Tài liệu Evaluation of possible Six Sigma implementation including a DMAIC project pptx

MASTER’S THESIS
2004:070 CIV
Evaluation of possible
SIX SIGMA implementation
including a DMAIC project
A case study at the Cage Factory, SKF Sverige AB
MASTER OF SCIENCE PROGRAMME
Department of Business Administration and Social Science
Division of Quality & Environmental Management
2004:070 CIV • ISSN: 1402 - 1617 • ISRN: LTU - EX - - 04/70 - - SE
MARTIN LENNARTSSON
ERIK VANHATALO

-TITLES-

Evaluation of possible SIX SIGMA implementation
including a DMAIC project
- A CASE STUDY AT THE CAGE FACTORY, SKF SVERIGE AB
Utvärdering av möjlig SEX SIGMA implementering
samt ett DMAIC-projekt
- EN FALLSTUDIE PÅ HÅLLAREFABRIKEN, SKF SVERIGE AB
This Master Thesis was carried out within the area of Quality Management
at Luleå University of Technology and SKF Sverige AB.
By:
Martin Lennartsson
Erik Vanhatalo
Luleå 2004-02-29
Supervisors:
Lars Palmqvist, SKF Sverige AB in Gothenburg
Karin Schön, Luleå University of Technology

-ABSTRACT-

ABSTRACT
Six Sigma is an initiative launched by Motorola in 1987, focusing on
reducing variation and continuously improving processes (Barney, 2002).
This thesis was carried out at the Cage Factory in Gothenburg, a smaller unit
within SKF Sverige AB with 124 employees. The factory manufactures the
cage component in the bearing, which will keep the rollers in place in the
complete bearing.
The purpose of this thesis is to investigate and make comparisons between
Six Sigma and the existing way of working with improvements and the
organization in the Cage Factory and to give recommendations on what
actions are needed to efficiently implement Six Sigma.
To aid in the fulfillment of the purpose a practical DMAIC (Define-
Measure-Analyze-Improve-Control) project was conducted. The project
aims to reduce customer complaints and downtime caused by the turning
activity in one of the production channels at the Cage Factory. Furthermore,
several interviews with strategically selected individuals were conducted.
The authors argue that Six Sigma could be implemented and integrated with
the existing improvement approach, Total Process Management, TPMG. Six
Sigma can be used to attack the most complex problems, while TPMG
handles the many day-to-day issues.
At the moment the implementation strategy "Strategically selected
individuals and projects" is the most applicable at the Cage Factory. The
authors argue that Six Sigma will provide a structure (DMAIC) and training
in tools, thereby ensuring that the tools are used at the right time and in the
right way at the Cage Factory.
In the future, there has to be a strategy for generating possible Six Sigma
projects at the Cage Factory. Also, it is important that a way for steering

different problems to different problem solving activities is developed.
However, some issues need to be considered if Six Sigma should work
efficiently at the Cage Factory. A training venture is needed to enable the
introduction of different roles in the organization. Also, all training should
be connected to practical experience. It is important that the Cage Factory
receives resource support from SKF Sverige and/or the SKF Group. Further,
a reliable measurement system must be implemented. The authors believe
that improved scrap reporting and improved use of SPC are important
actions that need to be taken.


-SAMMANFATTNING-

SAMMANFATTNING
Sex Sigma fokuserar på att reducera variation samt att kontinuerligt förbättra
företagets processer och introducerades av Motorola 1987 (Barney, 2002).
Detta examensarbete har utförts på Hållarefabriken i Göteborg som är en
mindre enhet inom SKF Sverige AB med 124 anställda. Fabriken producerar
hållarkomponenten i lagret som har till uppgift att hålla rullar på plats i det
färdiga lagret.
Syftet med examensarbetet är att utvärdera och jämföra Sex Sigma med det
rådande förbättringsarbetet och organisationen på Hållarefabriken samt att
ge rekommendationer på vad som bör åtgärdas för att effektivt kunna införa
Sex Sigma.
För att uppfylla syftet har bland annat ett DMAIC-projekt utförts. Projektets
syfte är att reducera antalet kundreklamationer och minska stopptid orsakade
av svarvning i en av produktionskanalerna på Hållarefabriken. Vidare har ett
antal intervjuer genomförts med strategiskt utvalda individer.
Författarna menar att Sex Sigma kan implementeras och integreras i det
nuvarande förbättringsarbetet, Total Process Management (TPMG). Sex

Sigma kan användas för att attackera de mest komplexa problemen inom
verksamheten, medan TPMG hanterar de vardagliga åtgärderna som krävs
för att utveckla verksamheten på lång sikt.
För tillfället är implementeringsstrategin "strategiskt utvalda individer och
projekt" den mest lämpliga för Hållarefabriken. Författarna anser att Sex
Sigma kommer att ge en struktur (DMAIC) och träning i verktyg, vilket i sin
tur innebär att verktygen används vid rätt tillfälle och på rätt sätt på
Hållarefabriken.
I framtiden måste det finnas en strategi för att generera möjliga Sex Sigma-
projekt på Hållarefabriken. Dessutom är det viktigt att det utvecklas ett sätt
att styra olika problem till olika problemlösningsalternativ.
Det finns dock en del åtgärder som måste vidtas för att Sex Sigma ska kunna
fungera effektivt på Hållarefabriken. En utbildningssatsning krävs för att
möjliggöra en introduktion av roller i organisationen. All utbildning bör
dessutom ske i samband med deltagandet i ett praktiskt projekt. Det är
viktigt att Hållarefabriken får stöd, i form av resurser, från SKF Sverige och
SKF-koncernen. Författarna anser att en förbättrad kassationsrapportering
och förbättrad användning av programvaran för SPS är viktiga åtgärder för
att skapa ett väl fungerande mätsystem.

