrial Engineering and Management Systems
in the College of Engineering and Computer Science
at the University of Central Florida
Orlando, Florida
Spring Term
2016
Major Professor: Ahmad Elshennawy
ii
ABSTRACT
Power Generation gas turbines used for heavy duty application mainly constitutes three
modules; compressor, combustion and turbine. Typically, all these parts are designed by OEM
companies for specific number of hours and cycles (also known as starts) before they become
dysfunctional. In addition, Gas Turbine (GT) also have intended repair interval depending upon
the type of part application and anticipated damages during service operation. Thus, GT parts
need inspections and repair (overhaul) after certain operating hours in order to recondition them
so that they can be fit for reoperation to produce power. In this dissertation, a unique six sigma
DFSS approach for development of GT parts overhaul is presented for total quality
improvement. In this dissertation report, a unique six sigma DFSS approach is presented
applicable to the development of repair processes for GT parts that can be used during
overhauling of the parts. All six sigma phases of the proposed DFSS approach along with repair
product development cycle are discussed. Various six sigma tools which yield significant
benefits for the process users are also discussed. Importantly, a statistical probabilistic life
analysis approach is proposed in order to verify the structural integrity of a repaired GT part.
Finally a case study of GT axial compressor diaphragms (stators) to illustrate various phases and
six sigma tools usage during each phase of the DFSS approach is discussed. The overall
significant benefit of the proposed DFSS approach was to achieve total quality improvement to
deliver final GT repair process, faster repair development cycle and end customer satisfaction.
iii
ACKNOWLEDGMENTS
I would like to thank my academic advisor Dr. Ahmad Elshennawy for his support and
guidance throughout various stages of this research.
Special thanks to Drs. Luis Rabelo, William Thompson and Sandra Furterer who served
on my dissertation committee and enriched this learning experience with their broad knowledge
and valuable support.
Finally, I would like to thank my family and friends members who have supported me
throughout this long journey. Special thanks to my father Dr. J.T. Ghunakikar for his constant
support during my education.
iv
TABLE OF CONTENTS
LIST OF FIGURES ..................................................................................................................................... vi
LIST OF TABLES ..................................................................................................................................... viii
LIST OF ACRONYMS AND ABBREVIATIONS ..................................................................................... ix
CHAPTER 1 INTRODUCTION .................................................................................................................. 1
1.1 Problem Statement ........................................................................................................................ 2
1.2 Difference Between Traditional DFSS And Proposed DFSS Approach ...................................... 3
1.3 Expected Benefits Of Proposed DFSS Approach ......................................................................... 5
1.4 Contribution To The Proposed DFSS Approach .......................................................................... 5
CHAPTER 2 SIX SIGMA LITERATURE REVIEW .................................................................................. 6
2.1 Methods For Literature Survey ........................................................................................................... 6
2.2 Summary Of Literature Survey ......................................................................................................... 10
2.3 Key Finding For Literature Survey ................................................................................................... 27
2.4 Literature Survey Conclusion ........................................................................................................... 35
CHAPTER 3 POWER GENERATION PARTS AND FAILURE MODES .............................................. 37
3.1 Introduction To Power Generation GT Parts .................................................................................... 37
3.2 GT Parts Familiarization And Repair Products ................................................................................ 38
3.3 GT Parts And Failure Modes ............................................................................................................ 40
3.3.1 Compressor Section ................................................................................................................... 40
3.3.2 Turbine Section .......................................................................................................................... 46
3.3.3 Combustion Section.................................................................................................................... 52
3.4 GT Parts Overhaul And Repair Importance And Role Of Six Sigma ............................................... 55
CHAPTER 4 METHODOLOGY ............................................................................................................... 56
4.1 Outline Of The Methodology ............................................................................................................ 56
4.2 Methodology Description ................................................................................................................. 57
4.3 Six Sigma Tools Used During Various Phases of New DFSS Approach ......................................... 60
v
CHAPTER 5 CASE STUDY: GT AXIAL COMPRESSOR DIAPHGRAM/STATOR REPAIR ............. 72
5.1 Case Study- Introduction ............................................................................................................ 72
5.2 Case Study - IPDcI Framework .................................................................................................. 76
5.3 Identify Phase .............................................................................................................................. 77
5.4 Prioritize Phase ........................................................................................................................... 79
5.5 Comprehensive Design Phase ..................................................................................................... 80
5.6 Proof Of Concept- Fixed Beam Example ................................................................................... 95
5.7 Benefits Of The Proposed PLA Method ......................................................................................... 102
5.8 Implement Phase ............................................................................................................................. 103
CHAPTER 6 CONCLUSIONS ................................................................................................................ 104
CHAPTER 7 RECCOMMENDATIONS FOR FUTURE WORK ........................................................... 105
REFERENCES ......................................................................................................................................... 106
vi
LIST OF FIGURES
Figure 1 Industrial Engineering Journals Surveyed Related to the Topic .................................................... 8
Figure 2 Impact Factors - Industrial Engineering Journals Surveyed Related to the Topic ......................... 9
Figure 3 A Typical Combined Cycle Power Plant Layout ......................................................................... 37
Figure 4 Typical Large Gas Turbine 3D Section (source Siemens PG website) ........................................ 38
Figure 5 Example Large Gas Turbine- Assembly (source Siemens PG website) ....................................... 39
Figure 6 Example Compressor Cross Section LGT (source gasturbinetutorial.blogspot.com-604) ........... 41
Figure 7 Example Compressor Diaphragms or Stators LGT (source ccj-online) ....................................... 42
Figure 8 Example of Failure Modes of Compressor Diaphragms or Stators (source ccj-online) ............... 43
Figure 9 Example of Compressor Seal Holder (source Siemens PG website) ............................................ 44
Figure 10 Example, Failure Modes of Compressor Seal Holders (source Siemens PG website) ............... 44
Figure 11 Example of a Compressor Casing (source oddstuffmagazine.com) ........................................... 45
Figure 12 Example, Failure Modes of a Compressor Casing (source ccj-online) ...................................... 45
Figure 13 Example Typical Cross Section of Turbine LGT (source etspower.com) .................................. 46
Figure 14 Example, Failure Modes of Turbine Casing (source Siemens PG website) ............................... 47
Figure 15 Example, Typical Inter Stage Seal Housing (source Siemens PG website) ............................... 47
Figure 16 Example, Failure Modes of Inter Stage Seal Housing (source Siemens PG website) ................ 48
Figure 17 Example of a Double Airfoil Turbine Vane (source liberdi.com) .............................................. 48
Figure 18 Example of Single Airfoil Turbine Vane (source liberdi.com) .................................................. 49
Figure 19 Example, Failure Modes of a Turbine Vane (sourc ccj-online) ................................................. 49
Figure 20 Example of Unshrouded Turbine Blade (source ge.com)........................................................... 50
Figure 21 Example of Shrouded Turbine Blade (source gasturbinepower.ASME) .................................... 50
Figure 22 Example, Failure Modes of a Turbine Blade- Platform, Angel Wing (source Siemens PG
website) ....................................................................................................................................................... 51
Figure 23 Example, Failure Modes of a Turbine Blade – Tip and Airfoil ( Siemens PG website) ............ 51
Figure 24 Example, Failure Modes of a Turbine Blade- Radial Tip & Platform Cracks (source ccj-online)
.................................................................................................................................................................... 52
Figure 25 Example, Typical Cross Section of a Combustion of LGT (source power-technology.com) .... 52
Figure 26 Example, Typical Parts of Combustion Section of LGT (source ccj-online) ............................. 53
Figure 27 Proposed DFSS Approach .......................................................................................................... 56
Figure 28 Power Generation Repair Development Life Cycle ................................................................... 60
Figure 29 IPDI- Identify Phase Features ..................................................................................................... 62
Figure 30 IPDI- Prioritize Phase Features .................................................................................................. 63
Figure 31 IPDI- Design Phase Features ...................................................................................................... 65
Figure 32 PLA Framework Steps 1-6 ......................................................................................................... 66
Figure 33 PLA Framework Steps 8-9 ......................................................................................................... 67
Figure 34 PLA Key Attributes (source www.ewp.rpi.edu) ........................................................................ 69
Figure 35 IPDI- Implement Phase Features ................................................................................................ 71
Figure 36 Case Study: Introduction ............................................................................................................ 72
vii
Figure 37 Case Study: Compressor Stator Nomenclatures and Zone Definitions ...................................... 73
Figure 38 Case Study: IPDcI Framework ................................................................................................... 76
Figure 39 Case Study: Identify – Statistical Failure Mode Assessment ..................................................... 77
Figure 40 Case Study: Identify- Damage Mode Contribution Example ..................................................... 78
Figure 41 Case Study: Crack Length Vs. EBH Box Plot Example............................................................. 79
Figure 42 Case Study: Design Phase- PLA Step 1 Requirement ................................................................ 81
Figure 43 Case Study: Design Phase- PLA Step 1 Requirement ................................................................ 82
Figure 44 Case Study: Design Phase- PLA Step 2 Requirement ................................................................ 83
Figure 45 Case Study: Design Phase- PLA Step 2 Load PDFs .................................................................. 84
Figure 46 Case Study: Design Phase- PLA Step 3 Requirement ................................................................ 85
Figure 47 Case Study: Design Phase- PLA Step 3 Strength PDF ............................................................... 85
Figure 48 Case Study: Design Phase- PLA Step 4 Requirement ................................................................ 86
Figure 49 Case Study: Design Phase- PLA Step 4 PDFs............................................................................ 87
Figure 50 Case Study: Design Phase- PLA Steps 5 and 6 Requirements ................................................... 88
Figure 51 Case Study: Design Phase- PLA Steps 5 and 6, SF and Dynamic Strain ................................... 88
Figure 52 Case Study: Design Phase- PLA Step 7 Requirement ................................................................ 89
Figure 53 Case Study: Design Phase- PLA Step 7 PDFs............................................................................ 89
Figure 54 Case Study: Design Phase- PLA Step 8 Requirement ................................................................ 90
Figure 55 Case Study: Design Phase- PLA Step 8; 1B, 1T and 2B Modes PDFs ...................................... 91
Figure 56 Case Study: Design Phase- PLA Step 8 Individual PDFs 1T and 2B Modes............................. 92
Figure 57 Case Study: Design Phase- PLA Step 8; Response Surface Plot 2B Mode ............................... 93
Figure 58 Case Study: Design Phase- PLA Step 8; %Probability of Failures ............................................ 93
Figure 59 Case Study: Proof Of Concept- Fixed Beam Example Introduction .......................................... 96
Figure 60 Case Study: Fixed Beam- Steady State Stress and Displacement .............................................. 97
Figure 61 Case Study: Fixed Beam- Free Vibration or Modal Analysis; First Six Modes ......................... 97
Figure 62 Case Study: Fixed Beam- Forced Vibration Analysis Dynamic Stress ...................................... 99
Figure 63 Case Study: Fixed Beam- Closed Form Vs Predicted Dynamic Stress .................................... 100
Figure 64 Case Study: Fixed Beam- Goodman Diagram .......................................................................... 100
viii
LIST OF TABLES
Table 1 Various DFSS Applications ............................................................................................................. 1
Table 2 Traditional DFSS Vs Proposed DFSS ............................................................................................. 3
Table 3 Number of Articles from Various Journals Surveyed .................................................................... 10
Table 4 Comparison Between Relevant Articles and Present Dissertation Findings ................................. 13
Table 5 Failure Modes for Various GT Parts .............................................................................................. 54
Table 6 Six Sigma Tools Used During Various Phases of DFSS Approach .............................................. 61
Table 7 Failure Modes % Distribution Assumption ................................................................................... 74
Table 8 Failure Modes % Distribution Actual Data Assumption ............................................................... 74
Table 9 Design of Experiment Matrix for Crack Lengths Vs No of Operating Hours ............................... 75
Table 10 Summary of Crack Lengths ......................................................................................................... 76
Table 11 Case Study: Prioritize Phase- Business Case Example ................................................................ 80
Table 12 Case Study: Design Phase- Overlapping Probability Failure Rate .............................................. 91
Table 13 Case Study: Design Phase- Proof of Concept, Fixed Beam-Free Vibration Analysis ................. 98
Table 14 Case Study: Design Phase- Fixed Beam-Forced Vibration Analysis .......................................... 99
Table 15 Case Study: Expected Benefits of Proposed PLA Method ........................................................ 102
ix
LIST OF ACRONYMS AND ABBREVIATIONS
ASQ- American society for quality
C&E- Cause and Effect Matrix
CTQ- Critical to Quality
DCOV- Design, characterize, optimize, verify
DFSS- Design for six sigma
DMADV- Define, measure, analyze, design, validate
DMAIC- Define, measure, analyze, improve, control
DOD- Domestic objet damage
EBH- Equivalent baseload hours
FEA- Finite Element Analysis
FMEA- Failure Mode Effective Analysis
FOD- Foreign Object Damage
GT- Gas turbines
HCF- High cycle fatigue
IDOV- Identify, design, optimize, validate
IPDcI or IPDI- Identify, prioritize, comprehensive design, implement
x
IRDDP-Identify, research, design, develop, production
Mpa- Mega pascal unit of stress and pressure
NDE- Non destructive evaluation
OEM- Original equipment manufacturer
PDF- Probability distribution function
PDPD- Plan, design, produce, deliver
PDS- Probability distributions
PG- Power generation
PLA- Probabilistic life assessment
SIPOC- Supplier, Inputs, Processes, Outputs, Customer
TMF- Thermal mechanical fatigue
UTS- Ultimate tensile strength
VoC- Voice of the Customer
YS- Yield Strength
1
CHAPTER 1 INTRODUCTION
Design for Six Sigma (DFSS) is a well-known six sigma approach, which is widely used
for design of new products and services by various industry sectors. Every organization defines
DFSS phases in order to develop a product or service. It is also known that the DFSS approach is
adopted by the organizations to achieve total quality improvement for a new product or service
for each customer requirement. This requires the organization management to understand the
customer needs (sometimes called as business needs) and implement them in the product design
and or service offered by the organization. There are various DFSS approaches that have been
used by the organizations. Table 1 shows a few examples of the DFSS approaches and their
applications to various industries.
Table 1 Various DFSS Applications
DFSS Approach Definition Applications
DMADV Define, Measure, Analyze, Design,
Validate
Aerospace, Automobile, medical
and heavy equipment industries
IDOV Identify, Design, Optimize, Validate Aerospace, Automotive and
Energy sector industries
PDPD Plan, Design, Produce, Deliver Tools and Fixtures Manufacturing
industries
IRDDP Identify, Research, Design, Develop,
Production
Pharmaceutical and Research
organizations
DCOV Define, Characterize, Optimize,
Verify
Shipping and Marine industries
2
Thus there are different variants of DFSS approaches followed by various industries, the
steps within each of those approaches uniquely designed by the respective industry depending
upon the customer requirements and end goals of a product or service.
1.1 Problem Statement
GT used for power generation are subjected to various operating and environmental
conditions during their service. These service conditions could cause GT parts to experience
various failure modes. The GT parts should be restored back to their original geometries during
the repair processes so that they could perform intended function during their next service
interval. Therefore, presently there is a need for a quality methodology or approach that would
integrate geographical customer requirements, development of repair and manufacturing
qualification steps altogether to deliver quality repair products.
Thus, as discussed above, presently the power generation service industry lacks a quality
methodology required for repair development of GT service operated parts. The streamlining of
existing processes used during the repair development cycles would help to form a quality
method using six sigma DFSS approach. In this dissertation, a new Six Sigma DFSS approach is
proposed to address the overhauling of GT parts used for power generation. A unique approach
called “IPDcI”; (Identify, Prioritize, Design (comprehensive design) and Implement) is proposed
in order to address repair development needs. Also, in this dissertation, discussed GT parts used
for power generation parts and their repair needs. In the methodology section a new DFSS
approach is described. Finally a case study is presented to demonstrate the use of the proposed
approach.
3
1.2 Difference Between Traditional DFSS And Proposed DFSS Approach
Table 2 outlines the difference between the most traditionally used DFSS approach
(DMADV) verses proposed DFSS approach called as IPDcI.
The proposed DFSS approach do not require pilot design and validation phase which saves
significant amounts of time and cost for the end customers- the power plant sites.
Table 2 Traditional DFSS Vs Proposed DFSS
Difference between traditionally used DFSS approach and newly developed
IPDcI approach
Traditional New Difference
DMADV IPDcI
Define Identify Traditional- Initiate, scope and plan the project.
New-In this step, geographical region needs are
obtained and studied to define and initiate the
task as opposed to traditional method.
