application of measurement system analysis at the abc...
TRANSCRIPT
Application of Measurement System Analysis at the ABC
Company
Mohammad Khanjani
Submitted to the Institute of the Graduate Studies and Research
In partial fulfillment of the requirements for the Degree of
Master of Science in
Industrial Engineering
Eastern Mediterranean University February 2009
Gazimağusa, North Cyprus
iv
Approval of the Institute of Research and Graduate Studies ______________________________ Prof. Dr. Elvan YILMAZ Director (a) I certify that this thesis satisfies all the requirements as a thesis for the degree of Master of Science in Electrical and Electronics Engineering ______________________________ Asst. Prof. Dr. Gökhan İZBIRAK Chair, Department of Industrial Engineering We certify that we have read this thesis and that in our opinion it is fully adequate in scope and quality as a thesis for the degree of Master of Master of Science in Industrial Engineering. ______________________________ Asst. Prof. Dr. Gökhan İZBIRAK Supervisor
Examining Committee _____________________________________________________________________ 1. Prof. Dr. Alagar RANGAN _____________________________ 2. Assoc. Prof. Dr. Bela VIZVARI _____________________________ 3. Asst. Prof. Dr. Gökhan İZBIRAK ____________________________
iii
ABSTRACT
Application of Measurement System Analysis at the ABC
Company
Mohammad Khanjani
M.S in Industrial Engineering Supervisor: Asst. Prof. Dr. Gökhan İZBIRAK
February 2009.
Keywords: Advanced Product quality Planning (APQP), Supply chain quality Management (SCQM), Measurement system analysis (MSA), Repeatability and Reproducibility of the Gage (GR&R), Analysis of the Variance (ANOVA)
One of the common technical design principles of management systems which
defined as Advanced Product Quality Planning (APQP) model is employed in the
suppliers of automotive industry.
Measurement plays a significant role in quality control and usually the gage study
needs to be conducted prior to any measurement for quality control. In this regard, to take
advantage of APQP as a management system other tools such as Measurement System
Analysis (MSA) should be utilized during stages of product realization.
This study has used application of repeatability, reproducibility and stability
methods to show that the current quality planning in a designer and manufacturer of gas
turbine blades needs to be improved by the implementation of APQP requirements.
iv
ÖZET
ABC Firmasında Ölçüm Sistemi Çözümleme Uygulaması
Yönetim sistemlerinin teknik tasarım prensiplerinden biri olarak bilinen İleri Ürün
Kalite Planlaması (APQP), otomotiv endüstrisinde faaliyet gösteren şirketler tarafından
kullanılmaktadır.
Kalite kontrolünde, ölçüm çok büyük önem taşımaktadır. Kalite kontrolü
yapılmadan önce ölçü ayarlama çalışmasının yapılması gerekmektedir. Bu bakımdan,
yönetim sistemi olarak APQP’den yararlanabilmek için Ölçüm Sistemi Çözümleme
(MSA) gibi diğer araçlar da ürün gerçekleştirme aşamaları boyunca kullanılmalıdır.
Bu çalışma, tekrarlanabilirlik, yeniden üretilebilirlik ve stabilite yöntemlerini
kullanarak, gaz türbin kesicilerinin tasarımcıları ve üreticileri tarafından kullanılan genel
kalite planlamasının, APQP kullanılarak geliştirilmesi gerekliliği gösterilmeye
çalışılmıştır.
v
ACKNOWLEDGEMENT
I would like to thank my supervisor Asst. Prof. Dr. Gökhan İZBIRAK for his
supportive and keen collaboration in this subject and for useful comments on the
structure of my thesis. My special thanks to Assoc. Prof. Dr. Bela VIZVARI for his
valuable hints and reminders on the Supply Chain Quality Management and for using his
precious times to read this thesis and gave his critical comments about problem statement
and for affording his time.
I gratefully thank Prof. Dr. Rangan ALAGAR for his valuable advice in science
discussion especially in normality assumption.