-ACKNOWLEDGEMENTS-

ACKNOWLEDGEMENTS
This Master Thesis was carried out during the period from late September
2003 to early February 2004 at the Cage Factory, SKF Sverige AB in
Gothenburg, Sweden.
We would like to take the opportunity to thank SKF Sverige AB and
especially the Cage Factory's entire staff for giving us the opportunity to
write our Master Thesis by studying their organization.
Furthermore, we also wish to thank our supervisor at Luleå University of

Technology, Luleå, LTU, Karin Schön as well as our supervisor at the Cage
Factory Lars Palmqvist for their valuable support during the completion of
this Thesis.
Moreover, the authors would like to express a special thanks to the members
of the improvement group taking part in the DMAIC project presented in
this Thesis. Without their committed participation this Thesis' completion
wouldn't have been possible. Also, we thank all the different individuals that
gave us some of their valuable time taking part in the interviews made in this
Thesis.
Last but not least, we would like to thank the Division of Quality and
Environmental Management at LTU for the opportunity to complete our
Master's study at their Division and for support during this and earlier
courses. Especially we would like to thank Görgen Edenhagen, Master
Thesis Coordinator and fellow course mates giving us feedback at Master
Student's seminars.
Gothenburg, February 2004


Martin Lennartsson Erik Vanhatalo




-CONTENTS-


CONTENTS
LIST OF FIGURES, PICTURES AND TABLES 1
FIGURES 1
PICTURES 2

TABLES 2
LIST OF ABBREVIATIONS 3
1. INTRODUCTION 4
1.1. BACKGROUND 4
1.1.1. The development of quality engineering 4
1.1.2. Company presentation 6
1.2. PROBLEM DISCUSSION 8
1.3. PURPOSE AND DELIMITATIONS 10
1.3.1. Purpose of the thesis 10
1.3.2. Delimitations 10
1.4. THE OUTLINE OF THE THESIS 11
2. METHODOLOGY 12
2.1. RESEARCH APPROACH 12
2.1.1. Positivism or Hermeneutics 12
2.1.2. Induction or Deduction 13
2.1.3. Quantitative or Qualitative method 14
2.2. RESEARCH STRATEGY 14
2.3. LITERATURE STUDY 15
2.4. CHOICE OF DATA COLLECTION METHOD 16
2.4.1. Primary data 16
2.4.2. Secondary data 18
2.5. METHODOLOGY AND CHOSEN TOOLS IN THE DMAIC PROJECT 18
2.6. RELIABILITY AND VALIDITY 19
2.6.1. Validity 19
2.6.2. Reliability 20
3. THEORETICAL FRAME OF REFERENCE 21
3.1. SIX SIGMA 21
3.1.1. The Six Sigma Framework 23
3.1.2. The Six Sigma Infrastructure 24
3.1.3. Strategies for Six Sigma Implementation 25

3.1.4. Critique to Six Sigma 27
3.2. THE DMAIC IMPROVEMENT CYCLE 28

-CONTENTS-


3.2.1. How to choose a Six Sigma project 29
3.2.2. The Define Phase 29
3.2.3. The Measure Phase 30
3.2.4. The Analyze Phase 31
3.2.5. The Improve Phase 31
3.2.6. The Control Phase 31
3.2.7. Examples of tools in the different phases 32
3.2.8. Tools Chosen in the DMAIC project 33
3.3. OTHER QUALITY INITIATIVES 34
3.3.1. Total Quality Management, TQM 34
3.3.2. Total Process Management and TPM, the SKF way 35
3.4. TURNING 37
4. THE EMPIRICAL STUDY 38
4.1. THE DMAIC PROJECT 38
4.1.1. The Define Phase 38
4.1.2. The Measure Phase 41
4.1.3. The Analyze Phase 47
4.1.4. The Improve Phase 49
4.1.5. The Control Phase 52
4.1.6. Evaluation of the DMAIC project 53
4.2. INTERVIEWS 54
4.2.1. Lars Arrenäs, TPMG Manager, SKF 55
4.2.2. Cecilia Lack, TPMG Coordinator at the Cage Factory 56
4.2.3. Folke Höglund, Quality Assurance Manager, SKF Group 57

4.2.4. Bo Bergman, Professor, Chalmers University of Technology, CTH 59
4.2.5. Laszlo Persson, Master Black Belt at Volvo Cars Engine in Skövde 61
4.3. FOCUS GROUP AT THE CAGE FACTORY 68
5. ANALYSIS AND RESULTS 72
5.1. EXPERIENCE FROM THE DMAIC PROJECT 72
5.1.1. The scope of the project 72
5.1.2. The cost issue 73
5.1.3. Collection and evaluation of data 73
5.1.4. DMAIC methodology, tools and results 74
5.2. ANALYSIS OF SIX SIGMA IMPLEMENTATION AT THE CAGE FACTORY 76
5.2.1. Why adopt Six Sigma? 76
5.2.2. The Six Sigma Framework at the Cage Factory 77
5.2.3. Organization and Six Sigma Roles 80
5.2.4. Methodology and Tools 81