Data on Failure modes of the power generation
GT parts collected up front and categorized the
modes based on their type and root cause.
Measure Prioritize Traditional- Understand customer requirements
and generate specifications.
New- Repair requirements, emergent needs are
determined to create repair development plan
and a preliminary business case is also created
along with potential technical and business
risks.
Analyze Not Applicable Traditional-Develop design concepts based on
measure data findings. Analyze the data to
make generate concepts.
New- Does not exist
Design Comprehensive
design
Traditional-Develop detailed design and
validation plan. Detailed product testing along
with a pilot design released. Very time
consuming and expensive step.
4
Difference between traditionally used DFSS approach and newly developed
IPDcI approach
Traditional New Difference
DMADV IPDcI
New- Repair concepts for GT parts are
brainstormed and developed.
A business case is updated to decide "no" or
"no go" for the program; this step is not
considered in the traditional DFSS approach.
If the business case is acceptable then detailed
design for a selected repair concept is executed
A minimum amount of testing such as dynamic
lab testing and material property testing usually
takes place.
A statistical probabilistic lifing approach using
past engine test results along with new material
test and lab tests data for final repair geometry
is used to predict failure rate for the fleet. This
evaluation acts as validation for the repaired
part.
There is no validation phase or pilot design
requirement, after the design phase; repair
process can be directly implemented for
production.
Validate Implement Traditional- Usually for validation of parts
requires product and or service process testing
and it takes several months or sometimes years
to complete validation of the product or service
process.
New- Few initial sets are repaired. Process
capability and statistical process control
activities are performed in this step.
Repair shop qualifications are performed and
conditional or un conditional release of the
manufacturing process takes place.
Based on initial three sets of repair, a feedback
regarding repair cost models and failure modes
given back to the program office to improve
the repair process in the future if required.
5
1.3 Expected Benefits Of Proposed DFSS Approach
A comprehensive approach which does not require a unit or engine test.
Shorter repair development cycles from identification to implementation.
Increased repair warranty benefits.
Reduction in premature failure (early life) of the repair parts and reduced scrap rates.
Less impact on gas turbine performance and efficiency.
Life extension of repair parts can be achieved.
Increased end customer satisfaction (power plant sites).
Usually no pilot design, the repair design could be directly implemented.
1.4 Contribution To The Proposed DFSS Approach
Proposed a new DFSS approach for power generation repair processes with specific
applications to gas turbines.
Developed a framework or cycle for new DFSS approach.
Discussed and developed detailed steps within the framework or cycle.
Proposed six sigma tools that can be used during proposed DFSS various steps.
Developed statistical probabilistic life assessment (PLA) approach for validation of the
repair concepts.
Completed a case study: Compressor stator vane repair using proposed DFSS approach
Demonstrated use of PLA step by step for the case study.
PLA approach was compared with a classical dynamics problem to determine the validity
of the methodology.
6
CHAPTER 2 SIX SIGMA LITERATURE REVIEW
2.1 Methods For Literature Survey
An extensive six sigma literature survey was performed in order to investigate and explore
the ways and applications of applying six sigma methods and approaches to improve the quality
and performance of the products and services. Below were the objectives of the literature survey,
Explore various six sigma methods and approaches presently being used for different
industries.
Once above step gets completed; check for any six sigma method approach that is
presently available or little modifications can be used for power generation GT repair
processes.
Based on the literature survey, the research gaps related to the topic under consideration were
identified. Below were the methods used for the literature survey,
Surveyed research articles/journal publications/conference proceedings available in the
industrial engineering domain with a main focus upon quality management, system or
technologies.
Surveyed doctoral and master’s thesis information available in the public domain from
the universities with the main focus upon quality management, system or technologies.
Surveyed power generation companies such as GE, Siemens, MHI, Rolls Royce, Alstom
white papers, articles and any information available in the public domain related to the
use of six sigma methods and techniques applied to the GT components.
7
The process of surveying journal articles began with a search for all industrial engineering
journals available in the public domain, then journals were selected which had more focus upon
quality management, system and or technology areas (e.g. ASQ journal of quality management,
quality technology, production and operations management and international journal lean and six
sigma) as shown in the Figure 1.0.
In addition, rankings of the journals were also checked available in the public domain (see
Figure 2.0) so as to emphasize for initial scrutiny of the articles.
Similar to journal article search process conference proceedings were searched for the topic
under consideration.
While surveying for prior doctoral and master’s thesis information related to quality
discipline, initial survey work began with UCF and then expanded to other universities.
Finally, power generation OEM companies white papers, articles available in the public
domain studied which had more emphasis upon quality methods and approaches.
8
Figure 1 Industrial Engineering Journals Surveyed Related to the Topic
9
Figure 2 Impact Factors - Industrial Engineering Journals Surveyed Related to the Topic
10
2.2 Summary Of Literature Survey
Overall 97 articles from various journals related to the quality and power generation
industry were surveyed to gather information on the six sigma methodologies and applications to
different industry sectors. Table 3 shows the journals and number of articles related to various
fields that discussed various six sigma methods or approaches.
Table 3 Number of Articles from Various Journals Surveyed
Journals\
Fields of
Interests
Au
tom
oti
ve
Rep
air
Pow
er G
en
New
bu
ild
Pow
er G
en
Ser
vic
e
Ele
ctro
nic
s
Aer
osp
ace
Hosp
itali
ty
Con
stru
ctio
n
Hea
lth
care
Fin
an
ce
Oth
er
Man
ufa
ctu
rin
g
Oth
er
Gen
eral
International
Journal of
Lean Six
Sigma
1 1 3 3 3
Asian
Journal on
Quality
2 1 2 4
International
Journal of
Quality &
Reliability
Management
2 1 1 2 2 1 1 4
International
Journal of
Quality
Science
1 1
Journal of
Quality in
Maintenance
Engineering
7 3 2 1 2
The TQM
Journal
1 1 2
The TQM
Magazine
1 1 9
International
Journal of
1 2 2 1 2 8
11
Journals\
Fields of
Interests
Au
tom
oti
ve
Rep
air
Po
wer
Gen
New
bu
ild
Po
wer
Gen
Ser
vic
e
Ele
ctro
nic
s
Aer
osp
ace
Ho
spit
ali
ty
Co
nst
ruct
ion
Hea
lth
care
Fin
an
ce
Oth
er
Ma
nu
fact
uri
ng
Oth
er
Gen
era
l
Six Sigma
and
Competitive
Advantage
International
Journal of
Productivity
and Quality
Management
1 1
International
Journal of
Quality
Engineering
and
Technology
International
Journal of
Quality and
Innovation
Journal
Aircraft
Engineering
and
Aerospace
Tech
2
Journal of
European
Society for
Engineering
Education
1
Quality
Management
Journal
1 5
American
Council for
an energy
efficient
economy
1 1
ASME
Proceedings
1
12
Journals\
Fields of
Interests
Au
tom
oti
ve
Rep
air
Po
wer
Gen
New
bu
ild
Po
wer
Gen
Ser
vic
e
Ele
ctro
nic
s
Aer
osp
ace
Ho
spit
ali
ty
Co
nst
ruct
ion
Hea
lth
care
Fin
an
ce
Oth
er
Ma
nu
fact
uri
ng
Oth
er
Gen
era
l
Combustion,
Fuels and
Emissions
Journal of
scientific and
industrial
research
1
Total 2 4 8 6 6 2 3 3 4 8 14 37
97
After reviewing various articles from several journals, a comparison among the most
relevant papers were created and shown in Table 4. The major difference appeared to be
customer agreement before product manufacturing and use of statistical analysis to justify the
product modifications. In the present dissertation, the proposed DFSS approach requires getting
the customer agreement before proceeding to the repair development process and included
probabilistic statistical method to justify the repair to predict the failure rate of the component
under consideration.
13
Table 4 Comparison Between Relevant Articles and Present Dissertation Findings
Authors\
Subjects
Article Journal/
Magazine
Yea
r
Ind
ustry
Six
Sigma
Method
Va
lidatio
n
Use of
Quality
Tools
(QFD,
VoC,
SPC)
Customer
Agreeme
nt before
Process/
Product
Manufact
ure
Proba
bilistic
Statisti
cal
Metho
ds
Min Chul
Lee,
Kwang
Ick Ahn
and
Youngbi
n Yoon
Developm
ent of Gas
Turbine
Combusti
on Tuning
Technolo
gy Using
Six Sigma
Tools
ASME
Proceeding
s
Combustio
n, Fuels
and
Emissions
, June 11–
15, 2012,
SBN: 978-
0-7918-
4468-7
2012 Power
Gen
New
Build
DFSS Y Y N N
John
Kane, Jr.,
E. I.
duPont
de
Nemours
& Co.
Using Six
Sigma to
Drive
Energy
Efficiency
Improvem
ents at
DuPont
American
Council
for an
energy
efficient
economy,
SS03_Pan
el2_Paper0
7, 2003
proceeding
s
2003 Power
Gen
New
Build
DMAIC Y Y N N
Kaushik
P,
Khanduj
a D
DM make
up water
reduction
in thermal
power
plants
using six
sigma
DMAIC
methodol
ogy
Journal of
scientific
and
industrial
research ,
vol 67, Jan
2008,
pp36-42
2008 Power
Gen
New
Build
DMAIC Y Y N N
14
Authors\
Subjects
Article Journal/
Magazine
Yea
r
Ind
ustry
Six
Sigma
Method
Va
lidatio
n
Use of
Quality
Tools
(QFD,
VoC,
SPC)
Customer
Agreeme
nt before
Process/
Product
Manufact
ure
Proba
bilistic
Statisti
cal
Metho
ds
Debasis
Das
Adhikary
,
Goutam
Kumar
Bose ,
Dipankar
Bose ,
Souren
Mitra
Multi
criteria
FMECA
for coal-
fired
thermal
power
plants
using
COPRAS-
G
Internation
al Journal
of Quality
&
Reliability
Manageme
nt, Vol. 31
Iss: 5,
pp.601 -
614
2014 Power
Gen
New
Build
DMAIC Y Y N N
Tore
Markeset
, Jorge
Moreno‐Trejo,
Rajesh
Kumar
Maintena
nce of
subsea
petroleum
productio
n systems:
a case
study
Journal of
Quality in
Maintenan
ce
Engineerin
g, Vol. 19
Iss: 2,
pp.128 -
143
2013 Power
Gen
Servic
e
DMAIC Y Y N N
D.
Papachri
stos, V.
Tsoukala
s, J.
Vlachogi
annis
Total
quality
plan
applied in
the
Hellenic
power
productio
n process
The TQM
Magazine,
Vol. 16
Iss: 2,
pp.136 -
144
2004 Power
Gen
Servic
e
DMAIC Y Y N N
E.V.
Gijo,
Ashok
Sarkar
Applicatio
n of Six
Sigma to
improve
the quality
of the
road for
wind
turbine
installatio
The TQM
Journal,
Vol. 25
Iss: 3,
pp.244 -
258
2013 Power
Gen
Servic
e
DMAIC Y Y N N
15
Authors\
Subjects
Article Journal/
Magazine
Yea
r
Ind
ustry
Six
Sigma
Method
Va
lidatio
n
Use of
Quality
Tools
(QFD,
VoC,
SPC)
Customer
Agreeme
nt before
Process/
Product
Manufact
ure
Proba
bilistic
Statisti
cal
Metho
ds
n
Pawan
Jaglan;
Dinesh
Khanduj
a;
Parbhaka
r
Kaushik
Capacity
waste at
thermal
power
plants in
India: a
Six Sigma
perception
Int. J. of
Six Sigma
and
Competitiv
e
Advantage
, 2013
Vol.8,
No.1,
pp.22 - 33
2013 Power
Gen
Servic
e
DMAIC Y Y N N
S.M.
Saeed
"The
Repair
and
Overhaul
of Gas
Turbines:
Some
Notes on
the
Organizati
on and
Facilities
Used in
the
Maintena
nce of Jet
Engines",
Aircraft
Engineerin
g and
Aerospace
Technolog
y, Vol. 25
Iss: 7,
pp.200 -
201
2006 Aeros
pace
DMAIC
/DFSS
Y Y N N
Oguzhan
Yilmaz,
Dominic
Noble,
Nabil
N.Z.
Gindy,
Jian Gao
A study of
turbomac
hinery
componen
ts
machining
and
repairing
methodol
ogies",
Aircraft
Engineerin
g and
Aerospace
Technolog
y, Vol. 77
Iss: 6,
pp.455 -
466
2005 Aeros
pace
DMAIC
/DFSS
Y Y N N
16
Authors\
Subjects
Article Journal/
Magazine
Yea
r
Ind
ustry
Six
Sigma
Method
Va
lidatio
n
Use of
Quality
Tools
(QFD,
VoC,
SPC)
Customer
Agreeme
nt before
Process/
Product
Manufact
ure
Proba
bilistic
Statisti
cal
Metho
ds
Aircraft
Engineeri
ng and
Aerospace
Technolo
gy
Johnson
M,
Dubikov
sky S
Incorporat
ing Lean
Six Sigma
into an
Aviation
Technolo
gy
Program
Journal of
European
Society for
Engineerin
g
Education
-SEFI,
2015
2015 Aeros
pace
DMAIC
/DFSS
Y Y N N
Premarat
ne
Samaran
ayake,
Senevi
Kiridena,
Aircraft
maintenan
ce
planning
and
schedulin
g: an
integrated
framewor
k
Journal of
Quality in
Maintenan
ce
Engineerin
g, Vol. 18
Iss: 4,
pp.432 -
453
2012 Aeros
pace
DMAIC Y Y N N
Ernest
Emeka
Izogo ,
Ike-
Elechi
Ogba
Service
quality,
customer
satisfactio
n and
loyalty in
automobil
e repair
services
sector
Internation
al Journal
of Quality
&
Reliability
Manageme
nt, Vol. 32
Iss: 3,
pp.250 -
269
2015 Auto
motiv
e
DMAIC N Y N N
Min
Zhang ,
Yueyue
Xie ,
Lili
Huang ,
Service
quality
evaluation
of car
rental
industry
Internation
al Journal
of Quality
&
Reliability
Manageme
2013 Auto
motiv
e
DMAIC N Y N N
17
Authors\
Subjects
Article Journal/
Magazine
Yea
r
Ind
ustry
Six
Sigma
Method
Va
lidatio
n
Use of
Quality
Tools
(QFD,
VoC,
SPC)
Customer
Agreeme
nt before
Process/
Product
Manufact
ure
Proba
bilistic
Statisti
cal
Metho
ds
Zhen He in China nt, Vol. 31
Iss: 1,
pp.82 -
102
Hakeem
Ur
Rehman,
Muham
mad
Asif,
Muham
mad
Aamir
Saeed,
Muham
mad
Asim
Akbar,
Muham
mad
Usman
Awan,
Applicatio
n of Six
Sigma at
cell site
constructi
on: a case
study
Asian
Journal on
Quality,
Vol. 13
Iss: 3,
pp.212 -
233
2012 Const
ructio
n
DMAIC N Y N N
Clement
L.W.
Wong,
Albert
H.C.
Tsang,
T.S.
Chung
A
methodol
ogy for
availabilit
y
assessmen
t of tunnel
designs
Internation
al Journal
of Quality
&
Reliability
Manageme
nt, Vol. 23
Iss: 1,
pp.60 - 80
2006 Const
ructio
n
DFSS N Y N N
Salman
T. Al‐Mishari,
Saad
Suliman
"Integrati
ng Six‐Sigma
with other
reliability
improvem
ent
methods
in
Journal of
Quality in
Maintenan
ce
Engineerin
g, Vol. 14
Iss: 1,
pp.59 - 70
2008 Other
Manu
factur
ing
DMAIC Y Y N N
18
Authors\
Subjects
Article Journal/
Magazine
Yea
r
Ind
ustry
Six
Sigma
Method
Va
lidatio
n
Use of
Quality
Tools
(QFD,
VoC,
SPC)
Customer
Agreeme
nt before
Process/
Product
Manufact
ure
Proba
bilistic
Statisti
cal
Metho
ds
equipment
reliability
and
maintenan
ce
applicatio
ns",
Journal of
Quality in
Maintena
nce
Engineeri
ng
Werner
Timans ,
Kees
Ahaus ,
Jiju
Antony
Six Sigma
methods
applied in
an
injection
moulding
company
Internation
al Journal
of Lean
Six Sigma,
Vol. 5 Iss:
2, pp.149 -
167
2014 Other
Manu
factur
ing
DMAIC Y Y N N
E.V. Gijo
,
Shreeran
ga Bhat ,
N.A.