I gratefully thank Ass. Prof. Dr. Majid HASHEMIPOUR for providing good
facilities to start and study in the Eastern Mediterranean University. I would also
acknowledge Naimeh, Vahid, Ehsan, Amir, Mohammad and Nima for their help.
vi
To my mother Sakineh,
To the memory of my father Hasan (1931-2007)
To my patient & devoted wife
Shokouh
To my children
Zahra(Sara)
Fatemeh(Sima)
Mohammad Sadegh
Sana
vii
TABLE OF CONTENTS
ABSTRACT ....................................................................................................................... iii
ÖZET ..................................................................................................................................iv
ACKNOWLEDGEMENT ................................................................................................... v
LIST of FIGURES ..............................................................................................................ix
LIST of TABLES ................................................................................................................xi
LIST OF ABBREVIATIONS AND SYMBOLS ............................................................ xiii
1. INTRODUCTION ........................................................................................................... 1
2. QUALITY IN THE SUPPLY CHAIN MANAGEMENT .............................................. 2
2.1 Quality Management System .................................................................................... 4
2.2 Advanced Product Quality Planning ......................................................................... 5
2.3 Six Sigma and APQP ................................................................................................ 9
3. MEASURMENT SYSTEM ANALYSIS ...................................................................... 11
3.1 Variation: Common and special causes .................................................................. 12
3.2 The process improvement cycle.............................................................................. 13
3.3 Quality of measurement data .................................................................................. 14
3.4 Capability index ...................................................................................................... 15
3.5 Repeatability and Reproducibility .......................................................................... 16
3.5.1 Range & Average Method ............................................................................... 18
3.5.2 Analysis of the Variance Method application in MSA .................................... 22
3.6 Stability ................................................................................................................... 25
3.6.1 Gage capability index ...................................................................................... 27
3.7 Relationships between capability of the manufacturing process and
viii
measurement system errors ................................................................................... 27
3.8 Advanced Product Quality Planning and Measurement System Analysis ............. 29
4. SYSTEM IDENTIFICATION AND PROBLEM STATEMENT ................................ 31
4.1 Overview of the XYZ and ABC company profiles ................................................ 31
4.2 Problem definition .................................................................................................. 32
4.2.1 Quality management system challenges ........................................................ 33
4.2.1.1 Maintenance of the Certificate .................................................................. 34
4.2.1.2 Process and integration ............................................................................. 40
4.2.1.3 Process planning and reliability ................................................................ 42
4.2.2 Technical problem statement ........................................................................... 39
5. CONCLUSION .............................................................................................................. 50
REFERENCES .................................................................................................................. 53
APPENDIX A: ISO 9001 AND ISO 14001 .................................................................... 536
APPENDIX B: VALUE OF 2d ......................................................................................... 67
APPENDIX C: STABILITY STUDY AND RESULTS ................................................... 68
APPENDIX D: NUMERICAL RESULTS AND DATA FORMAT ................................ 70
APPENDIX E: GAGE R&R STUDY-ANOVA MEHTHOD........................................... 72
APPENDIX F: XBAR AND R CHART ........................................................................... 74
APPENDIX G: NORMALITY TESTING BY ARENA .................................................. 76
APPENDIX H: FMEA INTERRELATIONSHIPS ........................................................ 767
APPENDIX I: PROBLEM SOLVING METHOD ......................................................... 80
ix
LIST OF FIGURES
Figure 1.1: The concept of degree in quality definition (David Hoyl, 2005a) ................... 3
Figure 1.2: Distribution of critical dimensions for transmissions....................................... 4
Figure 1.2: Internal supply chain (David Hoyl, 2005a) ..................................................... 5
Figure 2.1: SCQM evolution (Carol J. Robinson, et al., 2004) ....................................... 3