-CONTENTS-


6. CONCLUSIONS AND DISCUSSION 82
6.1. CONCLUSIONS 82
6.2. DISCUSSION 83
6.2.1. Thesis generalization 83
6.2.2. Criticism to sources and methodology 83
6.2.3. Problems with the purpose of the thesis 85
6.2.4. Future work 85
REFERENCES 86
BIBLIOGRAPHIC REFERENCES 86
ELECTRONIC REFERENCES 87
PAPER REFERENCES 88

ORAL REFERENCES 88
OTHER REFERENCES 89
SUPPLEMENTS 90
APPENDICES 102
GLOSSARY 110
INDEX 111



-CONTENTS-


LIST OF FIGURES, PICTURES AND TABLES
Figures
Page
FIGURE 1.1 DEVELOPMENT OF THE QUALITY CONCEPT 5
FIGURE 1.2 THE SKF BUSINESS CONCEPT 6
FIGURE 1.3 THE SKF QUALITY POLICY 7
FIGURE 1.4 SKETCH OF THE WORKFLOW IN CHANNEL 13 AT THE CAGE FACTORY 9
FIGURE 2.1 INDUCTIVE AND DEDUCTIVE APPROACH IN RESEARCH 13
FIGURE 3.1 NORMALLY DISTRIBUTED PROCESS PERFORMING AT SIX SIGMA LEVEL 22
FIGURE 3.2 COSTS, IN PER CENT OF TURNOVER, DEPENDING ON SIGMA LEVEL 22
FIGURE 3.3 THE CORPORATE FRAMEWORK OF SIX SIGMA 23
FIGURE 3.4 STRUCTURE OF A GENERAL SIX SIGMA PROJECT 25
FIGURE 3.5 STRATEGIES FOR SIX SIGMA IMPLEMENTATION 26
FIGURE 3.6 THE DMAIC METHODOLOGY 28
FIGURE 3.7 COMMONLY USED SOURCES FOR PROJECT GENERATION 29
FIGURE 3.8 VARIATION IN INPUT VARIABLES ARE TRANSFERRED TO THE OUTPUT 30
FIGURE 3.9 COMMONLY USED TOOLS IN A DMAIC PROJECT 33
FIGURE 3.10 QUALITY TOOLS IN THE DMAIC PROJECT 33

FIGURE 3.11 LINK BETWEEN VALUES, METHODOLOGIES AND TOOLS 35
FIGURE 3.12 THE OPERATOR MAINTENANCE STAIRCASE 36
FIGURE 3.13 THE CONCEPT OF LONGITUDINAL AND FACE TURNING 37
FIGURE 4.1 PARETO CHART OVER CUSTOMER COMPLAINTS 39
FIGURE 4.2 HISTORICAL DEVELOPMENT OF COMPLAINTS IN CHANNEL 13 39
FIGURE 4.3 PARETO CHART FOR IDENTIFIED COMPLAINT CAUSES IN CHANNEL 13 39
FIGURE 4.4 PROCESS MAP OF THE TURNING ACTIVITY 41
FIGURE 4.5 PIE CHART OF DOWNTIME IN DIFFERENT MACHINES IN CHANNEL 13 42
FIGURE 4.6 PARETO CHART OF DURATIONS FOR DIFFERENT CAUSES 43
FIGURE 4.7 RESULT OF THE FREQUENCY STUDY IN THE TURNING PROCESS 44
FIGURE 4.8 THE OUT OF CONTROL SITUATION FOR ONE OF THE STUDIED CAGES 49
FIGURE 4.9 THE KANO MODEL 60
FIGURE 4.10 SKETCH OF THE SIX SIGMA ORGANIZATIONAL STRUCTURE AT VCES 64
FIGURE 4.11 PROJECT LIFE AND RESPONSIBILITIES AT VCES 65
FIGURE 4.12 PROBLEM SOLVING ACTIVITIES AT VCES 65
FIGURE 5.1 HOW SIX SIGMA AND TPMG DEALS WITH PROBLEMS 77
FIGURE 5.2 TPMG AND SIX SIGMA AT THE CAGE FACTORY 79
FIGURE S.1 PROJECT CHARTER USED IN THE DMAIC PROJECT 92
FIGURE S.2 FREQUENCY STUDY FORM USED IN THE DMAIC PROJECT 93
FIGURE S.3 CAUSE-AND-EFFECT DIAGRAM FROM THE DMAIC PROJECT 95
-1-

-CONTENTS-


FIGURE S.4 PROCESS-FMEA DEVELOPED IN THE DMAIC PROJECT 98
FIGURE S.5 TREE DIAGRAM DEVELOPED IN THE DMAIC PROJECT 100
FIGURE S.6 MATRIX DIAGRAM DEVELOPED IN THE DMAIC PROJECT 101
FIGURE A.1 CONVERTING DPMO TO SIGMA VALUES 103
FIGURE A.2 THE SEVEN-TIMES-SEVEN TOOLBOX 104

Pictures
PICTURE 1.1 CAGE AND RADIAL BEARING 7
PICTURE 4.1 THE IMPROVEMENT GROUP 42
PICTURE 4.2 DEFECTS IN THE TURNING ACTIVITY 44
Tables
TABLE 4.1 CUSTOMER COMPLAINTS PER PRODUCTION CHANNEL 38
TABLE 4.2 POTENTIAL FAILURE CAUSES WITH THE HIGHEST RPN 48
TABLE 4.3 MEASURES FROM THE TREE DIAGRAM WITH 18 POINTS 50
TABLE 4.4 MEASURES CORRELATED TO PROBLEMS DETECTED IN THE FMEA 51
TABLE S.1 THE CUSTOMER COMPLAINT PROCESS 96
TABLE A.1 TRAINING CONTENT FOR DIFFERENT ROLES 103