Jnanesh
Applicatio
n of Six
Sigma
methodol
ogy in a
small-
scale
foundry
industry
Internation
al Journal
of Lean
Six Sigma,
Vol. 5 Iss:
2, pp.193 -
211
2014 Other
Manu
factur
ing
DMAIC Y Y N N
Ploytip
Jirasukpr
asert ,
Jose
Arturo
Garza-
Reyes ,
Vikas
Kumar ,
Ming K.
Lim
A Six
Sigma and
DMAIC
applicatio
n for the
reduction
of defects
in a
rubber
gloves
manufactu
ring
process
Internation
al Journal
of Lean
Six Sigma,
Vol. 5 Iss:
1, pp.2 -
21
2014 Other
Manu
factur
ing
DMAIC Y Y N N
19
Authors\
Subjects
Article Journal/
Magazine
Yea
r
Ind
ustry
Six
Sigma
Method
Va
lidatio
n
Use of
Quality
Tools
(QFD,
VoC,
SPC)
Customer
Agreeme
nt before
Process/
Product
Manufact
ure
Proba
bilistic
Statisti
cal
Metho
ds
Sukhwin
der Singh
Jolly ,
Bikram
Jit Singh
An
approach
to
enhance
availabilit
y of
repairable
systems: a
case study
of SPMs
Internation
al Journal
of Quality
&
Reliability
Manageme
nt, Vol. 31
Iss: 9,
pp.1031 -
1051
2014 Other
Manu
factur
ing
DMAIC Y Y N N
Sushil
Kumar,
P.S.
Satsangi,
D.R.
Prajapati
Improvem
ent of
Sigma
level of a
foundry: a
case study
The TQM
Journal,
Vol. 25
Iss: 1,
pp.29 - 43
2013 Other
Manu
factur
ing
DMAIC Y Y N N
Francesc
o
Aggogeri
, Marco
Mazzola,
James
O'Kane
Implemen
ting DFSS
to
increase
the
performan
ce level of
an
extrusion
process
Int. J. of
Six Sigma
and
Competitiv
e
Advantage
, 2009
Vol.5,
No.1,
pp.10 - 28
2009 Other
Manu
factur
ing
DFSS Y Y N N
Kioumar
s
Paryani,
Ali
Masoudi,
and
Elizabeth
Cudney
QFD
Applicatio
n in the
Hospitalit
y
Industry:
A Hotel
Case
Study
Quality
Manageme
nt Journal
- January
2010 , vol
17, #1
2011 Hospi
tality
DMAIC N Y N N
Robert E.
Carter,
Subhash
C.
Lonial,
and P. S.
Raju
Impact of
Quality
Managem
ent
on
Hospital
Performan
Quality
Manageme
nt Journal
-Oct 2010
, vol 17,
#4
Healt
hcare
DMAIC N Y N N
20
Authors\
Subjects
Article Journal/
Magazine
Yea
r
Ind
ustry
Six
Sigma
Method
Va
lidatio
n
Use of
Quality
Tools
(QFD,
VoC,
SPC)
Customer
Agreeme
nt before
Process/
Product
Manufact
ure
Proba
bilistic
Statisti
cal
Metho
ds
ce:
An
Empirical
Examinati
on
Rahul
Panat* ,
Valentin
a
Dimitrov
a , Tamil
Selvy
Selvamu
niandy ,
Kazuhik
o Ishiko ,
Dennis
Sun
The
applicatio
n of Lean
Six Sigma
to the
configurat
ion
control in
Intel’s
manufactu
ring R&D
environm
ent
Internation
al Journal
of Lean
Six Sigma,
Vol. 5 Iss:
4, pp.444 -
459
2014 Electr
onics
DMAIC Y Y N N
George
Besseris
Multi-
factorial
Lean Six
Sigma
product
optimizati
on for
quality,
leanness
and
safety: A
case study
in food
product
improvem
ent
Internation
al Journal
of Lean
Six Sigma,
Vol. 5 Iss:
3, pp.253 -
278
2014 Hospi
tality
DMAIC N Y N N
21
Authors\
Subjects
Article Journal/
Magazine
Yea
r
Ind
ustry
Six
Sigma
Method
Va
lidatio
n
Use of
Quality
Tools
(QFD,
VoC,
SPC)
Customer
Agreeme
nt before
Process/
Product
Manufact
ure
Proba
bilistic
Statisti
cal
Metho
ds
Chuen‐Sheng
Cheng,
Chi‐Ming
Kuan,
Research
on
product
reliability
improvem
ent by
using
DMAIC
process: A
case study
of cold
cathode
fluorescen
t lamp
Asian
Journal on
Quality,
Vol. 13
Iss: 1,
pp.67 - 76
2012 Electr
onics
DMAIC Y Y N N
Ayon
Chakrab
orty,
Michael
Leyer
Developin
g a Six
Sigma
framewor
k:
perspectiv
es from
financial
service
companie
s
Internation
al Journal
of Quality
&
Reliability
Manageme
nt, Vol. 30
Iss: 3,
pp.256 -
279
2013 Finan
ace
DMAIC N Y N N
Jiju
Antony,
Anmol
Singh
Bhuller,
Maneesh
Kumar,
Kepa
Mendibil
, Douglas
C.
Montgo
mery
Applicatio
n of Six
Sigma
DMAIC
methodol
ogy in a
transactio
nal
environm
ent
Internation
al Journal
of Quality
&
Reliability
Manageme
nt, Vol. 29
Iss: 1,
pp.31 - 53
2012 Finan
ace
DMAIC N Y N N
22
Authors\
Subjects
Article Journal/
Magazine
Yea
r
Ind
ustry
Six
Sigma
Method
Va
lidatio
n
Use of
Quality
Tools
(QFD,
VoC,
SPC)
Customer
Agreeme
nt before
Process/
Product
Manufact
ure
Proba
bilistic
Statisti
cal
Metho
ds
Alexandr
os G.
Psychogi
os, Jane
Atanasov
ski,
Loukas
K.
Tsironis
Lean Six
Sigma in
a service
context: A
multi‐factor
applicatio
n
approach
in the
telecomm
unications
industry
Internation
al Journal
of Quality
&
Reliability
Manageme
nt, Vol. 29
Iss: 1,
pp.122 -
139
2012 Electr
onics
DMAIC Y Y N N
Sabina
Potra;
Adrian
Pugna
DFSS in
marketing
:
designing
an
innovative
value co-
creation
campaign
Int. J. of
Six Sigma
and
Competitiv
e
Advantage
, 2015
Vol.9,
No.1,
pp.21 - 36
2015 Finan
ace
DFSS N Y N N
Shreeran
ga Bhat;
N.A.
Jnanesh
Enhancin
g
performan
ce of the
health
informatio
n
departmen
t of a
hospital
using lean
Six Sigma
methodol
ogy
Int. J. of
Six Sigma
and
Competitiv
e
Advantage
, 2013
Vol.8,
No.1,
pp.34 - 50
2013 Healt
hcare
DMAIC N Y N N
P.
Malliga,
S.P.
Srinivasa
n
The stock
service
improvem
ent by the
deployme
Int. J. of
Six Sigma
and
Competitiv
e
2007 Finan
ace
DMAIC N Y N N
23
Authors\
Subjects
Article Journal/
Magazine
Yea
r
Ind
ustry
Six
Sigma
Method
Va
lidatio
n
Use of
Quality
Tools
(QFD,
VoC,
SPC)
Customer
Agreeme
nt before
Process/
Product
Manufact
ure
Proba
bilistic
Statisti
cal
Metho
ds
nt of Six
Sigma
Advantage
, 2007
Vol.3,
No.2,
pp.103 -
119
Jaap Van
Den
Heuvel,
Ronald
J.M.M.
Does,
Henk De
Koning
Lean Six
Sigma in
a hospital
Int. J. of
Six Sigma
and
Competitiv
e
Advantage
, 2006
Vol.2,
No.4,
pp.377 -
388
2006 Healt
hcare
DMAIC N Y N N
Bikram
Jit Singh
, Yash
Bakshi
Optimizin
g backup
power
systems
through
Six
Sigma:
An Indian
case study
of diesel
genset
Internation
al Journal
of Lean
Six Sigma,
Vol. 5 Iss:
2, pp.168 -
192
2014 Other DFSS Y Y N N
Amir
Saeed
Noorami
n, Vahid
Reza
Ahouei,
Jafar
Sayareh
A Six
Sigma
framewor
k for
marine
container
terminals
Internation
al Journal
of Lean
Six Sigma,
Vol. 2 Iss:
3, pp.241 -
253
2011 Other DMAIC Y Y N N
24
Authors\
Subjects
Article Journal/
Magazine
Yea
r
Ind
ustry
Six
Sigma
Method
Va
lidatio
n
Use of
Quality
Tools
(QFD,
VoC,
SPC)
Customer
Agreeme
nt before
Process/
Product
Manufact
ure
Proba
bilistic
Statisti
cal
Metho
ds
Nicholas
Roth,
Matthew
Franchett
i
Process
improvem
ent for
printing
operations
through
the
DMAIC
Lean Six
Sigma
approach:
A case
study
from
Northwest
Ohio,
USA
Internation
al Journal
of Lean
Six Sigma,
Vol. 1 Iss:
2, pp.119 -
133
2010 Other DMAIC Y Y N N
Ching‐Kun Lin,
Hsien‐Ching
Chen,
Rong‐Kwei Li,
Ching‐Piao
Chen,
Chih‐Hung
Tsai
Research
on
Increasing
the
Productio
n Yield
Rate by
Six Sigma
Method:
A Case of
SMT
Process of
Main
Board
Asian
Journal on
Quality,
Vol. 10
Iss: 1, pp.1
- 23
2009 Other DMAIC Y Y N N
Jacquelin
e
Blackmo
re, Alex
Douglas
Towards a
“better”
University
: the Use
of the
EFQM
Model in
a UK
Higher
Education
Institution
Asian
Journal on
Quality,
Vol. 4 Iss:
2, pp.1 -
15
2003 Other DMAIC Y Y N N
25
Authors\
Subjects
Article Journal/
Magazine
Yea
r
Ind
ustry
Six
Sigma
Method
Va
lidatio
n
Use of
Quality
Tools
(QFD,
VoC,
SPC)
Customer
Agreeme
nt before
Process/
Product
Manufact
ure
Proba
bilistic
Statisti
cal
Metho
ds
Debapra
yag
Chaudhu
ri, Arup
Ranjan
Mukhopa
dhyay,
Sadhan
Kumar
Ghosh
Assessme
nt of
engineerin
g colleges
through
applicatio
n of the
Six Sigma
metrics in
a State of
India
Internation
al Journal
of Quality
&
Reliability
Manageme
nt, Vol. 28
Iss: 9,
pp.969 -
1001
2011 Other DMAIC N Y N N
Rupa
Mahanti,
Jiju
Antony,
Six Sigma
in the
Indian
software
industry:
some
observatio
ns and
results
from a
pilot
survey
The TQM
Journal,
Vol. 21
Iss: 6,
pp.549 -
564
2009 Other DMAIC N Y N N
Lawrenc
e O.
Jenicke,
Anil
Kumar,
Monica
C.
Holmes
A
framewor
k for
applying
six sigma
improvem
ent
methodol
ogy in an
academic
environm
ent
The TQM
Journal,
Vol. 20
Iss: 5,
pp.453 -
462
2008 Other DMAIC N Y N N
Andrius
Kubilius;
Keith
Winfrey;
Carrie
Mayer;
Gregory
Applying
Lean Six
Sigma
tools to
reduce the
rate of
slips, trips
Int. J. of
Six Sigma
and
Competitiv
e
Advantage
, 2015
2015 Other DMAIC N Y N N
26
Authors\
Subjects
Article Journal/
Magazine
Yea
r
Ind
ustry
Six
Sigma
Method
Va
lidatio
n
Use of
Quality
Tools
(QFD,
VoC,
SPC)
Customer
Agreeme
nt before
Process/
Product
Manufact
ure
Proba
bilistic
Statisti
cal
Metho
ds
Johnson;
Teena
Wilson
and falls
for Joint
Commissi
on field
staff
Vol.9,
No.1,
pp.37 - 55
Muthusw
amy
Shanmug
araja; M.
Nataraj;
Nallasam
y
Gunasek
aran
Total
performan
ce
excellence
- a model
to
implemen
t Six
Sigma in
service
organisati
ons
Int. J. of
Six Sigma
and
Competitiv
e
Advantage
, 2012
Vol.7,
No.2/3/4,
pp.209 -
242
2012 Other DMAIC Y Y N N
Ming-
Chang
Lee, To
Chang
Developin
g a lean
design for
Six Sigma
through
supply
chain
methodol
ogy
Int. J. of
Productivit
y and
Quality
Manageme
nt, 2010
Vol.6,
No.4,
pp.407 -
434
2010 Other DMAIC N Y N N
Ghunaki
kar S
A new
Six Sigma
DFSS
approach
for
overhauli
ng of
service
run
power
generatio
n parts
PhD
Thesis
2015 Powe
r Gen
Servi
ce
DFSS N Y Y Y
27
2.3 Key Finding For Literature Survey
As mentioned in section 2.2, overall 97 relevant journal articles related to six sigma
methods and approaches with practical applications to various fields were studied available in
the public domain.
The following paragraphs discuss the key findings of the six sigma literature survey.
I] Journal Articles/Conference Proceedings literature survey:
Arumugam V., Antony J., and Linderman K., 2014, systematically reviewed and
synthesized the academic research on six sigma and business performance. The article discussed
various six sigma approaches required for different business models used by the industries such
as aerospace, automotive, healthcare. The main finding from this article was the use of
traditional six sigma approaches and how they benefit the businesses.
Watson G., DeYong C., 2010, in their article described the historical approach to
concurrent engineering (CE) which has been used by product line management by various
organizations. The paper pointed out the need for rigorous examination of logical six sigma
models using DMAIC/DFSS methods that are proposed for guiding the practitioners while using
quality methods both for products engineering and business systems. The authors suggested
almost unique DFSS models for various industries.
Snee R, 2010, assessed lean six sigma methods and approaches to identify important
advances over the last 10 to 15 years and discussed emerging trends that suggested how the six
sigma methodologies need to evolve. The author found that organizations have many different
improvement needs that require objectives and methods contained in the lean and six sigma
28
methodologies (DMAIC/DFSS). The author’s findings suggested that improvement was most
effective when approached in an holistic manner addressing improvement in all parts of the
organization using an holistic improvement methodology which required a customized lean six
sigma DFSS approach.
Pulakanam V., 2012, mentioned that in the early 1990s several empirical studies have
been undertaken linking quality efforts to organizational performance such as total quality
management (TQM) and six sigma. The author claimed that the implementation of six sigma
delivered an average savings of 1.7 percent of revenues over the implementation period and
returned more than $2 in savings per dollar invested on Six Sigma. The author has shown the
benefits of use of DMAIC and DFSS six sigma methods to the organizations. However there is
no specific mention of the power generation industry.
Al‐Mishari S., Suliman S., 2008, reported in their paper weaknesses with existing
equipment reliability improvement methods through their integration into the six‐sigma DMAIC
methodology. The value of their paper lies in the introduction of six‐sigma into equipment
reliability/maintenance applications which was quite original since this methodology has
traditionally been limited to manufacturing. The outcome of this paper was of significant value
as it opens up a new perspective into the development of reliability improvement measures for
plant equipment.
Reosekar R., Pohekar S., 2014, authors’ study involved an analysis of 179 research
articles published from 1995 to 2011 in 52 selected reputable journals. The selected articles were
classified by time distribution of articles, research methodology, research stream, authorship
patterns, sector-wise focus of articles, integration with other manufacturing philosophies, and
29
implementation status and performance measurement of the models. Again, there was no
specific mention of the power generation industry.
Min L., Ahn K., Yoon Y., 2012, used DIDOV (Define, Identify, Design, Optimize and
Verify) DFSS six sigma approach to design power generation gas turbine combustion parts. In
the Define phase, the NOx reduction target was set, then the status of the power plant was
diagnosed in the identify phase. During the design phase important control parameters to meet
the defined target were determined based on the analysis of the correlation between the control
parameters and NOx emissions. For the optimize phase, the optimum condition was determined
using one of the six sigma tools. During the final phase the optimum condition parameters were
verified by applying the condition to the gas turbine combustion tests. This new six sigma DFSS
approach resulted in reduction of NOx emissions by 70% and the standard deviation was
improved by 60%. This was a very good example of the application of the DFSS approach for
new build products of power generation gas turbine parts.