Figure 2.2: APQP model (AIAG, APQP manual, 1998) .................................................... 6
Figure 2.3: Considering the APQP model with the IMS (M. Bobrek, et al. 2005) ............ 9
Figure 3.1: Process improvement cycle (AIAG, SPC, 1995) .......................................... 13
Figure 3.2: Relationships between precision and accuracy (MSA, 2002) ....................... 15
Figure 3.3: Process capability (Stefan steiner, et al., 2007) ............................................. 16
Figure 3.4: Repeatability (MSA Third Edition, 2002) ...................................................... 17
Figure 3.5: Reproducibility ............................................................................................... 18
Figure 3.6: the effect of Gage variation on the process capability ................................... 28
Figure 3.7: The effect of the measurement on the results (MSA, 1998) .......................... 28
Figure 4.1: Process subsequence of ABC ......................................................................... 37
Figure 4.4: Operator * Part interaction for D3 .................................................................. 45
Figure 4.5: Operator * Part interaction for D4 .................................................................. 46
Figure 4.6: Individual readings by operators for D1........................................................ 47
Figure 4.7: Individual readings by operators for D2........................................................ 47
Figure 4.8: Individual readings by operators for D3........................................................ 48
Figure A.1: General ISO 9001 model ............................................................................... 58
Figure A.2: General model of ISO 14001 ......................................................................... 64
x
Figure A.3: Process definition ......................................................................................... 65
Figure A.4: Process concept of AIAG point of view ....................................................... 66
Figure C.1: Six pack report of master D2 ........................................................................ 68
Figure C.2: Six pack report of master D4 ......................................................................... 69
Figure F.1: Xbar and R chart on D1 ................................................................................. 74
Figure F.2: Xbar and R chart on D2 ................................................................................. 74
Figure F.3: Xbar and R chart on D3 ................................................................................. 75
Figure F.4: Xbar and R chart on D4 ................................................................................. 75
Figure H.1: FMEA interrelationships, D. H. Stamatis (2003) .......................................... 78
Figure I.1: 8D procedure, Bern-Areno et al. (2007) ......................................................... 81
xi
LIST OF TABLES
Table 1.1: SCM in Dell and Wal-Mart (Jacobs, 2003) ....................................................... 2
Table 3.1: Acceptance levels of GR&R (MSA, 2002) ..................................................... 21
Table 3.2: Two way effect ANOVA model (MSA third edition, 2002) ........................... 23
Table 3.3: Indices of control chart limits .......................................................................... 26
Table 3.4: Relationship between Cp & %GR&R .............................................................. 29
Table 4.1: Categorization of company structure (PMBOK guide, third edition) ............. 33
Table 4.2: Product design process indicators in ABC ...................................................... 36
Table 4.5: The results of the average and range method .................................................. 42
Table 4.6: ANOVA results for D1, fixed effects model ................................................... 43
Table 4.7: ANOVA results for D2, fixed effects model .................................................. 44
Table 4.9: ANOVA results for D4, fixed effects model ................................................... 46
Table 4.10: Capability results on the D2 and D4 .............................................................. 49
Table B.1: Value of 2d ...................................................................................................... 67
Table C. 2: Stability results for D2 ................................................................................... 68
Table C. 3: Stability results for D4 ................................................................................... 69
Table D.1: Numerical result of 4 characteristics ............................................................. 70
Table D.2: Data format in Minitab ................................................................................... 71
Table E.1: ANOVA results for D1, fixed effects model ................................................... 72
Table E.2: ANOVA results for D2, fixed effects model ................................................... 72
Table E.3: ANOVA results for D3, fixed effects model ................................................... 73
Table E.4: ANOVA results for D4, fixed effects model ................................................... 73
Table G.1: output of Arena analyzer on D2 ...................................................................... 76
xii