-2-

-ABBREVIATIONS-


LIST OF ABBREVIATIONS
ABB Asea Brown Bovery
ANOVA Analysis of Variance
ANOM Analysis of Means
C
p
Process Capability Ratio
C
pk
Process Capability Ratio, considering centering
CTH Chalmers University of Technology
CTQ Critical To Quality
DFSS Design For Six Sigma

DMAIC Define, Measure, Analyze, Improve, Control
DPMO Defects Per Million Opportunities
DOE Design Of Experiments
FADE Focus, Analyze, Deploy and Evaluate
FMEA Failure Modes and Effects Analysis
H1, H2, H3 Production Section 1, 2 and 3 in the Cage Factory
ISO International Standardization Organization
KLEMM "Kvalitet" (Quality), "Leverans" (Delivery), "Ekonomi" (Economy), "Miljö"
(Environment) and "Medarbetare" (Co-worker)
KTI Kvalitets Tillstånds Information (Quality State Information)
LSL Lower Specification Limit
LTU Luleå University of Technology
MBNQA Malcolm Baldridge National Quality Award
MCSS Manufacturing Customer Service System
NC Numerically Controlled
PDCA Plan, Do, Check, Act
PTS Project Tracking System
QFD Quality Function Deployment
QIT Quality Improvement Teams
PCR Process Capability Ratio
R&R Repeatability & Rreproducibility
RPN Risk Priority Number
SEK Swedish Krona (Currency)
SIPOC Suppliers, Inputs, Process, Outputs, Control
SIQ Institutet För Kvalitetsutveckling
SKF Svenska Kullagerfabriken
SPC Statistical Process Control
TPM Total Productive Maintenance
TPMG Total Process Management
TQM Total Quality Management

TRIZ Theory of Inventive Problem Solving
USL Upper Specification Limit
VCES Volvo Cars Engine in Skövde
VMEA Variance Mode and Effect Analysis
VOC Voice Of Customer
-3-
-INTRODUCTION-


1. INTRODUCTION
This chapter begins with the background of the thesis, containing an overview of the quality
concept and a company presentation. Then a problem discussion will follow and finally the
purpose and delimitations of the thesis will be presented to the reader.
1.1. Background
1.1.1. The development of quality engineering
An important question that Company Managers are asking themselves is "How do
we become successful?", but a question of equal importance is “How do we stay
successful in the future?” (Pande, Neuman & Cavangh, 2000).
A company cannot survive without customers. According to Pyzdek (2003) it is
therefore very important that the company provides products that the customers are
willing to pay for. In plain language this means that the ultimate goal for the
company is to create value to the customer. Hence, the customer settles the quality
of a product.
The word quality has its origin from the Latin word "qualitas”, which means
“character” (Bergman & Klefsjö, 2001). There are several different definitions of
the Quality Concept and many different opinions of what should be included in the
concept of product quality. The authors have fallen for a definition of the quality of
a product from Bergman & Klefsjö (2001):
"The quality of a product is its ability to satisfy and preferably exceed the
needs and expectations of the customers".

(Translated from Bergman & Klefsjö, 2001 p.24)
The approaches that have been used to deal with quality problems have changed
over time. Bergman & Klefsjö (2001) mean that the dominating quality technique
used shortly after World War II was Quality Control of finished products, a
defensive technique. Since then, the development direction has been to increase the
efforts before the production process begins and also to work with continuous
improvements. This development is illustrated in Figure 1.1. The Swedish Institute
of Quality, SIQ, shares this point of view (SIQ, 2003).
-4-

-INTRODUCTION-



Quality Management
…continuous improvements before, during and after production
Quality Assurance
…before production
Quality Monitoring
…during production
Quality Control
…after production
Development

Figure 1.1 Development of the Quality Concept from the middle of the twentieth century.
Source: Bergman & Klefsjö (2001, p.94).
In the more recent history of the quality development, the quality improvement
program Six Sigma has been successful. The American company Motorola
developed Six Sigma as a consequence of poor quality and customer complaints,
which affected the competitive power of the company negatively (Barney, 2002).

In 1986 Bill Smith, engineer and statistician at Motorola, introduced the Six Sigma
concept aiming to attack the existing quality problems in the company.
Motorola began documenting their key processes, aligning them toward customer
requirements, measuring and analyzing to be able to improve their processes
continuously and reduce variation (Barney, 2002).
In 1988 Motorola won the Malcolm Baldridge National Quality Award, MBNQA
1

and the interest for Six Sigma increased (Pyzdek, 2001).
Since Motorola launched Six Sigma in 1987 and particularly from 1995, a growing
number of global companies have followed, developing Six Sigma programs of
their own (Magnusson, Kroslid & Bergman, 2003). Today, Six Sigma is well
established in the automotive, aviation, chemical, electronic and metallurgy
industries (ibid).
A recent successful example of Six Sigma implementation in Sweden is Volvo
Cars Corp., which after three years and around 500 completed Six Sigma projects
presents net savings of about one Billion SEK (Dahlquist, 2003).
Bergman & Klefsjö (2001) claims that the goal of Six Sigma is to substantially
reduce unwanted variation that either results in cost reductions or increased
customer satisfaction. The reduced variation may also lead to improved delivery
performance and increased process yield.