Dubikovsky J, 2015, has presented incorporation of six sigma methods into the aviation
gas turbine engines which benefited aircraft engine manufactures’ in several ways with design,
manufacturing, purchasing etc.
Kaushik P. and Khanduja D., 2008, have used the six sigma method in a thermal power
plant for energy conservation during DM (de mineralize) water process. They found that 0.1%
increase in DM water consumption increases power generation cost by Indian Rs. 82.82 lakh
annually. They also mentioned that the use of six sigma recommendations brought down the
requirements for makeup water from 0.9% to 0.54% MCR accruing it with comprehensive
energy savings of Indian Rs. 304.77 lakh annually.
30
Kane J., 2003, while working at DuPont applied six sigma problem methodologies to a
variety of business, technical, transactional, and process problems across the organization. He
applied six sigma methods and tools to analyze energy conversion processes such as steam
boiler, turbine generator, central refrigeration, compressed air, and HVAC systems. The author
mentioned that DuPont was able to gain and maintain remarkable energy savings using six sigma
methods. Author also applied six sigma methods to energy utilization processes such as the
heating and refining during manufacturing process. It appears that the six sigma energy project
savings of over $250,000 annually for an individual project. The author has also discussed the
six sigma methodology and presented case studies of several energy efficiency projects those
were succeeded.
Fouquet J., 2007, mentioned in the article the use of Lean Product Development (LPD)
and Design for Six Sigma (DFSS) approaches to satisfy customer expectations for small
industries. The author’s study was focused on identifying various dissimilarities and similarities
between six sigma methodologies and the correlation between them. This comparison showed by
the author is of high importance to six sigma practitioners while deciding on a strategy for their
product development. LPD and DFSS methodologies helped to reduce waste and development
time and increased the quality of a product in the beginning of the product development.
Ngapuli I., Sinisuka, H. and Nugraha, 2013, presented an approach to predict the life
cycle cost (LCC) on the power generation operating engines The article was mainly related to
estimating the total cost of ownership which included project or asset acquisition cost,
operational and maintenance cost, disposal cost. The interesting point was noticed that LCC cost
estimation included deterministic costs (such as acquisition costs, improvement costs and
31
disposal costs) and probabilistic (such as failure cost, spare, repair, downtime cost and gross
margin) cost. This was the first article that depicted usage of six sigma tools that could be
applicable to the present dissertation discussed in this report.
Premaratne S. and Senevi K., 2012, presented an approach using the six sigma frame
work for aircraft maintenance and repair planning and scheduling. Although the presented
framework was mainly applicable to the aerospace industry, however some of the steps could be
used by other industry sectors including power generation.
Andrew T., Barton R., Byard P., 2008, have discussed an integrated six sigma
maintenance (SSM) model for the manufacturing industry mainly for the UK manufacturing
sector. The proposed six sigma model combines contemporary business management techniques
along with TQM strategies and provides repair and maintenance managers and engineers a
strategic framework to increase efficiency and output.
Yamashina H., Mizuyama H., Byard P., 1998, in their article presented how the number
of short stops could be reduced by placing inspection and repair stations strategically for an
automated assembly line. They used the DMAIC methodology to achieve the change. Authors
developed an algorithm to optimize the steps during the improve phase to solve the assembly
sequence issue. Again the methodology developed to optimize the sequence was specific to
automate an assembly system.
Bañuelas R., Antony J., 2004, have presented six sigma frameworks for software design
processes. This was a unique six sigma methodology applicable to software designing. They
32
have clarified some of the myths of their methodology and suggested practical challenges
However their approach is not suitable for the power generation industry.
Gijo E., Sarkar A., 2013, in their article they presented the DMAIC methodology for the
development of wind farm roads, manufacturing, installation and service of windmills. Even
though this particular paper was related to power generation service industry however, no new
six sigma approach was developed.
Aggogeri F., Mazzola M., O'Kane J., 2009, in their article presented a DFSS
methodology to improve business performance and customer satisfaction at a small
manufacturing company. They designed an extrusion process by applying project management,
along with DFSS steps such as brainstorming and statistical tools (ANOVA, DOE). An
important finding to notice from this article was the use of various quality tools that could be
used for DFSS steps.
Shahin A., 2008, in the article presented a survey for Design for Six Sigma (DFSS)
methods used by world class companies. The author’s key finding was the companies’ use of a
unique DFSS approach suitable to their application, in some cases creating a new DFSS
approach. This finding confirms that power generation repair industry requires a unique DFSS
approach to suit the business needs.
Booysen C., Heyns P., Hindley M., Scheepers R., 2015, presented a probabilistic lifing
approach as a six sigma tool that can be used in fatigue life prediction for a steam turbine blade
subjected to dynamic loadings. They used finite element analysis models of the blade to generate
the transient response of the blades. The authors also mentioned use of random curve data fitting
33
for material properties using finite element analysis results. They also used confidence intervals
in statistical models to account for uncertainty in the material properties. However, this approach
can only be applied to new products, as in the paper there was no mention about but repair
processes applied to the steam turbine parts.
Zhao J., 1995, presented a generic probabilistic life evaluation model that can be used as
a six sigma tool for part life estimations. The author specifically developed the approach for
rotorcrafts; however the framework can also be used in the power generation field. There are
lots of similarities between power generation and aircraft engine parts even though the
functionalities are different. The six sigma tool presented in the paper predicts the probabilistic
failure rate taking into considerations from design requirements, failure criteria, material
properties, analysis and test conditions using statistical probabilistic. It was assumed that all the
data used followed normal distributions. However, the particular framework presented in the
paper was very generic and mainly focused upon new build parts.
II] Doctoral and master’s theses survey:
Baral M. ULBS, 2014, aimed the research to investigate interactive phenomena of
knowledge management (KM) concepts with the six sigma methods, and how KM concepts
including updated elements could be integrated in a structured, systematic and effective way with
the six sigma framework for project deployment. The research aimed for application of KM
concepts within the Six Sigma projects deployed to the textile manufacturing process. Author
proposed a structured integrated conceptual model; namely DMAIC- KM model that could be
used for textile manufacturing. Again this was a custom six sigma method for textile
manufacturing.
34
Chakrabarthy A., 2009, in the doctoral research work presented results from two aspects.
The first concern was with the estimation of success and progress of service organizations due to
six sigma implementation. The author accomplished this activity by conducting a large-scale
survey of service organizations situated in different geographic locations. The author mentioned
that the results obtained by analyzing the responses indicated that mainly mass services have
implemented six sigma throughout the organization and they were the most successful and
progressive. The use of tools and techniques was also different among successful and less
successful organizations. Successful service organizations use a smaller tools and techniques
compared to less successful organizations. Author presented 18 cases studies in his theses which
encompass various service industries using traditional DFSS and DMAIC methods.
III] Power Generation Industries white papers/articles/information pertaining to applications of
six sigma methods and approaches while designing their products and processes:
An article from GE on gas turbine repair technology “GER-3957B” (reference 22)
described some of the repair processes and technologies currently used by GE’s Global
Inspection and Repair Services Operation used for power generation GT parts. This article
describes the repair processes such as coating, welding, brazing, drilling and destructive analysis.
Also, described in this article an approach to assess the repaired parts integrity. However, major
missing factor from the article was the failure mode identification and prediction of failure rate
after repair. In such a case six sigma DFSS approach should certainly help in order to enhance
quality of the overhauled parts and increase customer satisfaction.
Another article from GE described a DFSS approach which was DMADOV (define,
measure, analyze, design, optimize and verify) used specifically during the design of new parts
35
(reference 23). The author of the article also showed a frame work for the six sigma process
along with tools used in each step. It appeared that GE has customized the DFSS approach for
designing new build parts for power generation purposes. The article does not describe any six
sigma approach for repair process or products.
An article from Motorola (reference 24), described evolution of DFSS six sigma
approach and the author mentioned that Motorola began to quantitatively benchmark its
performance against companies that held a much stronger competitive position before using
traditional DFSS approach. The author said that after implementation of six sigma methods,
Motorola was changed and made the company very competitive with competitors. This article
presents evolution of six sigma DMAIC method and DFSS approach at Motorola.
An article from Honeywell’s Dr. D. Houston (reference 25) on a case study using
traditional DMAIC approach on simulation of software system usage, based on data elicited from
regular system users, provided a sound, quantitative basis for estimating productivity increases
and justifying upgrade investments. Honeywell is a small engine manufacturer for power
generation sector. The case study presented was for software used during new build product
design. The missing factor from the article was measurement of actual savings (business case)
and use of a control plan for monitoring variation in improved steps of the process (lacked
implementation).
2.4 Literature Survey Conclusion
Above discussions showed various six sigma methods and approaches with practical
applications to different industries in the public domain. However, there was insufficient
36
evidence found for the six sigma approaches and applications to power generation repair
processes (parts being overhauled at an intended interval). There was a need to develop a unique
six sigma approach for power generation repair products and processes, since the currently
available six sigma DFSS approaches are not adequate to design and develop repairs for the parts
used in the power generation GT industries. The major reasons for inadequacy for application to
power generation repair products and processes are; different business models involving
customers across the globe, project requirements driven by failure modes; emergent needs and
long term strategy plans required in order to address various failure modes. Usually a pilot
design is required for the product and process development using the present DFSS approach
whereas GT repair customers cannot wait for the pilot design/process to be proven. In addition a
validation (engine testing) is required for the present DFSS approach which can take several
months and with huge expense; whereas the GT repair customers cannot wait for such a long
time period and do not wish to pay for the testing. Thus there was a need to develop a quick and
robust validation approach which is a missing factor for present DFSS approach.
These GT parts even though originally designed for a certain number of cycles and hours
they still go through repair at certain time before they retire or go out of functional requirements
as described in the introduction section. Such type of business model where in which original
design expects that there will be some kind of repair required for the parts at a certain point of
time in the product life cycle requires a new DFSS approach. A new six sigma DFSS approach
should eliminate the above mentioned deficiencies from the present DFSS approach.
37
CHAPTER 3 POWER GENERATION PARTS AND FAILURE MODES
3.1 Introduction To Power Generation GT Parts
A power plant is a facility for electricity generation. It can be steam turbine or gas turbine
driven or sometimes a combination of both steam and gas turbines in cases of a combined cycle
power plant. Figure 3 shows a combined cycle power plant facility which includes several parts
in addition to steam and gas turbines such as a generator, cooling facilities, compressor,
transformers, condensers and heat recovery systems. During operation, air is taken at the gas
turbine inlet and fed through the GT unit to produce power, exhaust gas heat is utilized to heat
the water through a heat recovery steam generator and fed to the steam turbine to produce power
(electricity) which is collected at a transformer. In this dissertation, the major focus was upon the
GT part rather than steam turbine or any other power plant part.
Figure 3 A Typical Combined Cycle Power Plant Layout
38
3.2 GT Parts Familiarization And Repair Products
Power generation gas turbine (heavy duty application) mainly constitutes three modules;
compressor, combustion and turbine sections. Each section has primary parts such as blades,
stator vanes, rotors, seals, casings, in case of compressor and turbine sections. The combustion
section includes baskets, transitions, support housing and fuel nozzles as the main parts. The
Figure 4 shows a typical large gas turbine three dimensional cut section to illustrate the
complexity of the power engine.
Figure 4 Typical Large Gas Turbine 3D Section (source Siemens PG website)
Figure 5 depicts a large gas turbine assembly layout example along with various parts
that are typically overhauled at a certain point of time during the life cycle of a gas turbine.
39
Figure 5 Example Large Gas Turbine- Assembly (source Siemens PG website)
As discussed above, below are the major parts of a GT unit and discussions about these
parts along with their failure modes are presented in the subsequent sections
Compressor: Casings, Diaphragms/Stators, Blades, Seal holders, rotors
Turbine: Casings, Blades, Vanes, Ring segments, Inter stage seal housing, rotors
Combustor: Fuel nozzles, Baskets, Support housing
Typically, the GT parts are designed by an OEM for a specific number of hours and
cycles (also known as starts) before their performance degradation. In addition, they also have an
intended repair interval depending upon the type of the part. For example, typically combustion
parts can have 8,000 or 12,000 hours of operation per repair interval with maximum of 36,000 to
40
48,000 hours of service operation. Thus combustion parts can be serviced by 3 to 4 times before
they complete their intended total design life. Turbine parts usually have 48,000 to 72,000 hours
for total design life with repair intervals at every 24,000 hours and compressor parts are designed
for 96,000 hours of operation with the repair interval at every 48,000 hours. The above
mentioned design life and component repair intervals can change from one OEM to another
OEM, however primary GT components shall require repairs at some point of time during their
life cycle.
Therefore, any overhauled GT part is called a repair product and it goes under a repair
process to be functionally fit to re-operate in a GT unit.
3.3 GT Parts And Failure Modes
In this section, three major sections or modules of a GT unit sub component and their
distress modes called as failure or damage modes during service operating conditions are
described.
3.3.1 Compressor Section
Figure 6, depicts a typical cross section of a compressor GT unit; mainly it consists of
casings, rotors, stators, blades and seal holder as sub components. The purpose of the compressor
section is to deliver compressed air to the combustion section to burn fuel. Also, sometime cold
compressor air from front stages of the compressor are used for cooling purposes in the turbine
section.
41
Figure 6 Example Compressor Cross Section LGT (source gasturbinetutorial.blogspot.com-604)
Figure 7 shows a stator or diaphragm sub component of the compressor section. The
stator plays a crucial role during service; it guides the airflow through the compressor to improve
efficiency of the unit. As air passes through the stator vanes, it smoothens out the flow to
increase the pressure ratio in subsequent rotor blades and stator vanes.
42
Figure 7 Example Compressor Diaphragms or Stators LGT (source ccj-online)
Typical failure modes experienced by stators or diaphragms are shown in the Figure 8
during service operation and they can be FOD, wear, erosion and cracks due to the operating
conditions while compressing air. Usually, these failure modes are caused by high cycle fatigue
loading or impact from other objects or sometimes sand particles from the air may erode the part
surfaces.
43
Figure 8 Example of Failure Modes of Compressor Diaphragms or Stators (source ccj-online)
Figure 9 shows the seal holder subcomponent of the compressor section. The function of
the seal holder is to seal the airflow between stator and rotor parts so that there is no airflow
leakage and thus minimize performance loss to the GT unit.
44
Figure 9 Example of Compressor Seal Holder (source Siemens PG website)
The failure mode experienced by the seal holders is mainly wear as shown in the Figure
10 and can be found in the hooks where they get interfaced with the stator parts.
Figure 10 Example, Failure Modes of Compressor Seal Holders (source Siemens PG website)
Figure 11 shows a casing subcomponent of the compressor section. Casings are outside
covers which are thick in the cross section to protect the inner parts of the unit.
45
Figure 11 Example of a Compressor Casing (source oddstuffmagazine.com)
Typically casings exhibit pitting erosion and wear failure modes as shown in the Figure
12 during operating conditions.
Figure 12 Example, Failure Modes of a Compressor Casing (source ccj-online)
46
3.3.2 Turbine Section
Figure 13 below depicts a typical cross section of the turbine module of a GT unit;
mainly it consists of casings, rotor, stators, blades and inter stage seal housing as sub
components. The functionality of the turbine section parts is to expand the air so as to drive the
turbine and compressor and at the same time the turbine shaft gets attached to generator to
produce power.
Figure 13 Example Typical Cross Section of Turbine LGT (source etspower.com)
Figure 14 shows typical failure modes of turbine casing which are usually wear, erosion,
cracks and FOD. The turbine casing protects inner assembly parts similar to compressor casing
in a GT unit.
47
Figure 14 Example, Failure Modes of Turbine Casing (source Siemens PG website)
Figure 15 depicts the inter stage seal housing subcomponent of a turbine module. The
functionality of this sub component is similar to that of a seal holder to seal the air between rotor
and stator parts to minimize performance loss.
Figure 15 Example, Typical Inter Stage Seal Housing (source Siemens PG website)
48
Typically, inter stage seal housing experience wear, erosion and seal damage as failure
modes and are shown in the Figure 16.
Figure 16 Example, Failure Modes of Inter Stage Seal Housing (source Siemens PG website)
Figure 17 and 18 depict the stator vanes (single and double airfoils) sub component of the
turbine model. The function of the vanes is to guide the hot gas in the unit and improve the
efficiency of the turbine section.
Figure 17 Example of a Double Airfoil Turbine Vane (source liberdi.com)
49
Figure 18 Example of Single Airfoil Turbine Vane (source liberdi.com)
The failure modes of vanes observed to be linear cracks that can be classified as narrow
(crack width <0.15 mm) or wide (crack width >0.15 mm) gap cracks; erosion as shown in the
Figure 19.