Table G.2: output of Arena analyzer on D4 ...................................................................... 76
Table H.1: MSA plan sample............................................................................................ 79
xiii
LIST OF ABBREVIATIONS AND SYMBOLS
2A : Constants base on subgroup size
AIAG : Automotive International Action Group
ANOVA : Analysis of variance method
APQP : Advanced Product Quality Planning
ASC : Automotive Supply Chain
ATO : Assemble-to-Order
CB : Certification Body
2d : Estimated with Z and W
4D : Constants base on subgroup size
3D : Constants base on subgroup size
EFQM : European Foundation for Quality Management
8D : Eight Discipline method
FMEA : Failure Mode and Effective Analysis
GR&R : Repeatability and Reproducibility of Gage
I : Interaction between the appraisers and the parts
IATF : International Automotive Task Force
IMS : Integrated Management System
IQNET : International Quality Network
ISO : International Standard for Organization
ISO 9001 : The international Organization for Standardization, Quality Management
ISO 14001: The international Organization for Standardization, Environment
management system
RLCL : Lower Control Limit of Range
XLCL : Lower Control Limit
MSA : Measurement System Analysis
MSB : Mean square of parts
MSE : Mean square of Gage
MSAB : Mean square of interaction between Appraisers and Parts
xiv
MTS : Make-to-Stock
ndc : Number of Data Category
OEM : Original Equipment Manufacturer
OHSAS: Occupational Health and Safety Assessment
R : Mean of ranges
pR : Difference between the largest average part measurement and the smallest
SCM : Supply chain Management
SCQM : Supply Chain Quality Management
SSA : Sum of square of Appraisers
SSAB : Sum of square of interaction between Appraisers and Parts
SSE : Sum of square of Gage
SPC : Statistical Process Control
TQM : Total Quality Management
TV : Total variability, measurement system variability and part variation
RUCL : Upper Control Limit of Range
XUCL : Upper Control Limit
PV : Part variation
W : The number of trials
WBS : Work Breakdown Structure
X : Overall mean
rangeX : Average of the difference in the average measurements between the appraiser
with the highest average measurements, and the appraiser with the lowest
average measurements, for all appraisers and parts
Z : The number of parts times the number of appraisers
1
CHAPTER 1
INTRODUCTION
In the 1980s, acute global competition forced business organizations to offer
high quality products at low cost. As competition in the 1990s intensified further,
manufacturing organizations began to realize the potential benefits and importance of
strategic and cooperative buyer supplier relationships (Tan, K.-C., Kannan. et. al.,
1999).
The definition of supply chain is varying from organization to organization, and
within an organization, from person to person. The American Production and Inventory
Control Society (APICS, 2001), which known as the association for operation
management, defines supply chain in following way: “The global network used to
deliver products and services from raw material to end customers through an engineered
flow of information, physical distribution, and cash.” As seen in the APICS definition,
physical, information, and financial flow are dimensions of the supply chain. The
viewpoint, a very common one, of supply chain as only physical distribution is too
limiting. As a summary definition, Supply Chain Management (SCM) is: Design,
Maintenance, and operation of supply chain processes, including those that compensate
extended product features, for satisfaction of end-user needs (James B. Ayers, 2001).
2
Quality is determined by the extent to which a product or services successfully
serve the purpose of the user during usage (not just at the point of sale). It means that,
price and delivery are both temporary features, whereas the impact of quality is
maintained long after the utilization (David Hoyle, 2005a).
The word quality has many meanings:
• A degree of excellence
• Conformance with requirements
• The totality of characteristics of an entity that bear on its ability to satisfy stated
or implied needs.
• Fitness for use
• Fitness for purpose
• Freedom from defects imperfections or contamination
• Delighting customers
The fundamental and vocabulary of quality management system (ISO 9000) defines
quality as “the degree to which a set of fundamental characteristics fulfils the
requirements” (the former definitions focused on an object that was described as
product or services). With this new definition, quality is relative to what something
should be and what is it. The something may be a product, service, document,
information or any output from a process (David Hoyle, 2005a).
The output is expressed as its characteristic(s). Obviously, to judge the quality of
any output we need to measure it and a basis for comparison is also needed. The
concept of “degree” is illustrated in Figure 1.1. The diagram expresses three facts:
• Needs, requirements and expectations are constantly changing
3
• To go at the same rate with the needs, performance to be constantly
changing.
• Quality is variation between the standard required and standard reached.
This means that all of the related techniques, methodology and tools in the
field of Quality Management utilized for one purpose, that of enabling
organization to close the gap between the standard required and the standard
reached. Therefore, environmental, safety and health problem are quality
problem (David Hoyle, 2005a).