1
The MBNQA is an annual quality award in the USA and was established in 1987 when the sitting
president Ronald Reagan signed the MBNQ improvement Act. The award is named after a former
American Secretary of Commerse, Malcolm Baldridge (Dale, 1999).
-5-
-INTRODUCTION-

1.1.2. Company presentation
Findings of simple forms of ball bearings from archeological excavations have
been dated to the days of the Roman Empire. The basic function of a bearing is to
facilitate the rotation of wheels and axles. When the bicycle with pedals was
introduced during the 19
th
century, the importance of bearings grew to make it
easier for the user to ride the bike. This sped up the development of ball bearings.
In 1907 the Swedish engineer Sven Wingquist introduced the first self-aligning
2

ball bearing and later that year he founded "Svenska KullagerFabriken", (the
Swedish Ball Bearing Factory), SKF (SKF, 2003a).
In 2002 The SKF Group, having business activities at 79 locations around the
world, had an annual turnover exceeding SEK 42 Billion and presented a pre-tax
profit of SEK 3.5 Billion. The SKF Group employs 39,700 people, of which 4,600
in Sweden (SKF, 2003e).
The SKF Group has five subdivisions: Automotive, Electrical, Industrial, Service
and Aero and Steel Division. These divisions also include a number of subsidiaries
located in different countries around the world.
In Sweden the activities are centered round SKF Sverige AB and its thirteen
subsidiaries, which work with everything from refining raw material to business
development (SKF, 2003c).
The SKF business concept is presented in Figure 1.2.
SKF's mission is to enhance and develop global leadership in bearings, seals, related products,
systems and services.
Values
- Empowerment
- High ethics

- Openness
-Teamwor
k
D
rivers
- Profitability
- Quality
- Innovation
- Speed
(SKF, 2003e p.1)
Our aim is to be the best in the industry at:
- providing customer value
- developing our employees
- creating shareholder value


Figure 1.2 The SKF Business Concept.
SKF's attitude and commitment to quality is communicated through the Quality
Policy, which is displayed in Figure 1.3.


2
A self-aligning bearing is characterized by its ability to adjust to skewed axles. This is possible because
the inner ring of the bearing is spherical which allows the inner and outer ring to interact (Huttunen,
2003a).
-6-
-INTRODUCTION-


Aim for Total Quality in everything we do

Market only products and services that will ensure customer satisfaction by:
- Operating reliable and capable processes
- Maintaining a program of continuous improvement
(SKF, 2003d)

Figure 1.3 The SKF Quality Policy
SKF's largest individual customer is the French Railway. Other important customer
segments are for example crushing machine manufacturers, paper mills and wind
power plants (Huttunen, 2003).
This thesis is focusing on the situation at the Cage Factory, SKF Sverige in
Gothenburg. The factory is manufacturing the cage component in a bearing. The
function of a cage is to keep the rollers in place in the complete bearing. Picture 1.1
shows a cage for a radial bearing. There are 124 employees at the factory and the
annual turnover is about SEK 170 Million. The factory is divided into three
sections with a total of 14 production flows. There are about 200 standard versions
of cages in production with dimensions 100-2,008 mm in diameter (Huttunen,
2003).
70 per cent of the production moves on to assembly at SKF's other factories in
Gothenburg. More then 20 per cent are assembled at factories in the USA and the
remaining part goes to factories in England and Malaysia. The annual production is
roughly 2.5 Million cages and the dominating types are cages for radial bearings
3
,
which constitute about 95 per cent of the production (ibid). A cage for a radial
bearing and a complete radial bearing can be viewed in Picture 1.1.
(a) (b)
Picture 1.1 (a) Cage for a radial bearing. (b) Complete radial bearing.

3
The radial bearings are designed to carry load in radial direction of the shaft.

-7-
-INTRODUCTION-


1.2. Problem discussion
There are many ways to deal with quality problems in a manufacturing company.
The past 15-20 years many new methods, strategies and tools have emerged in the
quality area, for example Total Quality Management, TQM, has been a popular
approach that many companies have adopted (Pande et al., 2000).
In recent years Six Sigma has grown in popularity especially in the US and
companies like General Electric and Motorola have obtained significant
improvements in their performance (Pande et al., 2000). A reason for their success
is probably Six Sigma's ability to prove reduced costs or higher profits in
economical measures.
The SKF Group has taken a decision to start a Six Sigma initiative within the entire
group during 2004. Each division will be responsible for its own time plan to
launch Six Sigma (de Laval, 2003).
Earlier, the Automotive Division has decided to start a Six Sigma initiative within
the division and has recently started a Black Belt training program. Chicago
Rawhide (USA) for example, working in the seal area has completed roughly, 300-
400 Six Sigma projects (Nielsen, 2003).
The Cage Factory is a part of the Industrial Division, which has shown an early
interest for investigating how Six Sigma could be implemented in the division.
Therefore an evaluation of Six Sigma implementation in the Cage Factory is
valuable.
In a traditional organization the structure is designed to carry out routine tasks. In
Six Sigma most activities and problems are unique (Pyzdek, 2003). What is this
demanding of the Six Sigma Company?
A company that aims for an implementation of Six Sigma also has to consider
ongoing improvement activities and systems. Management must therefore plan how

overlapping activities should be organized in a way to prevent confusion and reach
a positive cooperation between the current system and Six Sigma (Pyzdek, 2003).
There is a five years old Total Process Management initiative running at the Cage
Factory, which has to be taken into consideration. Other aspects, such as different
culture and employee devotion, also have to be considered.
As mentioned previously, there are fourteen different production flows in the Cage
Factory. The workflow is principally divided into the same main activities.
These are shown in Figure 1.4 (main activities in bold font). The figure displays
the principle process in the different lines and specifically the flow in Channel 13.
-8-
-INTRODUCTION-