Figure 19 Example, Failure Modes of a Turbine Vane (sourc ccj-online)
50
Turbine blade features with and without shrouds are shown in the Figure 20 and 21
respectively. The turbine blade plays a crucial role in expanding the hot gases while producing
power.
Figure 20 Example of Unshrouded Turbine Blade (source ge.com)
Figure 21 Example of Shrouded Turbine Blade (source gasturbinepower.ASME)
51
Typical failure modes for a turbine blade are radial tip crack, tip wear, platform cracks
and angel wing cracks. The typical failure modes of a turbine blade are shown in the Figures 22,
23 and 24.
Figure 22 Example, Failure Modes of a Turbine Blade- Platform, Angel Wing (source Siemens PG website)
Figure 23 Example, Failure Modes of a Turbine Blade – Tip and Airfoil ( Siemens PG website)
52
Figure 24 Example, Failure Modes of a Turbine Blade- Radial Tip & Platform Cracks (source ccj-online)
3.3.3 Combustion Section
Figure 25 shows a cross section of a typical combustion parts mainly consist of support
housing, pilot nozzles, baskets and transitions.
Figure 25 Example, Typical Cross Section of a Combustion of LGT (source power-technology.com)
53
Figure 26 shows some details of the parts fuel nozzles, basket and support housing. The
functionality of combustion parts is to support burning of fuel and contain as well as transfer hot
expanding air to the turbine section.
Figure 26 Example, Typical Parts of Combustion Section of LGT (source ccj-online)
Table 5 shows the summary of failure modes for all sections of a GT unit with a focus on
main components. GT parts typically experience these failure modes during their service
operation and require them to be overhauled during repair intervals so that they can become
functional for the next service interval.
54
Table 5 Failure Modes for Various GT Parts
GT Section Part Failure Modes
Compressor Diaphragms Airfoil- FOD, linear cracks, erosion and corrosion
Shrouds- Wear, corrosion, cracks
Seal Holders Wear
Casings Pitting erosion, small linear cracks, wear
Turbine Casings Wear, erosion, cracks, out of roundness
Interstage seal
housing
Wear, erosion, cracks, out of roundness
Vane Airfoil- FOD, craze cracks, zipper cracks, narrow and wide gap
cracks, erosion, TBC spallation
Platform- Erosion, narrow and wide gap cracks, TBC spallation
Blade Tip cracking and wear, platform erosion and cracking, TBC
spallation
Combustion Baskets Cracks in the frame and liner, bent thermocouple tubes burning
of the exit end, wear, misalignments, FOD
Fuel Nozzles Cracks, wear, oxidation,
Support Housing Cracks, wear, oxidation,
55
3.4 GT Parts Overhaul And Repair Importance And Role Of Six Sigma
During service operation GT parts experience various types of damages which can be
broadly categorizes into two; continuous operation and cyclic damages. The continuous
operation damages typically consist of oxidation, erosion, wear, FOD (foreign object damage) or
DOD (domestic object damage), HCF (high cycle fatigue) and TMF (thermal mechanical
fatigue). The cyclic damages may involve TMF (thermal mechanical fatigue), wear and erosion.
These damages can result in cracks or loss of material which depends upon several parameters
such as geographical location (Asia/ North America/Middle East or Europe etc.) of the power
generation unit, an operational anomaly. The repairable limits such as damage
(crack/FOD/DOD/erosion/oxidation) length, depth and surface area can be decided based upon
repair technology and capability of the repair facility and effect of repair on part’s integrity.
Designing of the repair limits are one of most challenging factors during overhauling. Because
any excessive repair on the parts can trigger immediate or accelerated complete failures of the
parts which may cause significant amount of financial loss to the power plant owners and gas
turbine manufacturers. Therefore, power plants and GT manufacturers negotiate on warranties
which are often very expensive. Thus overhaul and repair of the heavy duty gas turbine parts
play a very crucial role for the power generation industry.
It is known that six sigma methods can be used during the design of GT parts and there
are many journal articles available that describe the systematic use of six sigma processes such
as DMAIC as well as DFSS. However, it is required to have a six sigma approach that will work
for repair process development of GT parts to ensure the quality, lead time reduction and lower
the cost of the repaired parts.
56
CHAPTER 4 METHODOLOGY
4.1 Outline Of The Methodology
The proposed DFSS approach mainly consists of four phases; identify, prioritize, design
(comprehensive design) and implement. Figure 27 shows the skeleton of proposed DFSS
approach.
Figure 27 Proposed DFSS Approach
Mainly the DFSS approach was derived based upon operational experience across several
continents and customers’ demands. GT operations are always driven by geographical region
needs. That means power plant sites located in the Middle East, Europe, North America, South
America and Asia have different overhaul and repair requirements for the GT parts because of
types and severity of failure modes can be different from one site to another site. It is very
57
important to recognize such needs before repair development planning takes place. Once repair
needs are identified, prioritization of these requirements is required in order to determine short
term (also called as emergent) and long term needs. Based on categories of repair requirements
the next step is to determine the repair development plan and business case creation which
involves technical and commercial risks so as to continue on the program. Once repair
development trials are completed then in the final phase repair facilities should be qualified for
future orders.
4.2 Methodology Description
The six sigma approach used here was abbreviated as “IPDcI” of “IPDI” (Dc stands for
comprehensive design) as described in the section 4.1. Figure 28 shows the detailed repair
process development life cycle. It should be noticed that there was no validation phase included
since the repair process development of the GT components should be inexpensive as compared
to the new build part cost and development time so that the part would be put back into the
engine for operation. Therefore a comprehensive design phase is developed which includes
preliminary design, business case, final optimized design and verification using finite element
simulations, probabilistic life calculations (sometimes simple mechanical testing to validate the
FE models (no expensive rig or actual gas turbine unit test). The major advantage of the
comprehensive design phase is the ability to predict the failure rate using past GT test experience
and scaling those results using simulations and six sigma statistical tools (design of experiment).
The first phase, ‘Identify’ begins with customer location This matters a lot as geographical
region repair needs of the customer could be different because of extent and nature of the failure
modes on the parts. Thus during the identify phase, customers and their repair needs should be
58
determined using classical six sigma tools such as VoC (voice of the customer), SIPOC (basic
process map) for the components. The customer needs could be collected from the marketing
team, field service team and regional repair network facilities. Once all customer requirements
have been sorted out then the next step of the ‘IPDcI’ approach was to ‘Prioritize’. During this
phase emergent needs (field issues, customer repair lead time) and specific repair requirements
from various power plant sites are analyzed. This is done using the classical six sigma tool such
as a CTQ tree, Kano model and C&E (cause and effect) matrix. This phase determines the repair
product strategy for the customers (gas turbine power plant sites). Once repair product strategy is
set up then, it is required to translate the emergent needs into a priority list. This is done using
the QFD six sigma tool. The deliverable of the prioritize phase is usually a repair development
plan and preliminary business case (repair vs replacement cost, risks). The repair development
plan includes the repair trials (or technology) completion schedule, budget (cost to achieve repair
development of the part under study), manufacturing qualifications. As soon as the repair
development plan is prepared, the next important step of the proposed ‘IPDI’ cycle is the
comprehensive design (Dc) phase. The comprehensive design (Dc) is initiated with the
preliminary design in which repair concepts are brainstormed and evaluated. During
brainstorming sessions, various experts (design, materials, manufacturing, field service, budget
owners, and repair development engineers) should participate to come up with the concepts.
Once the repair concept is finalized then the business case can be updated with the proposed
repair cost. This is a ‘Go/No Go’ toll gate to decide if the repair program is feasible to move
forward or stop at this point. Again QFD is a very useful six sigma tool to make decisions during
this process. If the updated business case review gets passed then the final design phase begins.
During the final design phase FE simulations are done to evaluate mechanical integrity of the
59
repair concept. A design of experiment can be performed to optimize the repair geometry. Also,
in many cases it is required to perform small lab testing to validate FE models. This depends
upon complexity and technical risks associated with the part operation during service. In addition
sometimes a material testing is done to explore material properties such as tensile and fatigue
tests of the repaired sample specimens. Once FE simulation, lab and material test data are
collected then finally part failure rate is predicted using prior GT test data. This is usually done
via scaling the results for repair parts using probabilistic life calculations. The failure or scrap
rate for the fleet is one of the deliverables from the design phase. In addition, repair trials are
performed to check if the proposed repair process is feasible in terms of time and quality
requirements. The final phase of the “IPDcI” approach is the ‘Implement’ step in which a
customer set is identified for repair and procurement of necessary tooling and fixtures for the
repairs. The first requirement of the implement phase is the limited or conditional manufacturing
qualification in which process capability analysis is carried out to verify manufacturing
expectations. The final requirement of the ‘Implement’ phase is unconditional manufacturing
qualification in which the repair facility is allowed to continue parts’ repairs to remove limited
qualification status. However, during unconditional qualification, it is required to conduct a
capability analysis periodically to ensure the quality of the repair products and process. Figure 28
shows the power generation repair product development life cycle phases and time lines. This
product development life cycle is an average life cycle based on experience of overhauling
several power generation GT parts.
60
Figure 28 Power Generation Repair Development Life Cycle
4.3 Six Sigma Tools Used During Various Phases of New DFSS Approach
Table 6 shows the various tools that could be used during different phases of the
proposed DFSS steps.
61
Table 6 Six Sigma Tools Used During Various Phases of DFSS Approach
Identify- During the first step, as discussed in the above section (4.2), regional customer
needs are identified via voice of the customer sometimes called as voice of the repair
market segment. In addition an internal repair facility needs and field operational issues
are also gathered as part of the VoC process. A SIPOC tool is used after VoC typically to
arrange a process map for the present repair process for a given GT section or module.
Figure 29 depicts the key features of the identify phase. As discussed above, collection of
geographical repair needs and goals is the objective of the identify phase. There are four
62
categories of inputs in this phase; repair shop requirements which includes failure or
damage mode data collection during inspections that are not identified during
geographical requirements, determine repair vs replace strategy; business needs which
include cost reduction and new technology developments; geographical fleet
requirements include collection of failure or damage modes and performance issues; new
product requirements include redesigns and modifications in the fleet.
Figure 29 IPDI- Identify Phase Features
Prioritize- In this step overhaul and repair needs are prioritized using CTQ (critical to
quality) factors. Primarily CTQ factors are driven by part life, repair cost, GT
performance. Kano model and C&E (cause and effect) matrix can be used to arrange the
prioritized customer needs in a certain fashion so that they can be ranked depending upon
short or long term goals. This decides repair product or process strategy for the GT parts
under consideration. Once repair product strategy gets decided then the next step is to
prepare a repair development plan and preliminary business case. The repair development
63
plan includes various toll gates for quality checks and these toll gates are repair product
design and qualification process reviews. Process map, project plan, QFD (quality
function deployment) tools are most suitable for this purpose. Figure 30 shows the
prioritize phase key features. There are two outputs of this phase; repair development
plan and preliminary business case. For the repair development plan, there are four inputs
required such as geographical repair needs (this is the output of the identify phase),
emergent needs (forced outages), repair facility resource availability and manufacturing
requirements (mainly tooling, new repair methods). The preliminary business case
requires replacement cost, potential risks (technical and financial) and estimated repair
cost as inputs.
Figure 30 IPDI- Prioritize Phase Features
Comprehensive Design- This phase includes three steps. It begins with preliminary
design in which team members along with various discipline experts develop repair
concepts, tooling requirements, potential risks and also determine preliminary costs for
64
the concepts. Pugh’s matrix and TRIZ methods can be used for the preliminary design
and finalizing the concept. Once the final repair concept is selected the next step is to
update the business case for the cost, lead times and repair forecasts. This is really a ‘GO’
or ‘No GO’ gate in order to release the repair development program, if the business case
fails then most likely the program gets closed or requires re-evaluation of the financial
calculations or objectives of the program. Mainly the QFD tool can be used for this step.
If the business case gets approved then the next step is to pursue a detailed design. The
repair concept gets analyzed via finite element analysis and sometimes CFD
(computational fluid dynamics) modeling to estimate the part life after repair and impact
on GT performance is evaluated for the parts under consideration. Various DOE (design
of experiment) tools and capability analysis can be used as six sigma tools for this
purpose. Since there is no validation phase for the proposed DFSS approach, it is required
to estimate failure rates for the proposed repair approach. Many times it is done via
probabilistic life assessment which primarily uses existing material testing data, prior GT
test results. Typically, GT OEM companies keep the records for engine testing for all the
parts and this data along with analytical calculations can be used to predict the failure rate
for the given part under consideration. The probabilistic lifing approach saves significant
amount of time and cost as it serves the purpose of validation. Figure 31 shows the key
features of the design phase. There are three deliverables proposed from this phase which
are final repair design, risk assessment and final business case. A repair design is
finalized based on repair concept selection and repair development trials supported by
repair development plan (output of prioritize phase). It is important that the final repair
design should be validated in some cases. The validation can be done by probabilistic life
65
assessment (PLA).
Figure 31 IPDI- Design Phase Features
PLA as mentioned above uses prior testing experience on the same product and
FEA (finite element analysis) or closed form analysis calculations. The Proposed
framework for PLA is shown in the Figures 32 and 33. It must be noted that the proposed
PLA framework is only applicable to high cycle fatigue (HCF) failure mode. It has
overall eight steps and each step is discussed in the below paragraphs in detail.
Step 1: Design Requirements- In the first step prior design experience related to the part
under consideration is collected. This involves collection of data such as failure criteria,
service operating conditions, frequencies or vibratory modes of interest. It is important
that prior design models shall be studied so as to develop knowledge about the part
design under consideration.
66
Step 2: In the second step, prior power generation GT unit test data shall be collected and
studied. This includes collection of strain gage instrumentation test data and any other
data such as air flow pressure changes, temperatures during testing. In this dissertation
the major focus is upon the use of strain gage data collection and its usage for various
modes that can be used further to predict repaired part responses. The probability
distributions (PDS) shall be plotted against various modes of vibration for steady state
baseload and start or shut down transient operating conditions.
Figure 32 PLA Framework Steps 1-6
67
Figure 33 PLA Framework Steps 8-9
Step 3: Material properties of the part under consideration shall be collected which
includes elastic properties such as dynamic young’s modulus, poison’s ratio; physical
properties such as density, coefficient of thermal expansion and tensile properties such as
0.2% yield strength (YS) , ultimate tensile strength UTS. Similar to step 2, probability
distribution curves can be generated with +/- 3 sigma standard deviation for tensile
properties.
Step 4: FEA/Lab Test Data- During repair development, it is extremely difficult to
conduct another power generation GT test for a repaired part due to the cost and time.
However, for many cases it is possible to develop finite element models or some type of
closed form models to assess the repair concepts from the mechanical integrity
perspective. In some instances, laboratory testing can be done in case of more complex
repair to validate FEA models. In this step, it is proposed to develop FEA or analytical
model for the baseline as well as repair geometries. Once the FEA model is created then
dynamic strain values can be found at the repair locations and strain gage locations for
68
the baseline model. The strain gage locations can be obtained from the baseline unit
testing (step 2).
Step 5: Scale Factors- In this step scale factors are generated between repair location
dynamic strain and prior GT test strain gage location dynamic strain using the FEA
model. The FEA model is usually created for free vibration or modal analysis so as to
collect the various frequencies of vibration. The scale factors can be used to predict
service operating dynamic strains at the repair locations as it is very difficult to develop
FEA models for forced vibration or dynamic conditions during operation.
Step 6: In this step, dynamic strain distribution functions at repair locations are developed
using scale factors from step 5 for critical modes of vibration obtained from step 1.
Step 7: In this step data from step 3 and 6 is used to perform boolean (subtraction)
operations for various vibratory modes or frequencies of vibration. Probability
distribution functions for strength (material properties, step 3) and loading (step 6) for
each critical mode of vibration data is used to perform probability subtractions. PDF can
be converted to standard normal probability distributions if required due to different
scales for strength and loading curves. The subtractions of PDFs produce predictions for
probabilities of failures for each mode of vibration.
Step 8: Combined probabilities of failures- In this step all probabilities of failures are
collected for each mode of vibration and summed up to estimate combined probability of
failure for a given operating condition such as steady state or start/shutdown transients.
Once combined probabilities of failures are estimated then confidence intervals such as
69
50%, 90% or 95% can be plotted to see the variations in the failure rates. Finally these
failure rates are compared against the allowable failure rate which can be set by risk
assessment and failure cost to the power generation GT unit.