Figure 1.1: The concept of degree in quality definition (David Hoyl, 2005a)
David Hoyle, (2005a) stated that the final judger of quality is the customer. The
customer either provides feedback directly or by loss in sales, reduction in market share
and, finally, loss of business. The customer could be defined as an organization that
receives a product or service includes: purchaser, consumer, client, end user, retailer, or
The performance level
The need, requirements or expectation
Time
Standard
the degree to which a set of fundamental characteristics fulfils a need or expectation that is stated, generally implied or obligatory
4
beneficiary.
Montgomery (2005a) has defined quality term as a proportion to variability. This
definition implies that if variability in the key characteristic of a product decrease, the
quality of the product increases.
As an example of this definition, a few years ago, one of the automobile
companies in the United State performed an empirical study of a transmission which was
manufactured in a domestic plant and by a Japanese supplier. The warranty claim and
repair costs indicated that there was an obvious difference between the two sources of
production, with the Japanese-produced transmission having much lower costs. To
discover the causes, the random samples of transmissions have been selected from each
plant, disassembled them, and measured several key quality characteristics. Figure 1.2 is
generally representative of the results of this study. Note that the first graph (i.e. United
State) takes up about 75% of the specification width while the second graph covers
about 25% of the specification band.
USLLSL
Target
Japan
United
States
Figure 1.2: Distribution of critical dimensions for transmissions.
5
Although, customer in ISO 9000, is considered as internal and external, some
authors e.g. David Hoyle believe that a customer is a stakeholder. But the internal one,
who receives a product are not stakeholder. For example, where an employee receives a
technical drawing from a designer the employee could be regarded as a customer and
she/he does not pay anything to the designer and there is no any contract between them.
Figure 1.2: Internal supply chain (David Hoyl, 2005a)
The notion of internal and external customer illustrated in Figure 1.2 in the upper
diagram requirements are passed through the supply chain and if at each stage there is
some interpretation by the time the last person in the chain receives the documents, they
may be very much different from what the customer originally wanted.
In the reality of extreme global competition, SCM principles are turning to the
business excellence models (Total Quality Management). Highly business companies
External customer
What we think the customer ordered
Customer Customer Customer
Supplier Supplier Supplier
Inside the organization
Customer
Supplier
External customer
Exactly what the customer ordered
Customer Customer Customer
Supplier Supplier Supplier
Customer
Supplier
Calibration of requirements
Inside the organization
2
such as Wal-Mart and Dell Computer (Table 1.1) have integrated their supply chains to
make efficient use of information and technologies while harmonizing all activities of
the chain (Lee, 2000) ; (Kinsella, 2003); (Carol J. Robinson et al., 2004).
Dell computer
Wal-Mart
Inventory management
Dell manufactures more than 50,000 computers every day, but carries only four days of inventory (competition carries 20–30 days)
Wal-Mart uses cross-docking and hub-and spoke distributions centers to eliminate unnecessary handling and storage of product while targeting a large geographical area.
Supplier management
Production management
Only about 30 Dell suppliers provide 75% of direct material purchased. If supplier levels exceed 10 days, Dell works with the supplier to lower inventory. Dell took a Make-To-Stock (MTS) industry and shifted it to Assemble-To-Order (ATO). Orders are pulled through manufacturing based on actual orders.
Wal-Mart gives better payment terms to suppliers for their use of electronic ordering and information sharing between Wal-Mart and the supplier. (e.g. Proctor & Gamble).
Wal-Mart initiated the practice of ‘‘everyday low prices’’ in which there’s no need for weekly sales or special promotions (now almost standard in the retail industry).
Information management
More than 50,000 orders come through the Internet. Dell’s legacy order management System records all the orders and releases them to manufacturing. Production lines are scheduled every two hours.
Wal-Mart launched its own satellite creating a communication network to monitor orders and shipments with all stores and suppliers ensuring the quality of data.
Technology management
Technology in Dell’s supply chain process provides efficiencies, immediate communication with suppliers and improved operations internally.
Wal-Mart issued a RFID technology mandate to the top 100 suppliers by 2005 (Wal-Mart technology standard).