Washing:
Remove superfluous
oil
Raw material
in:
- Discs
Cup drawing:
Sheet pressing of raw
material
Blasting:
Remove burrs and
sharp edges
Turning:
- Correct dimensions for the cage
- Eliminate non-conformities in surface
Bevelling:
"Bars" are adjusted to

rollers
Delivery:
Customer
shipment
Notching:
Roller pockets are
p
unched out

Piercing:
Hole diameter is cut to
the right dimensions
Phosphating:
Surface layer is
added
Packing:
Packing of complete
cages
A
E
C
D F
B

Figure 1.4 Sketch of the workflow in Channel 13 at the Cage Factory, SKF Sverige AB,
Gothenburg.
There has been an increase of complaints in Channel 13 until October 2003. Most
of these complaints can be derived from Turning (activity F in Figure 1.4.). The
cages are turned on two sides as showed in Picture 1.2. The two sides of the Cage
are called small end and large end. There are both esthetical and capability

problems. There are also efficiency losses, in form of downtime, connected to these
problems.
The use of the five-phase improvement cycle DMAIC
4
, within Six Sigma
companies, has become increasingly common (Pande et al., 2000).

Small end
of cage
Large end of
Cage
Picture 1.2 Turning surface on large end and small end of the Cage. Source: Own work
This particular problem provides an opportunity to further investigate if and how
the Six Sigma methodology (DMAIC) and tools can be used successfully in the
Cage Factory.


4
The acronym DMAIC is an abbreviation for the five phases in the Six Sigma improvement project,
namely Define, Measure, Analyze, Improve and Control.
-9-
-INTRODUCTION-


1.3. Purpose and delimitations
1.3.1. Purpose of the thesis
The main purpose of this thesis is to evaluate and make comparisons between Six
Sigma and the existing way of working and the organization in the Cage Factory.
This will end up in recommendations on how Six Sigma can be implemented in the
Cage Factory and which actions need to be taken to efficiently work with Six

Sigma in the organization.
To aid in the fulfillment of the main purpose a practical improvement project will
be carried out. In this project the DMAIC methodology will be used. The practical
part of the project will deal with quality shortages in the production process
concerning the turning device, analyzing the reasons for these shortages and
suggesting actions to improve the situation.
1.3.2. Delimitations
When studying the possibility of Six Sigma implementation the focus will be on
how, and in which form, a Six Sigma venture can exist within the current
organization and how a general Six Sigma project can be conducted.
The authors have chosen not to study the possibility of implementing Design For
Six Sigma, DFSS, at the Cage Factory. This choice was made since the authors
have a limited time of 20 weeks to conduct the study. Also, a natural starting point
of a Six Sigma venture is the use of Six Sigma in the production and not in the
design phase (Professor Bergman, 2003).
The practical improvement project will deal with quality shortages in the turning
device in Production Channel 13, since there has been an increase of customer
complaints in the Channel until October 2003.
The authors decided to study only one of the turning devices in one of the
production channels because of the complexity of the process and the limited time
available. The types that are being produced in Channel 13 during the time of the
study limit the types of Cages that will be included in the study.
-10-
-INTRODUCTION-


The improvement project will be carried out between 2003-10-01 and 2004-01-13.
Since the authors have a limited amount of time to conduct the practical
improvement project the Control Phase in the DMAIC cycle will not be carried out.
This phase need several months of monitoring to be properly evaluated. The

authors will instead give recommendations on how this phase could be conducted.

The thesis will not take all possible Six Sigma tools into account but a selection of
appropriate tools will be used in the different phases.
1.4. The outline of the thesis
In this section the outline of the thesis is presented to the reader. An overview of
the following chapters is given in Figure 1.5. The figure is also an attempt to
present a composition of the thesis to the reader.
METHODOLOGY Chapter 2
THEORETICAL FRAME
OF REFERENCE
Chapter 3

Deduction
Evaluation of Six Sigma
implementation at the Cage
Factory
Chapter 4
ANALYSIS Chapter 5
CONCLUSIONS & DISCUSSION
Chapter
6
Improvement
Project
DMAIC
INTRODUCTION TO THE THESIS

Chapter 1

Figure 1.5 The outline of the thesis.
The first chapter gives the reader an introduction to the thesis. The second chapter
discusses the methodology chosen to solve the problem. Since the authors have
chosen a deductive approach (see Section 2.1.2), the theoretical frame of reference
is then presented to the reader in chapter three. Chapter 4 contains the results of the
empirical study at the Cage Factory. These results are then analyzed in Chapter 5
and finally, in Chapter 6, the authors' conclusions and a general discussion are
presented.
-11-
-METHODOLOGY-