Again, the above steps and PLA is applicable to only the HCF failure mode. For
other failure modes similar to PLA can be developed and those are not covered in this
dissertation. The purpose of this PLA is to showcase an example for a failure mode. A
case study for compressor stator using this PLA is discussed in chapter 5. Figure 34
depicts the key features of the PLA framework. These key features are strength and
loading probability density functions for transient and steady state operating conditions.
Both probability distributions are required in order to predict the failure rate for the
repaired part.
Figure 34 PLA Key Attributes (source www.ewp.rpi.edu)
70
Implement- In the final phase, once the repair concept is fully developed and product
definition which includes drawings, CAD models and repair instructions are released; a
conditional qualification is performed. During limited or conditional qualification few
sets are identified for the repair of the part under consideration and capability analysis is
performed on dimensional inspections. If the data from the conditional qualification
appear to meet the manufacturing qualification requirements then the repair facility gets
un-conditionally approved to perform repairs on the future parts with statistical process
control requirements. Figure 35 shows the key features of the implement phase. This
phase is focused upon manufacturing qualifications and repair production. The
manufacturing qualifications involve final repair design drawings, generating repair
instructions and development of critical to quality dimensions from the repair
perspective. During actual production, classical statistical process control can be used
along with six sigma tools discussed in Table 6. In this dissertation, implement steps are
not discussed in detail as these are very specific to a given part. Only a general structure
is proposed for the discussion purposes to showcase key features or points to be
considered during the implement phase for GT part repair process.
71
Figure 35 IPDI- Implement Phase Features
72
CHAPTER 5 CASE STUDY: GT AXIAL COMPRESSOR
DIAPHGRAM/STATOR REPAIR
5.1 Case Study- Introduction
The case study demonstrates the use of the IPDI approach discussed in chapter 4.0. The
data used in this case study was found from information available in the public domain
(internet search). The particular case study discussed was focused upon service operated GT
compressor stator or diaphragms repair. One of the major focuses in the case study was the
use of a PLA (probabilistic lifing analysis) approach to justify the repair design so that the
stator could be directly used for field operation. Also, the case study demonstrates the one
type of loading condition in the PLA framework which is HCF (high cycle fatigue). Figure
36 shows details of the design, functionality and typical failure modes for GT compressor
stators.
Figure 36 Case Study: Introduction
73
Typically, GT compressor stators are used for 100,000 hours of operation with the major
inspections at 50,000 hours of operation. This is very common for large gas heavy duty gas
turbine compressors. It was assumed in the case study that the compressor stators under
study had gone through first service intervals and would get repaired at the repair facility.
For evaluation purposes two different GT units are considered; one from North American
region and another one from Middle East area. The major reason behind considering different
GT units from two different geographical regions was to show the difference between the
failure or damage modes due to different environmental conditions such as sand storms
(Middle East region) or snow storms (North American region).
Figure 37 Case Study: Compressor Stator Nomenclatures and Zone Definitions
A typical GT compressor stator has three geometry features airfoil, inner shroud and
outer shroud. The structure is subjected to various airflow fluctuations during service
operations that can cause cracking to the airfoil due to dynamic loads. Figure 37 shows the
nomenclature of the compressor stator parts and for damage evaluation purposes the stator
74
airfoil was divided into various zones such as A, B, D and E. The zones distributions (20%,
60% and 20%) were based upon typical compressor airfoil failure modes intensities observed
during service operation. Table 7 shows the failure mode matrix assumptions and Table 8
depicts actual damage mode data used for Middle East and North American regions.
Table 7 Failure Modes % Distribution Assumption
Table 8 Failure Modes % Distribution Actual Data Assumption
No of Damages LE Zone A TE Zone B AF OD Zone D AF ID Zone E
FOD/DOD 60% 25% 10% 5%
Cracks 25% 20% 50% 5%
Erosion pits 60% 25% 10% 5%
No of Damages LE Zone A TE Zone B AF OD Zone D AF ID Zone E
FOD/DOD 75% 10% 10% 5%
Cracks 25% 20% 50% 5%
Erosion pits 70% 15% 10% 5%
Failure Modes Matrix % Distribution- North American Site
Failure Modes % Distribution Matrix- Middle East
No of Damages LE Zone A TE Zone B AF OD Zone D AF ID Zone E No of damages
FOD/DOD 60 25 10 5 100
Cracks 5 4 10 1 20
Erosion pits 6 3 1 1 10
No of Damages LE Zone A TE Zone B AF OD Zone D AF ID Zone E
FOD/DOD 38 5 5 3 50
Cracks 10 8 20 2 40
Erosion pits 70 15 10 5 100
Failure Modes Matrix Actual Data- Middle East
Failure Modes Matrix Actual Data- North American Site
75
The design of experiment matrix (only for North American site) used for various EBH
(equivalent baseload hours) is shown in the Table 9 and Table 10 depicts the summary of
crack length data used in the case study.
Table 9 Design of Experiment Matrix for Crack Lengths Vs No of Operating Hours
M=1, STD= 1 M=2, STD=1 M= 3, STD= 1 M= 4, STD= 1
EBH 24000 34000 40000 50000
Set # 1 2 3 4
1 2 2 4
4 3 2 8
1 2 5 3
1 2 4 4
1 1 3 3
1 1 3 5
2 2 4 9
2 4 3 4
1 1 2 3
1 3 3 4
2 3 1 5
1 1 5 4
2 2 3 4
1 3 2 3
2 3 5 10
1 2 4 5
2 1 3 4
1 3 2 4
2 1 5 3
1 3 1 3
1 1 3 4
2 3 4 9
1 3 5 5
1 2 3 4
2 3 3 5
1 2 3 3
1 3 2 2
1 2 3 3
1 2 2 9
1 1 1 6
1 3 2 4
1 6 4 4
1 3 4 3
3 2 4 2
1 7 3 4
1 2 6 3
1 2 4 9
1 4 5 4
1 3 1 9
2 1 5 4
2 4 4 5
2 2 5 3
1 2 1 6
1 2 2 6
1 2 3 4
1 1 3 3
2 2 4 3
1 1 3 4
1 3 3 4
1 3 3 6
76
Table 10 Summary of Crack Lengths
5.2 Case Study - IPDcI Framework
As discussed in the methodology section of chapter 4.0, the IPDcI (or IPDI) approach
was used to evaluate the case study. Using the proposed IPDI approach, the case study
framework was developed and is shown in the Figure 38.
Figure 38 Case Study: IPDcI Framework
Set 1 Set 2 Set 3 Set 4
Minimum 0.50 0.63 0.76 2.12
25th Percentile 0.50 1.66 2.14 3.36
Median 1.01 2.25 3.14 4.01
75th Percentile 1.50 2.81 3.73 5.13
Maximum 4.36 7.00 5.52 10.00
Summary of Crack Length
77
During the identify phase failure mode data is collected. The prioritize phase is focused
upon repair development plan and the preliminary business case development;
comprehensive design phase involves repair concept development, verification, risk
assessment and finalizing the business case; Implement phase is focused upon repair
qualifications using statistical process control. The details of each of these steps are discussed
in this chapter. There is a large emphasis on probabilistic life assessment of repair geometry
and the case study discusses the step by step proposal for failure rate evaluation.
5.3 Identify Phase
The identify phase used the data from section 5.2 for failure mode assessment such as
FOD/DOD, cracks and erosion pits. Figure 39 shows the failure mode assessment for two
different sites.
Figure 39 Case Study: Identify – Statistical Failure Mode Assessment
78
It is evident that the airfoil LE zone A has a maximum number of damages for both site
parts. However, the key difference was the type of a failure mode contributing to the
damages; for the North American site FOD/DOD type of damage was predominant whereas
Middle East site parts received erosion pits as significant failure mode.
Figure 40 shows the damage modes contribution for both site parts for various failure
modes. It is evident that in both cases FOD/DOD and erosion pits were a significant
contributor to the overall damages to the airfoil.
Figure 40 Case Study: Identify- Damage Mode Contribution Example
A box plot was created using the data from Table 9 and 10 discussed in section 5.2 for
the North American site only. The crack length range was in the range 1 mm to 10 mm.
Three different crack lengths were chosen for further evaluation to check the limiting crack
79
length beyond which part could not be repaired. The different crack lengths chosen were 3
mm, 6mm and 9 mm.
Figure 41 Case Study: Crack Length Vs. EBH Box Plot Example
5.4 Prioritize Phase
As discussed in Figure 30, deliverables of the prioritize phase are repair development
plan and preliminary business case. The repair development plan includes region and
emergent needs, resource availability and manufacturing requirements. It also includes
proposal for repair trials and schedule, initial repair concepts, critical path, lead times and
materials requirements.
Table 7 shows the business case example that can be used in the Prioritize phase. It
includes development costs and repair project return over the period of several years and
payback period of investment. In addition, it also includes potential technical and commercial
80
risks to the overall repair scope. The details of this phase are not discussed in this dissertation
since risks may vary from part to part and it can be difficult to discuss all of those details in
the dissertation. However, one can find lots of information in the public domain regarding
risk assessment and business case creation.
Table 11 Case Study: Prioritize Phase- Business Case Example
5.5 Comprehensive Design Phase
The design phase general roadmap is shown in Figure 31 and it has three deliverables;
validated repair design, final business case and risk assessment. In this dissertation, the
evaluation of repair concepts and validation approach has been discussed and business case
and risk assessment details has not been discussed in detail as it was discussed in chapter
4.0. During the identify phase three different repair geometries were identified for the
evaluation purposes which were 3 mm, 6 mm and 9 mm crack lengths. For the case study
purposes, the three crack geometries assumed to be at the airfoil leading edge and at the
81
center. Therefore, three different repair models were analyzed from the repair acceptability
perspective. The purpose of the analysis was to decide the crack length beyond which the
repaired part could not be accepted and should be scrapped.
Figure 42 Case Study: Design Phase- PLA Step 1 Requirement
As stated earlier, this section (design) mainly focused upon probabilistic evaluations
since it is important to discuss the repair verification approach and the PLA supports the
verification approach. Figures 32 and 33 show the PLA steps discussed earlier during
methodology section of the chapter 4. In this section the step by step evaluation process is
described for 9 mm crack length and the results for other two geometries (3 mm and 6 mm)
have been shown for comparison purposes.
82
Step 1: In the first step, stator airfoil design requirements needed to be found. For this
example, three modes of vibration 1B (bending), 1T (torsion) and 2B (bending) were
assumed to be critical during unit start or shutdown transient and baseload or steady state
operating conditions. Again the loading condition for this PLA framework was assumed to
be high cycle fatigue. Figure 43 shows the step 1 requirement which is discussed above
regarding collection of critical information related to the design of the part and loading
conditions.
Figure 43 Case Study: Design Phase- PLA Step 1 Requirement
Step 2: In the second step, GT unit test data needed to be collected such as frequency vs
dynamic strain as shown in the Figure 44. In this example test data was created in MS- Excel
assuming strain range between 50-300 micro strains. The strain range assumption was based
83
upon historical GT testing experience. Fifty data points were created for each mode of
vibration (1B, 1T and 2B) and for every operating condition. A normal distribution was
assumed.
Figure 44 Case Study: Design Phase- PLA Step 2 Requirement
The data created using a normal distribution assumption is shown in the Figure 45. It can
be noticed that there are a total of six PDFs are created for three modes and two operating
conditions (start transient and steady state). These data points represent airfoil responses
during operation at a location on the airfoil that could capture 1B, 1T and 2B modes of
vibration. Repair location was assumed to be at the center of the airfoil on the leading edge.
The strain levels at the repair location are usually unknown in practice. In this case, it was
assumed that the repair location strains were not known and to be found via the proposed
PLA approach.
84
Figure 45 Case Study: Design Phase- PLA Step 2 Load PDFs
Step 3: Material property data was collected in this step. For the evaluation purposes of
this example, it was assumed that the airfoil material to be 304 SS (stainless steel). As shown
in Figure 46, endurance strength (fatigue strength of the material) of the airfoil was 241 Mpa
and the source of this information is shown in Figure 46. The weld debited endurance
strength of the airfoil was assumed to be 181 Mpa (25% weld debit) in order to analyze the
repair location. The 25% weld debit factor was assumed based upon type of weld and post
weld inception method. In this case, it was assumed that after repair X-ray NDE being used
for inspections. A standard deviation of 5% was assumed on the endurance strength of the
material and PDF curves for 241 Mpa and 181 Mpa with 5% standard deviation and +/- 3
sigma and normal distribution assumption are shown in the Figure 47.
85
Figure 46 Case Study: Design Phase- PLA Step 3 Requirement
Figure 47 Case Study: Design Phase- PLA Step 3 Strength PDF
86
Figure 48 Case Study: Design Phase- PLA Step 4 Requirement
Step 4: In this step, FEA dynamic strains data was used from free vibration analysis so as
to create scale factors for vibratory stress predictions. Figure 48 shows the FEA and lab
testing set up that can be used to validate FE model as an example. Figure 49 shows the free
vibration dynamic strains (also known as modal strains) for three modes of vibration (1B, 1T
and 2B) and two different operating conditions; transient and steady state created during this
step. Again the strain range was assumed to be 50 to 300 micro strains while generating the
data. Normal distribution was assumed while creating data points (50 data points for each
mode).
87
Figure 49 Case Study: Design Phase- PLA Step 4 PDFs
Step 5: Scale factors were created between GT test reference strain gage on the airfoil
and repair location using modal analysis results. Figure 50 and 51 show the scale factors
(step 5) for three modes of vibration at two different operating conditions.
Step 6: In this step, forced response dynamic strains were predicted at repair locations
using the scale factors generated step 5. Figures 50 and 51 show the mean forced response
dynamic strains values for two operating conditions and three different modes of vibration as
discussed in step 5. Forced response dynamic strain indicates actual strain values that can be
seen during service operation. In addition, vibratory stresses were also predicted in this
example using forced response dynamic strains obtained from scaling results.
88
Figure 50 Case Study: Design Phase- PLA Steps 5 and 6 Requirements
Figure 51 Case Study: Design Phase- PLA Steps 5 and 6, SF and Dynamic Strain
89
Step 7: In this step, two PDF curves strength (step 3) and loading (step 6) plotted on a
chart and performed Boolean operation of subtraction on both curves to find out overlapping
probability for a given mode. Thus in this case, six overlapping probabilities are found.
Overlapping probability has standard deviation of strength 2 + load
2. Figures 52 and 53 depict
the 1B mode overlapping probability as an example.
Figure 52 Case Study: Design Phase- PLA Step 7 Requirement
Figure 53 Case Study: Design Phase- PLA Step 7 PDFs
90
Step 8: As discussed in step 7, all probabilities of failures (overlapping probability curve)
for all modes found, these probabilities were combined to find total probability of failure.
Also various confidence intervals were considered (50%, 90% and 95% or 99%) to determine
bands on the probability failure rates. Figure 54 shows the example of overlapping
probability for different operating conditions and material strength. Table 12 shows the
calculations for all three modes and two operating conditions probability of failures. It should
be noted that the 2B mode showed the highest probability of failure. For the other modes and
at any operating conditions, probability failure rates were low (much below 1%) in fact some
cases were almost zero. If one accepts the model based on the reliability of the original test
data then in this case the failure rate of 22% would not be acceptable as this repair (9 mm of
crack length) could cause more susceptibility for failure during service operation.
Figure 54 Case Study: Design Phase- PLA Step 8 Requirement
91
Table 12 Case Study: Design Phase- Overlapping Probability Failure Rate
Figure 55 shows the PDFs for 1B, 1T and 2B modes as well for material strength for both
operating conditions transient and steady state. From this figure it should be evident that
there is always a probability of failure during transient whereas during steady state, there was
no overlapping between strength and loading curves. Therefore, there is no probability of
failure during steady state operating condition. Again this example was shown for 9 mm
airfoil damage mode repair at the leading edge of the airfoil. There were different probability
of failures for other crack lengths 3 mm and 6 mm.
Figure 55 Case Study: Design Phase- PLA Step 8; 1B, 1T and 2B Modes PDFs
92
Figure 56 Case Study: Design Phase- PLA Step 8 Individual PDFs 1T and 2B Modes
Figure 56 shows the PDFs for individual 1T and 2B modes as well for material strength
for both operating conditions transient and steady state. Again from this figure it can be
evident that there is always a probability of failure during start transient operating condition
however probability of failure was very low.