Quality management
To address quality issues Dell launched the Critical Supplier Partnership Program resulting in improvement in quality metrics and continuity of supply. This program reduced early field failures by 37% and manufacturing line failures fell from 15,000 to 3000 defective parts per million (dppm).
Wal-Mart achieves a very high degree of quality with respect to loading pallets and merchandise in correct condition on its trucks that accurately match the bill of lading. High quality procedures minimize loss or damage during material handling within the warehouses and during transportation.
Table 1.1: SCM in Dell and Wal-Mart (Jacobs, 2003)
As stated before, these two companies are well known as pioneers of SCM. The
performance indicators of Dell show that this company manufactures more than 50,000
computers every day, but carries only four days of inventory (competition carries 20–30
days). From quality issues point of view Dell launched the Critical Supplier Partnership
3
Program resulting in improvement in quality metrics and continuity of supply. This
program reduced early field failures by 37% and manufacturing line failures fell from
15,000 to 3000 defective parts per million (dppm).
Analogously, SCM is the most developed in the automotive industry (Krisztina
et al. 2006). The automotive industry is the biggest industry in the world and constantly
changing. Over 8 million people working for 50 manufactures produced over 60 millions
vehicles in 2003 with production rising by 6% by mid-2004 (David Hoyle. 2005b). Due
to its global nature, OEM’s (Original Equipment Manufacturer) in automotive supply
chain (ASC), widely use techniques for control and improvement of the suppliers (such
as: Advanced Product Quality Planning (APQP), Measurement System Analysis
(MSA), Statistical Process Control (SPC), Failure Mode and Effective Analysis
(FMEA), which published by Automotive International Action Group (known as AIAG,
American Automotive association).
In AIAG, statistical tools, control of process variation is taken into account by (a)
using of Plan, Do, Study, Action philosophy (see Chapter 3, Fig. 3.1) not only for control
of special causes but also for improvement via reducing of common variation and (b)
defining and monitoring of related indicators such as Gage R&R (repeatability; the
variation observed when the same operator measures the same part repeatedly with the
same device and reproducibility of gage; the variation observed when different operators
measure the same parts using the same device.), gC & gkC (capability index of the gage
which in SPC case are applied as pC & pkC ) and etc..
The purpose of this thesis is to find a solution for improvement of the controls
among the product realization in designer and manufacturer of the gas turbine blades
4
with applying the two Automotive Supply Chain tools which are known as Measurement
System Analysis and Advanced Product Quality Planning. This idea is taken from works
of Carol J. Robinson et al. (2004) in which illustrate the Supply Chain Quality
Management characteristics and works of M. Bobrek et al. (2005) in which investigate
the APQP model in other sectors as a basic concept for designing and implementation of
Integrated Management System (IMS). Our aim is to focus on the engineering and
designing processes of the ABC company which should be taken in consideration with
existing quality and environmental management system (ISO 9001, ISO 14001). We
also concentrate on presenting the variation of quality control with applying
Measurement System Analysis (MSA). Further details on the subject will be given in
relevant sections throughout the dissertation.
The dissertation is organized as follows: In Chapter 2 a review of Quality and
Quality System in the Supply Chain Management is presented. Chapter 3 describes one
of the automotive supply chain tools which is known as MSA. Chapters 4 provide
problem statement and application of the MSA technique in quality control department
of the ABC Company. Conclusion and further research plans are also presented in
Chapter 5.
2
CHAPTER 2
QUALITY IN THE SUPPLY CHAIN MANAGEMENT
Developing and maintaining strong relationships between firms and their
suppliers, as well as among suppliers at different layers (first tier is a supplier which
provides main product(s) directly to the OEM, second tier is a supplier for the first tier
and so on) of the supply chain, has become an important strategic issue. Many have
suggested that supply chain management can lead to faster product development,
decreased production lead-times, reduced cost, and increased quality (Choi. T. Y, 1999).