2. METHODOLOGY
In this chapter the methodology of the thesis is presented. Different aspects of the research
approach and research strategy are discussed. The chapter also describes the study of literature,
choice of data collection, methodology and chosen tools in the DMAIC project and finally a
discussion of the validity and reliability of the thesis.
Holme & Solvang (1991) argue that methodology is a tool or a way to solve
problems and thereby get new knowledge. Everything that is helping the
researchers to reach their goals is methodology.
2.1. Research approach
A governing thought within modern science is that research results should be
published and used freely to aid the growth of science. Other researchers must be
able to review models, methods and results. Are the data valid? Are the
interpretations and analyses reliable? Are the conclusions only applicable under
certain circumstances or are they of a more general nature? (Wiedersheim-Paul &
Eriksson, 1993)
2.1.1. Positivism or Hermeneutics
Humans observe the world in different ways and that's the reason why there are
different perspectives of science and knowledge. There are two extremes of

research directions that are dominating today, namely Positivism and Hermeneutics
(Thurén, 1991)
The meaning of Positivism is that a science thesis only has a value if it could be
empirically verified. Everything that can't be tested empirically, like feelings,
values, religious and political statements don't belong to the scientific sphere
(Wallén, 1993). Thurén (1991) means that Positivism's only sources of knowledge
are observations and logic.
The opposite of Positivism is Hermeneutics, which can be translated to
interpretation science and origins in theories about bible and other text
interpretations. Hermeneutics is about interpretation of meanings in its widest sense
(Wallén 1993).
This thesis, investigating a possible Six Sigma implementation at the Cage Factory,
has got elements of both Hermeneutics and Positivism. Qualitative data is collected
through interviews, a focus group and by running a DMAIC project. The authors'
interpretations and analysis of these sources of data, founded upon interpretations
of existing theories, are examples of a hermeneutic approach.
-12-

-METHODOLOGY-


However, there are also elements of Positivism. The DMAIC project concerning
the turning related problems also includes quantitative data and logical and fact-
based decisions, a positivistic approach. In fact, the evaluation of a possible Six
Sigma implementation must be based on observations and empirical results from
the current situation at the Cage Factory, consequently a positivistic approach.
2.1.2. Induction or Deduction
Rationalism is a philosophical branch, which claims that it is possible to obtain
knowledge about reality only by using reason (Nationalencyklopedin, 2003).
Empiricism on the other hand is a branch that, in contrast to rationalism,

emphasizes experience as the base of our knowledge (Prawitz, 2003). According to
Wiedersheim-Paul & Eriksson (1993) empiricism and rationalism lead to two
principally different approaches to research, see Figure 2.1, Deduction and
Induction:
− Deduction: Hypotheses, that can be verified, are formed on the basis of a
theory. Results will be obtained through logical contemplation.
− Induction: On basis of separate phenomena in reality, more general
conclusions are drawn and theories and models are formed.

REALITY
Generalisation
Observations
Hypotheses
Observations
INDUCTIVE APPROACH
DEDUCTIVE APPROACH
THEORY

Figure 2.1 Inductive and deductive approach in research. Source: Wiedersheim-Paul & Eriksson
(1993, p. 150).

In the deductive approach to research, theory has got a more important and
independent position than the inductive approach (Wallén, 1993).
This thesis has got a deductive approach. The starting-point of the thesis is a review
of existing theories in the area. Then the empirical studies are conducted, after
which the empirical results are analyzed on the basis of existing theories.
-13-
-METHODOLOGY-

2.1.3. Quantitative or Qualitative method
Holme & Solvang (1991) argue that there are principally two different types of
research methods, quantitative and qualitative methods.
Qualitative data consists of detailed descriptions of for example situations, events
and people. It can give information about different people's experiences, attitudes,
opinions and thoughts. Quantitative data give information about how many, how
much, amount, frequency and the distribution of the data (Merriam, 1994).
Qualitative methods have primarily a purpose of understanding and by using
different data collection methods get a deeper understanding of the studied problem
(Holme & Solvang, 1991). Quantitative methods are more formalized and
structured and often mean more control from the researcher. Statistical methods
play an important role within quantitative research (ibid).
According to Holme & Solvang (1991) it can be successful to combine qualitative
and quantitative methods because advantages and drawbacks with each method
complement each other.
In this thesis there are elements of both qualitative and quantitative data or
information. The majority of the data or information gathered is of a qualitative
nature, in the form of interviews, focus groups and the use of qualitative tools in the
improvement project. However, quantitative data or information is also gathered by
the use of check sheets and review of quantitative data provided by SKF from the
turning process.
2.2. Research strategy
There are several ways of doing research. Examples are experiments, surveys,
historical studies, analyses of archival information and case studies (Yin, 1994).
According to Yin (1994), case studies are the preferred strategy when the questions
"how" or "why" are being asked. In this thesis, questions like "how can the Six
Sigma methodology be implemented to improve the existing production process?"
and "how and why should SKF introduce Six Sigma in the organization?" are valid.
To be able to fulfill the purpose of the thesis, the authors concluded that a thorough
study of the company's current way of working was needed. Likewise the authors