Figure 57 shows the response surface for 2B mode. The response surface was plotted
using strength, loading and repair location stress. From this figure, it can be evident that there
is a significant damage that occurred to the airfoil up to 180 Mpa, which is endurance
strength of the material with weld debit factor.
93
Figure 57 Case Study: Design Phase- PLA Step 8; Response Surface Plot 2B Mode
Figure 58 Case Study: Design Phase- PLA Step 8; %Probability of Failures
94
Figure 58 shows the percent probability of failures for three different repair models that
were discussed in the introduction section of the case study. The repair geometries were
considered in the case study had 3 mm, 6 mm, 9 mm crack lengths for the evaluation
purposes. Steps 1 to 7 calculations were shown in this section discussed results for crack
length of 9 mm on the airfoil leading edge and at the center. These steps were repeated for 3
mm (concept 1) and 6 mm (concept 2) crack lengths. From this Figure 58, it can be evident
that the concept 3 with the crack length geometry of 9 mm was the worst case scenario and
had predicted probability of failure rate as 22%. Concept 2 with crack length of 6 mm had
predicted failure rate of 7% and concept 1with crack length of 3 mm predicted 1% failure
rate. All these failure rates were nominal. The values with the confidence interval such as
50%, 90% and 95% are also shown in the plot. An acceptance level of 5% was defined from
practical perspective. Thus any probability of failure concept above 5% would be rejected.
With this criterion, concept 1 could be accepted for repair whereas concept 3 should be
certainly rejected. In case of concept 2, where the probability of failure was 7%, the repair
crack length of 6 mm might be accepted if the risk/cost ratio below than 1.0. This would
require further study in order to accept the concept 2. Thus the PLA methodology was
helpful in order to assess the risk of repairing the part if it had to operate in the field in the
future. Also, the PLA methodology proposed in this section eliminates the need for GT unit
testing or validation and parts can be repaired and sent to the field operation right away
without the need of pilot design as the calculations are based upon the original GT unit test
data.
95
5.6 Proof Of Concept- Fixed Beam Example
In this section, a simple example is discussed to compare the results using the
methodology from the section 5.5 and classical analytical evaluation. The objective of this
section was to show the percentage error results from classical methods of analysis and PLA
approach for a fixed beam example. A fixed beam as shown in Figure 59 has a very similar
characteristic as that of compressor stator airfoil. For the actual part, it is very difficult to
predict the gas load dynamic conditions and any analytical calculations done might not
predict reliable results. Therefore it is always recommended to conduct GT unit testing
before original product release to assess the technical risks. Whereas many times the repair
justification is based upon FEA analysis in such cases it is very important to take into
consideration the previous GT testing experience with the original product design. In this
section as shown in Figure 59 a compressor stator vane is shown along with the loadings and
also a fixed beam along with the gas load, geometry and material details have been shown.
The fixed beam example was used as proof of concept in this scenario. The material of the
fixed beam was the same as that of stator vane discussed in the case study (304 SS) and the
material data is shown in the figure. A reference gage was placed at approximately 25% from
one of the end of the fixed beam. Repair geometry with 5 mm crack length was considered to
demonstrate the concept. Weld repair could be performed in this case (more material required
to be removed at least four times than the crack length during welding) and a debit factor of
25% was applied as discussed in the case study.
96
Figure 59 Case Study: Proof Of Concept- Fixed Beam Example Introduction
Three different analyses were executed using ANSYS FEM software in order to obtain
numerical results for the fixed beam and at the same time a scaling approach was used as
discussed in the case study to compare with numerical results. Initially steady state analysis
was carried out with pre stress “effects on” for the subsequent vibration analysis so as to
obtain dynamic responses. Figure 60 shows the maximum steady state stresses at the repair
location which was found to be 81 Mpa (equivalent stress). Maximum deflection was found
to be 1 mm and observed at the center of the fixed beam. After steady state analysis, free
vibration analysis was executed with pre stress “effects on”. The first six modes of vibration
1B, 1T, 2B, 2T, 3B and 3T were considered for study purposes and the modes shapes as well
frequency values are shown in the Figure 61.
97
Figure 60 Case Study: Fixed Beam- Steady State Stress and Displacement
Figure 61 Case Study: Fixed Beam- Free Vibration or Modal Analysis; First Six Modes
98
Table 13, shows the summary of results of modal or free vibration analysis for fixed
beam. The modal strains were collected at reference gage and repair location (average strain)
at each mode of vibration and a scale factor was created by taking the ratio of repair strain to
reference gage strain. This scale factor was used in further calculation. After modal analysis a
forced response or harmonic response analysis was executed using ANSYS FEM software.
The vibratory stresses observed in forced vibration analysis for each mode are shown in the
Figure 62. It can be noticed that 3B predicted dynamic stress of 169.8 Mpa at the repair
location which was very close to the weld debited endurance strength of the material (181
Mpa). The dynamic stress predictions from forced response analysis were the numerical
solution of the problem. These results were compared with the PLA method scaling
approach. During forced vibration analysis dynamic strains were also collected at the
reference gage and repair location so as to use for dynamic stress predictions using the scale
factor generated in the previous step (refer Table 13).
Table 13 Case Study: Design Phase- Proof of Concept, Fixed Beam-Free Vibration Analysis
99
Figure 62 Case Study: Fixed Beam- Forced Vibration Analysis Dynamic Stress
Table 14 Case Study: Design Phase- Fixed Beam-Forced Vibration Analysis
Table 14, shows the summary of results of forced vibration analysis for fixed beam. The
dynamics stresses shown in the table above were obtained from the forced vibration analysis.
These stresses were compared with the predicted stresses using PLA scaling approach.
100
Figure 63 Case Study: Fixed Beam- Closed Form Vs Predicted Dynamic Stress
Figure 64 Case Study: Fixed Beam- Goodman Diagram
101
Figure 63 shows the predicted vibratory stress using the PLA approach and FEA
vibratory (dynamic) stress for each mode of vibration. The predicted vibratory stress using
PLA approach was within 1% of error as compared with FEA numerical solutions results.
Thus the PLA approach predicts dynamic stresses very close to the actual fixed beam
responses.
Figure 64 shows the Goodman diagram for the predicted stresses obtained from PLA
approach for the fixed beam example. Goodman diagram constitutes of steady state and
vibratory stresses and shows safety margin for the part under study from dynamic loading
conditions perspective. In this case, it is evident that the 3B dynamic stress found to be above
the safety margin and the repair process is not acceptable. However, the purpose of fixed
beam example was to showcase the PLA scaling approach only.
The fixed beam example has articulated a proposed PLA method in this dissertation and
predicted results were very close with respect to the FEA calculations.
While analyzing power generation GT components for the particular loading condition
defined by high cycle fatigue, it is expected to have more error in the results due to FEA
modeling errors therefore it is required to estimate a confidence interval in order to establish
high or low limits on the probability failure rates. This was shown in the case study
calculation of step 8 discussed in the section 5.6.
102
5.7 Benefits Of The Proposed PLA Method
This section discusses expected benefits from the proposed PLA methodology and Table
15 describes the expected benefits. It can be noted that the PLA methodology proposed in
this dissertation is applicable to GT repair processes only. Design cost involves repair
concept generation and development trials. The comparison shown in the below table is
between current approach used for validation primarily using unit or engine testing and the
proposed PLA method is this dissertation.
Table 15 Case Study: Expected Benefits of Proposed PLA Method
Parameter Current Approach Proposed Approach
Validation Yes No
Pilot Design Yes No
Unit Testing Yes No
% Quality Yield 99.9% 99.9%
New build Yes No
Repair Yes Yes
Time for Eval Large Less
Immediate Product
Release
No Yes
Risk Low, (medium some
cases)
Low-Medium
Design cost High Low-Medium
103
5.8 Implement Phase
The implement phase discussed in the dissertation is very short as there is more emphasis
on the validation approach using PLA. A lot of aspects of implement phase are very similar
to a manufacturing process qualifications and the details of which can be found on the
internet. A general procedure is described in this section for manufacturing qualifications and
involves following activities,
Final repair geometry/Drawings
CTQ (critical to quality) dimensions and attributes
MQCP (Metrology quality control plan)
Fixtures/Tooling
Equipment/Machine
Manufacturing FMEA
Materials analysis
Final dimensional inspections and attributes
NDE (non destructive evaluation)
Statistical analysis for critical dimensions
104
CHAPTER 6 CONCLUSIONS
The existing repair development steps used for the power generation gas turbine parts
were streamlined using the six sigma DFSS methodology in order to achieve faster repair
development cycles and the total quality improvement.
The proposed DFSS cycle consisted four steps; Identify, Plan, Design & Implement
(called as IPDI or IPDcI).
The use of various six sigma tools was depicted throughout the different steps of the
proposed DFSS approach.
A probabilistic life assessment methodology was developed in order to verify the repair
concepts and choose the final repair geometry.
The probabilistic life assessment methodology proposed in this dissertation is applicable
to high cycle fatigue failure modes of gas turbine compressor airfoils.
A case study of compressor stator airfoil was used to demonstrate the step by step
application of the six sigma various phases using IPDI cycle.
Percent probabilities of failures were calculated for the case study repair concepts and
methodology was validated using classical fixed beam example.
It is expected to achieve reduction in premature failure (early life) of the repair parts
using the proposed six sigma approach as well as anticipated reduction in scrap rate.
105
CHAPTER 7 RECCOMMENDATIONS FOR FUTURE WORK
The proposed six sigma approach was applied to power generation GT parts. However, a
similar DFSS approach can be developed for repair processes used in the aircraft engine
parts as an aircraft engine is essentially a gas turbine.
The aircraft engine repair development cycle can also be streamlined in a similar way that
has been discussed in this dissertation.
The proposed PLA framework can be applied to other different damage modes such as
TMF (thermal mechanical fatigue), Creep etc. There is an opportunity for other
researchers to develop a similar PLA framework; however failure criteria and other
mathematical equations will be different.
The proposed PLA framework in the dissertation is applicable to crack initiation at the
repair location. However crack growth or propagation should be considered in addition to
the crack initiation for the calculation of final fracture failure rate which would represent
final failure of the part.
In addition to above, the proposed PLA framework can also be applied to other gas
turbine components from turbine and combustion modules. The failure modes of these
parts can be different from the case study discussed in this dissertation. It will be required
to make suitable changes in the PLA framework based on initial design requirements and
failure modes in order to predict failure rates for the other components.
106
REFERENCES
1. Arumugam V., Antony J., and Linderman K., 2014, “A Multilevel Framework of Six
Sigma: A Systematic Review of the Literature, Possible Extensions, and Future
Research”, Quality Management Journal - October 2014 , vol 24, # 4.
2. Watson G., DeYong C., 2010, “Design for Six Sigma: caveat emptor”, International
Journal of Lean Six Sigma, Vol. 1 Iss: 1, pp.66 – 84.
3. Snee R., 2010, “Lean Six Sigma – getting better all the time”, International Journal of
Lean Six Sigma, Vol. 1 Iss: 1, pp.9 – 29.
4. Pulakanam V., 2012, “Costs and Savings of Six Sigma Programs: An Empirical Study”,
Quality Management Journal - October 2012 , vol 19, #4.
5. Al‐Mishari S., Suliman S., 2008, “Integrating Six‐Sigma with other reliability
improvement methods in equipment reliability and maintenance applications", Journal of
Quality in Maintenance Engineering”, Journal of Quality in Maintenance Engineering,
Vol. 14 Iss: 1, pp.59 – 70.
6. Reosekar R., Pohekar S., 2014, “Six Sigma methodology: a structured review”,
International Journal of Lean Six Sigma, Vol. 5 Iss: 4, pp.392 – 422.
7. Min L., Ahn K., Yoon Y., 2012, “Development of Gas Turbine Combustion Turning
Technology Using Six Sigma Tools”, ASME Proceedings, Combustion, Fuels and
Emissions, June 11-15, 2012, SBN:978-0-7918-4468-7.
8. Dubikovsky J., 2015, “Incorporating Lean Six Sigma into an Aviation Technology
Program”, Journal of European Society for Engineering Education-SEFI, 2015.
107
9. Kaushik K., Khanduja D., 2008, “DM make up water reduction in thermal power plants
using six sigma DMAIC methodology”, Journal of Scientific and Industrial Research, vol
67, Jan 2008, pp36-42
10. Kane J., 2003, “Using Six Sigma to Drive Energy Efficiency Improvements at DuPont”,
American Council for an Energy Efficient Economy, SS03_Panel2_Paper 07, 2003
proceedings.
11. Fouquet J., 2007, “Design for Six Sigma and Lean Product Development : Differences,
Similarities and Links”, Asian Journal on Quality, Vol. 8 Iss: 3, pp.23 – 34.
12. Sinisuka N., Nugraha H., 2003, “Life cycle cost analysis on the operation of power
generation”, Journal of Quality in Maintenance Engineering, Vol. 19 Iss: 1, pp.5 – 24.
13. Samaranayake P., Kiridena S., 2012, “Aircraft maintenance planning and scheduling: an
integrated framework”, Journal of Quality in Maintenance Engineering, Vol. 18 Iss: 4,
pp.432 – 453.
14. Andrew T., Barton R., Byard P., 2008, “Developing a Six Sigma maintenance model”,
Journal of Quality in Maintenance Engineering, Vol. 14 Iss: 3, pp.262 – 271.
15. Yamashina H., Mizuyama H., “An inspection and repair strategy in automated assembly
systems”, Journal of Quality in Maintenance Engineering, Vol. 4 Iss: 2, pp.95 – 106.
16. Bañuelas R., Antony J., 2004, “Six sigma or design for six sigma?”, The TQM Magazine,
Vol. 16 Iss: 4, pp.250 – 263.
17. Gijo E., Sarkar A., 2013, “Application of Six Sigma to improve the quality of the road for
wind turbine installation”, The TQM Journal, Vol. 25 Iss: 3, pp.244 – 258.
108
18. Aggogeri F., Mazzola M., O'Kane J., 2009, “Implementing DFSS to increase the
performance level of an extrusion process”, Int. J. of Six Sigma and Competitive
Advantage, 2009 Vol.5, No.1, pp.10 – 28.
19. Shahin A., 2008, “Design for Six Sigma (DFSS): lessons learned from world-class
companies”, Int. J. of Six Sigma and Competitive Advantage, 2008 Vol.4, No.1, pp.48 –
59.
20. Baral M., 2014, ULBS, Phd Dissertation, “Integrating Knowledge Management Concepts
with Six Sigma Framework to Apply for Textile Manufacturing Processes”
21. Chakrabarthy A., 2009, National University of Singapore, PhD Dissertation, “Six Sigma
In Service Organizations: A Conceptual Framework Based On Aspects Of
Implementation And Performance”
22. Pallos K., 2001, “GE Gas Turbine Repair Technology”, Retrieved from http://site.ge-
energy.com/prod_serv/products/tech_docs/en/downloads/ger3957b.pdf.
23. Gardner M., 2006,“ Design for Six Sigma: A Key for Growth“, Retrieved from
“http://www.slideshare.net/Vijay_Bijaj/design-for-six-sigma-a-key-for-growth”.
24. Harry M., 2004, “Six Sigma Story- The Days at Motorola”, Retrieved from
http://www.mikeljharry.com/story.php?cid=8.
25. Houston D., 2003, “A Case Study in Software Enhancements as Six Sigma Process
Improvements: Simulating Productivity Savings”, Retrieved from
http://www.iienet2.org/uploadedfiles/IIE/Education/Six_Sigma_Black_Belt_Transition/p
rosim03_houston.pdf.
26. Ghunakikar S, 2015, “A Six Sigma DFSS Approach for Industrial Gas Turbine Repair
Products”, ASQ Lean and Six Sigma World Conference, Houston, TX, USA.
109
27. Singh R., Elrod C., Cudney E., 2012, “Comparative Analysis of Quality Function
Deployment Methodologies: A Case Study Analysis”, Quality Management Journal -
January 2012, vol 19, #1.
28. Latham J., 2012, “Management System Design for Sustainable Excellence: Framework,
Practices and Considerations”, Quality Management Journal - April 2012 , vol 19, #2.
29. Zhang W., Hill A., Gilbreath G., 2011, “A Research Agenda for Six Sigma Research”,
Quality Management Journal - January 2011 , Vol 18, #1.
30. Paryani K, Masoudi A., Cudney E., 2010, “QFD Application in the Hospitality Industry:
A Hotel Case Study”, Quality Management Journal - January 2010 , vol 17, #1.
31. Carter R., Lonial S., Raju P., 2010, “Impact of Quality Management on Hospital
Performance: An Empirical Examination”, Quality Management Journal -Oct 2010 , vol
17, #4.