saw the need for a careful analysis of the DMAIC project.
-14-
-METHODOLOGY-


This motivated the choice of a Case Study as the starting point of the data
collection methods used in the thesis.
Yin (1994 p. 3) describes the Case Study strategy as follows:
"In brief, the Case Study allows an investigation to retain the holistic and
meaningful characteristics of real-life events - such as individual life cycles,
organizational and managerial processes, neighborhood change,
international relations and the maturation of industries."
According to Wallén (1993) the advantages of a Case Study are that the research is
carried out under real circumstances and it provides the possibility to generate in-
depth knowledge about the research object. The Case Study is often very useful in
such research. The strategy is then used as an alternative approach together with
other methods.
The purpose of a Case Study is simply to take a small part of a whole course of
events to let the case represent and describe reality (Ejvegård, 1996).
Merriam (1994) argue that all types of methods to gather scientific information can
be used in a Case Study, both qualitative and quantitative ones.
According to Yin (1994 p.9-10) the weaknesses of the Case Study are:
− It is not uncommon that equivocal information and/or biased views are
allowed to influence the conclusions of the Case Study.
− Case studies provide a rather little basis for scientific generalization.
− Case studies can become massive and take too long time, hence being hard to
analyze.
These shortages need to be considered by the authors as they can affect both the
validity and the reliability of the thesis.
2.3. Literature study

As the authors have chosen a deductive approach, which uses theories as a starting
point of the research, there was a need for an extensive literature study.
Since Six Sigma is an approach with many different advocates, the authors decided
to look for information about Six Sigma from a number of authors to get an own
objective point of view. Three books about Six Sigma used frequently by the
authors are: Magnusson, Kroslid & Bergman (2003), Pande, Neuman & Cavangh
(2000) and Pyzdek (2003). The first one displays a Swedish point of view of Six
Sigma while the other two are American.
-15-
-METHODOLOGY-


These books were chosen because Six Sigma was originally an American
"invention" and the Swedish book will give a Swedish perspective of the Six Sigma
concept.
Other books in the quality field that will be referred to are literature about TPM,
TQM and Statistical Process Control, SPC. Among these are Bergman & Klefsjö
(2001), Ljungberg (2000), Dale (1999) and Montgomery (2001).
Used reference books are mainly the online editions of Nationalencyklopedin, but
also Encyclopedia Britannica. The use of two different Encyclopedias will
guarantee that the correct explanation of the search word is used in doubtful cases.
The literature was found by web search in the catalogues, CHANS, at the library at
Chalmers University of Technology, Gothenburg and LUCIA, the library catalogue
at Luleå University of Technology, Luleå. Some references were found from
previous works of the authors and from tips and advices from colleagues and
friends and a few references were also available at the Cage Factory.
2.4. Choice of data collection method
According to Wiedersheim-Paul & Eriksson (1991) and Arbnor & Bjerke (1994)
the data collected in the research can be of two kinds:
− Primary data: This is new information that the research team has to collect.

− Secondary data: This is information, already gathered by someone else, which
can be used in the research.
Wiedersheim-Paul & Eriksson (1991) mean that it is often appropriate to use
secondary data at the beginning of the research because it is easier and cheaper.
According to Yin (1994) data for case studies can come from six sources:
documents, archival records, interviews, direct observation, participant-observation
and physical artifacts.
2.4.1. Primary data
A basic method to find what people are experiencing is to ask them. This can be
done by a standardized interview or by a questionnaire. The possibility to adjust the
questions to a single individual is important. Qualitative interviews of this kind are
called in-depth interviews (Wallén, 1993).
Holme & Solvang (1991) argue that the strength with the qualitative interview is
that the interview situation is similar to an everyday situation or conversation.
-16-
-METHODOLOGY-


In depth interviews were carried through with a TPMG
5
Manager, a TPMG
Coordinator at the Cage Factory, a Quality Assurance Manager and a Professor in
Quality Engineering at Chalmers University of Technology. The purpose of these
interviews was to get information about the existing TPMG approach, a group view
of Six Sigma and its future in SKF and to get an external point of view about Six
Sigma, which also will secure the validity of the thesis (see Section 2.6.1).
An interview was also made with a Master Black Belt at Volvo Cars Engine in
Skövde. This was done in order to benchmark the experiences of their Six Sigma
venture.
According to Holme & Solvang (1991) a group interview or group discussion is

similar to any other form of interview. The basic difference is the social dimension
added between the participants.
The authors have also collected primary data by putting together a focus group with
four white-collar workers in the Cage Factory. The discussed topic was: "how can
Six Sigma be implemented in the Cage Factory?" The evaluation of the DMAIC
project together with the Improvement Group was done in a similar manner.
The interviews and focus group were semi-structured since the authors prepared
questions or topics for discussion in advance. The questions were of a more open
nature to invite the interviewee to respond with his or her own words.
Primary data in the DMAIC project was gathered by the use of selected tools at
different stages in the DMAIC methodology. Such tools are Process Mapping,
Pareto Charts, Cause-and-Effect Diagram, Check Sheets, Process FMEA and Tree
Diagrams, described closer in Chapters 3 and 4. These tools were used in the
improvement group assembled to deal with the quality shortages in the turning
device. Beside the two authors, the group consisted of two operators, a production
technician, a tool technician and a quality engineer. The authors prepared the use of
these tools by studying them in advance and then acting as supervisors as well as
team members. The authors mapped the turning process by directly observing the
process.
The authors also performed a measurement of defective cages after the turning
activity in Production Channel 13 during a three-week period. This was done by
designing a form that enabled the operators to specify the number of defects
(rework or rejected) and the type of defects for the different types of cages.


5
The authors have decided to abbreviate Total Process Management as TPMG (TPM Gothenburg) to
distinguish between Total Productive Maintenance and the local TPMG at SKF.
-17-