32. Saeed S., 2006, “"The Repair and Overhaul of Gas Turbines: Some Notes on the
Organization and Facilities Used in the Maintenance of Jet Engines", Aircraft
Engineering and Aerospace Technology, Vol. 25 Iss: 7, pp.200 – 201.
33. Yilmaz O., Noble D., Nabil N., Gao G., 2005, “A study of turbomachinery components
machining and repairing methodologies", Aircraft Engineering and Aerospace
Technology”, Aircraft Engineering and Aerospace Technology, Vol. 77 Iss: 6, pp.455 –
466.
34. Timans W. , Ahaus K. , Antony J., 2014, “Six Sigma methods applied in an injection
moulding company”, International Journal of Lean Six Sigma, Vol. 5 Iss: 2, pp.149 –
167.
110
35. Singh B. , Bakshi Y., 2014, “Optimizing backup power systems through Six Sigma: An
Indian case study of diesel genset”, International Journal of Lean Six Sigma, Vol. 5 Iss:
2, pp.168 – 192.
36. Gijo E. , Bhat S. , Jnanesh N., 2014, “Application of Six Sigma methodology in a small-
scale foundry industry”, International Journal of Lean Six Sigma, Vol. 5 Iss: 2, pp.193 –
211.
37. Panat R., Dimitrova V. , Selvamuniandy T. , Ishiko K., Sun D., 2014, “The application
of Lean Six Sigma to the configuration control in Intel’s manufacturing R&D
environment”, International Journal of Lean Six Sigma, Vol. 5 Iss: 4, pp.444 – 459.
38. Jirasukprasert P. , Garza-Reyes J. , Kumar V. , Lim M., 2014, “A Six Sigma and
DMAIC application for the reduction of defects in a rubber gloves manufacturing
process”, International Journal of Lean Six Sigma, Vol. 5 Iss: 1, pp.2 – 21.
39. Besseris G., 2014, “Multi-factorial Lean Six Sigma product optimization for quality,
leanness and safety: A case study in food product improvement”, International Journal of
Lean Six Sigma, Vol. 5 Iss: 3, pp.253 – 278.
40. Nooramin A., Reza V., Sayareh J., 2011, “A Six Sigma framework for marine container
terminals”, International Journal of Lean Six Sigma, Vol. 2 Iss: 3, pp.241 – 253.
41. Roth N., Franchetti M., 2010, “Process improvement for printing operations through the
DMAIC Lean Six Sigma approach: A case study from Northwest Ohio, USA”,
International Journal of Lean Six Sigma, Vol. 1 Iss: 2, pp.119 – 133.
42. Rehman H., Asif M., Saeed M., Akbar M., Awan M., 2012, “Application of Six Sigma
at cell site construction: a case study", Asian Journal on Quality, Vol. 13 Iss: 3, pp.212 –
233.
111
43. Cheng C., Kuan C., 2012, “Research on product reliability improvement by using
DMAIC process: A case study of cold cathode fluorescent lamp", Asian Journal on
Quality, Vol. 13 Iss: 1, pp.67 – 76.
44. Lin C., Chen H., Li R., Chen C., Tsai C., 2009, “ Research on Increasing the Production
Yield Rate by Six Sigma Method: A Case of SMT Process of Main Board”, Asian
Journal on Quality, Vol. 10 Iss: 1, pp.1 – 23.
45. Lee K., Ree S., Park Y., 2005, “A Study of Lean DFSS through the Creative Value
Design”, Asian Journal on Quality, Vol. 6 Iss: 3, pp.121 – 131.
46. Kim J., Yim D., Choi J., Kim K., 2005, “A Methodology for Integrating Business Process
and Simulation for Business Process Redesign”, Asian Journal on Quality, Vol. 6 Iss: 1,
pp.74 – 97.
47. Park S., Park Y., 2004, “A Case Study of Six Sigma R&D Improvement Projects: Design
Optimization of Inner Shield Omega CPT”, Asian Journal on Quality, Vol. 5 Iss: 2, pp.63
– 69.
48. Blackmore J., Douglas A., 2003, “Towards a “better” University: the Use of the EFQM
Model in a UK Higher Education Institution”, Asian Journal on Quality, Vol. 4 Iss: 2,
pp.1 – 15.
49. Jichao X., Yumin L., Li Z., 2001, “A Classification Techniques For Quality
Improvement” Asian Journal on Quality, Vol. 2 Iss: 2, pp.24 – 33”
50. Emeka Izogo E. , Ogba I., 2015, “Service quality, customer satisfaction and loyalty in
automobile repair services sector”, International Journal of Quality & Reliability
Management, Vol. 32 Iss: 3, pp.250 – 269.
112
51. Singh S. , Singh B., 2014, “An approach to enhance availability of repairable systems: a
case study of SPMs”, International Journal of Quality & Reliability Management, Vol. 31
Iss: 9, pp.1031 – 1051.
52. Adhikary D. , Bose G. , Bose D. , Mitra S., 2014, " Multi criteria FMECA for coal-
fired thermal power plants using COPRAS-G", International Journal of Quality &
Reliability Management, Vol. 31 Iss: 5, pp.601 – 614.
53. Zhang M., Xie Y. , Huang L. , He Z., 2013, “Service quality evaluation of car rental
industry in China”, International Journal of Quality & Reliability Management, Vol.
31 Iss: 1, pp.82 – 102.
54. Assarlind M., Gremyr I., Bäckman K., 2013, “Multi‐faceted views on a Lean Six Sigma
application”, International Journal of Quality & Reliability Management, Vol. 30 Iss: 4,
pp.387 – 402.
55. Chakraborty A., Leyer M., 2013, “Developing a Six Sigma framework: perspectives from
financial service companies”, International Journal of Quality & Reliability Management,
Vol. 30 Iss: 3, pp.256 – 279.
56. Antony J., Bhuller A., Kumar M., Mendibil K., Montgomery D., 2012, “Application of
Six Sigma DMAIC methodology in a transactional environment”, International Journal of
Quality & Reliability Management, Vol. 29 Iss: 1, pp.31 – 53.
57. Psychogios A., Atanasovski J., Tsironis L., 2012, “Lean Six Sigma in a service context:
A multi‐factor application approach in the telecommunications industry” , International
Journal of Quality & Reliability Management, Vol. 29 Iss: 1, pp.122 – 139.
113
58. Chaudhuri D., Mukhopadhyay A., Ghosh S., 2011, “Assessment of engineering colleges
through application of the Six Sigma metrics in a State of India”, International Journal of
Quality & Reliability Management, Vol. 28 Iss: 9, pp.969 – 1001.
59. Ahuja I., Khamba J., 2008, “Total productive maintenance: literature review and
directions” International Journal of Quality & Reliability Management, Vol. 25 Iss: 7,
pp.709 – 756.
60. Antony J., Kumar M., Cho B., 2007, “Six sigma in service organisations: Benefits,
challenges and difficulties, common myths, empirical observations and success factors”,
International Journal of Quality & Reliability Management, Vol. 24 Iss: 3, pp.294 – 311.
61. Wong C., Tsang A., Chung T., 2006, “A methodology for availability assessment of
tunnel designs”, International Journal of Quality & Reliability Management, Vol. 23 Iss:
1, pp.60 – 80.
62. Seth N., Deshmukh S., Vrat P. , 2005, “Service quality models: a review”, International
Journal of Quality & Reliability Management, Vol. 22 Iss: 9, pp.913 – 949.
63. Battikha M., 2003, “Quality management practice in highway construction”, International
Journal of Quality & Reliability Management, Vol. 20 Iss: 5, pp.532 – 550.
64. Srinidhi B., 1998, “Strategic quality management”, International Journal of Quality
Science, Vol. 3 Iss: 1, pp.38 – 70.
65. Madu C., 1996, “A framework for environmental quality assessment”, International
Journal of Quality Science, Vol. 1 Iss: 3, pp.24 – 38.
66. Khalifa M. , Khan F. , Thorp J., 2015, “Risk-based maintenance and remaining life
assessment for gas turbines”, Journal of Quality in Maintenance Engineering, Vol. 21 Iss:
1, pp.100 – 111.
114
67. Markeset T., Moreno‐Trejo J., Kumar R., 2013, “Maintenance of subsea petroleum
production systems: a case study”, Journal of Quality in Maintenance Engineering,
Vol. 19 Iss: 2, pp.128 – 143.
68. El‐Thalji I., Liyanage J., 2012, “On the operation and maintenance practices of wind
power asset: A status review and observations”, Journal of Quality in Maintenance
Engineering, Vol. 18 Iss: 3, pp.232 – 266.
69. Márquez A., De León P., Gómez Fernández J., Márquez C., López M., 2009, “The
maintenance management framework: A practical view to maintenance management”,
Journal of Quality in Maintenance Engineering, Vol. 15 Iss: 2, pp.167 – 178.
70. Dohi T., Ashioka A., Osaki S., Kaio N., 2001, “Optimizing the repair‐time limit
replacement schedule with discounting and imperfect repair “, Journal of Quality in
Maintenance Engineering, Vol. 7 Iss: 1, pp.71 – 84.
71. Sherwin D., 2000, “A review of overall models for maintenance management”,
Journal of Quality in Maintenance Engineering, Vol. 6 Iss: 3, pp.138 – 164.
72. Kumar D., 1999, “New trends in aircraft reliability and maintenance measures”, Journal
of Quality in Maintenance Engineering, Vol. 5 Iss: 4, pp.287 – 295.
73. Ahmad S., 1996, “Life assessment study of a 400MW steam boiler at Ghazlan Power
Plant”, Journal of Quality in Maintenance Engineering, Vol. 2 Iss: 4, pp.48 – 58.
74. Goh M., Lim F., 1996, “Implementing TQM in an aerospace maintenance company”,
Journal of Quality in Maintenance Engineering, Vol. 2 Iss: 2, pp.3 – 20.
75. Thia C., Chai K., Bauly J., Xin Y., 2005, “An exploratory study of the use of quality tools
and techniques in product development”, The TQM Magazine, Vol. 17 Iss: 5, pp.406 –
424.
115
76. Papachristos D., Tsoukalas V., Vlachogiannis J., 2004, “Total quality plan applied in the
Hellenic power production process”, The TQM Magazine, Vol. 16 Iss: 2, pp.136 – 144.
77. Hong G., Goh T., 2003, "Six Sigma in software quality", The TQM Magazine, Vol. 15
Iss: 6, pp.364 – 373.
78. Bañuelas R., Antony J., 2003, “Going from six sigma to design for six sigma: an
exploratory study using analytic hierarchy process”, The TQM Magazine, Vol. 15 Iss: 5,
pp.334 – 344.
79. Ingle S., Roe W., 2001, “ Six sigma black belt implementation”, The TQM
Magazine, Vol. 13 Iss: 4, pp.273 – 280.
80. Xie M., Goh T., 1999, “Statistical techniques for quality”, The TQM Magazine, Vol. 11
Iss: 4, pp.238 – 242.
81. Zhang Z., 1998, “Application of experimental design in new product development”, The
TQM Magazine, Vol. 10 Iss: 6, pp.432 – 437.
82. Spring M., McQuater R., Swift K., Dale B., Booker J. 1998, "The use of quality tools
and techniques in product introduction: an assessment methodology", The TQM
Magazine, Vol. 10 Iss: 1, pp.45 – 50.
83. Starkey M., Aughton J., Brewin R., 1997, “Extending process thinking: design of
experiments in sales and marketing”, The TQM Magazine, Vol. 9 Iss: 6, pp.434 – 439.
84. Friday‐Stroud S., Sutterfield J., 2007, " A conceptual framework for integrating six‐
sigma and strategic management methodologies to quantify decision making". The TQM
Magazine, Vol. 19 Iss: 6, pp.561 – 571.
85. Kumar S., Satsangi P., Prajapati D., 2013, “Improvement of Sigma level of a
foundry: a case study”, The TQM Journal, Vol. 25 Iss: 1, pp.29 – 43.
116
86. Mahanti R., Antony J., 2009,”Six Sigma in the Indian software industry: some
observations and results from a pilot survey”, The TQM Journal, Vol. 21 Iss: 6, pp.549 –
564.
87. Jenicke L., Kumar A., Holmes M., “A framework for applying six sigma improvement
methodology in an academic environment”, The TQM Journal, Vol. 20 Iss: 5, pp.453 –
462.
88. Potra S, Pugna A., 2015, “DFSS in marketing: designing an innovative value co-creation
campaign”, Int. J. of Six Sigma and Competitive Advantage, 2015 Vol.9, No.1, pp.21 -
36.
89. Kubilius A., Winfrey ., Mayer C., Johnson G.,Wilson T.,2015, “ Applying Lean Six
Sigma tools to reduce the rate of slips, trips and falls for Joint Commission field staff”,
Int. J. of Six Sigma and Competitive Advantage, 2015 Vol.9, No.1, pp.37 – 55.
90. Jaglan P., Khanduja D., Kaushik P., 2013, “Capacity waste at thermal power plants in
India: a Six Sigma perception”, Int. J. of Six Sigma and Competitive Advantage, 2013
Vol.8, No.1, pp.22 – 33.
91. Shanmugaraja M., Nataraj M.,Gunasekaran N., 2012, “Total performance excellence - a
model to implement Six Sigma in service organisations”, Int. J. of Six Sigma and
Competitive Advantage, 2012 Vol.7, No.2/3/4, pp.209 - 242.
92. Bhat S., Jnanesh N., 2013, “Enhancing performance of the health information department
of a hospital using lean Six Sigma methodology”, Int. J. of Six Sigma and Competitive
Advantage, 2013 Vol.8, No.1, pp.34 - 50.
117
93. Shanmugaraja M., Nataraj M., Gunasekaran N.,2012, “Total performance excellence - a
model to implement Six Sigma in service organisations”, Int. J. of Six Sigma and
Competitive Advantage, 2012 Vol.7, No.2/3/4, pp.209 - 242.
94. Singh B., Khanduja D., 2010, “DMAICT: a road map to quick changeovers”, Int. J. of
Six Sigma and Competitive Advantage, 2010 Vol.6, No.1/2, pp.31 – 52.
95. Ray S., Das P., 2009, “Improving efficiency and effectiveness of APQP process by using
DFSS tools”, J. of Six Sigma and Competitive Advantage, 2009 Vol.5, No.3, pp.222 –
236.
96. Cronemyr P., 2007, “DMAIC and DMADV – differences, similarities and synergies”,
Int. J. of Six Sigma and Competitive Advantage, 2007 Vol.3, No.3, pp.193 – 209”.
97. Malliga P., Srinivasan S., 2007, “ The stock service improvement by the deployment of
Six Sigma”, Int. J. of Six Sigma and Competitive Advantage, 2007 Vol.3, No.2, pp.103 -
119.
98. Ferryanto L., 2007, “Analytical Design for Six Sigma for multiple response products”,
Int. J. of Six Sigma and Competitive Advantage, 2007 Vol.3, No.1, pp.13 - 32.
99. Heuvel J., Does R., Koning H., 2006, “Lean Six Sigma in a hospital”, Int. J. of Six Sigma
and Competitive Advantage, 2006 Vol.2, No.4, pp.377 - 388.
100 Kovach J., Cho B., 2006, “A D-optimal design approach to robust design under
constraints: a new Design for Six Sigma tool”, Int. J. of Six Sigma and Competitive
Advantage, 2006 Vol.2, No.4, pp.389 - 403.
101 Hu M., Pieprzak J., 2005, “Using axiomatic design to improve conceptual design
robustness in Design for Six Sigma (DFSS) methodology”, Int. J. of Six Sigma and
Competitive Advantage, 2005 Vol.1, No.3, pp.245 – 262.
118
102 Sarkar A., Mukhopadhyay A., Ghosh S., 2014, “Measurement system analysis for
implementing design for Six Sigma”, Int. J. of Productivity and Quality Management,
2014 Vol.14, No.3, pp.373 – 386.
103 Lee M., Chang T., 2010, “Developing a lean design for Six Sigma through supply chain
methodology”, Int. J. of Productivity and Quality Management, 2010 Vol.6, No.4,
pp.407 - 434.
104 Booysen C., Heyns P., Hindley M., Scheepers R., 2015, “Fatigue life assessment of a
low pressure steam turbine blade during transient resonant conditions using a
probabilistic approach”, International Journal of Fatigue, 73 (2015) 17-26
105 Zhao J., 1995, “Development and Demonstration of Advanced Structural Reliability
Methodologies for Probabilistic Fatigue Damage Accumulation of Aerospace
Component”