For more than three decades organizations have been dominated by quality
management and improvement. Tan et al. (1999) conducted a survey on quality directors
and vice-presidents from a broad range of industries, and concluded that successful
management and well-defined linkages between Total Quality Management (TQM)
practices and performance is the key to long-term success. In addition, they concluded
that many strategic quality approaches and Supply Chain Management tools are
positively correlated with firm performance. Their results show that quality management
and supply based management techniques and tools must be implemented conjointly to
achieve superior financial and business performance.
Some of the related questions which supported this idea are:
- Training in basic statistical techniques such as histograms and control charts.
3
- Training in advanced statistical techniques (design of experiments and
regression).
- Quality awareness provided to managers and supervisors.
- Development of procedures for monitoring key indicators of competitor and
customer satisfaction performance.
- Quality department plays an active role in providing specific training such as
SPC.
Carol J. Robinson, et al. (2004) conducted a comprehensive review in quality
management system among SCM and provided a new definition for it as Supply Chain
Quality Management (SCQM). The evolution of SCQM is illustrated in Fig. 2.1.
Meanwhile, they logically categorized their results into the themes of (1) communication
and partnership activities, (2) process integration and management, (3) management and
leadership, (4) strategy, and (5) best practices.
Figure 2.1: SCQM evolution (Carol J. Robinson, et al., 2004)
-Acceptance Sampling - Control charts - Statistical quality Control - Inspection
-Zero defects - Program solving - Quality circles - SPC - DOE
-TQM - ISO9001 - Baldrige Award - Six-sigma
Supply Chain Management
Supply Chain Quality Management (SCQM)
1920-1960 Internal Organization
1960-1980 Internal Organization
1980-1990 - Supply-base - Organization - Customer exception
1990-present All supply channel members and mostly internal organization
2004-present All supply channel members and mostly external organization
Programs
Years
Focus
4
Carol J. Robinson, et al. (2004) believe that the traditional quality programs
focusing on approaches such as TQM, the Malcolm Baldrige National Quality Award
(MBNQA) and ISO 9001 (international quality management system standard), must now
transform to a supply chain perspective in order to simultaneously make use of supply
chain partner relationships and quality improvement gains.
In order to better illustrating the SCQM themes a case study of a firm that is a
first-tier supplier in an offshoot of automotive supply chain is presented by the authors.
Information for this case study was gathered during ISO 9001:2000 pre-assessment
auditing and a detailed structured interview with the Assistant Vice President of the
firm.
2.1 Quality Management System
A system is an ordered set of ideas, principles and theories or a chain of
operations that produce specific results, and to be a chain of operations they need to
work together in a regular relationship (David Hoyle, 2005a). Deming (1989) defined a
system as a series of functions or activities within an organization that work together for
the aim of the organization. These two definitions appear to be consistent although
worded differently. In fact, a quality management system (QMS) is not a random
collection of procedures, tasks or documents (which many quality systems are). Quality
management systems are like air-conditions systems- they need to be designed (David
Hoyle, 2005a).
To date, over half a million organizations in over 150 countries have achieved
quality registration through ISO standards. Over 50,000 companies within the United
5
States alone have obtained the new ISO 9000:2000 registration (IQNet, 2006).
Although, for many firms, obtaining acceptable levels of quality starts with the
registration of a QMS for itself and its suppliers to ISO 9001 and the standard is still
subject to controversy for individual firms and supply chains, a widespread criticism of
the program is that it is not connected directly enough to product quality. For example, a
registered company can still have substandard processes and products because
registration does not tell a company how to design more efficient and reliable products
(Robert Sroufe et al., 2007). In fact, Quality Assurance registration does not necessarily
ensure product quality, but gives guidance on the implementation of the systems needed
to trace and control quality problems.
From reliability (how often does the product fail?) point of view, J. D. Booker et
al. (2001) argued that, reliability prediction will remain a controversial technique until
the statistical methods for quantifying design parameter becomes embedded in everyday
engineering.
Attention needs to be focused on the quality and reliability of the design as early as
possible in the product process development. A lack of understanding of variability in
manufacturing and service conditions at the design stage is a major contributor to poor
product quality and reliability. In order to analyzing and communicating reliability
problem, various well-known tools and techniques exist, For instance Quality Functional
Deployment (QFD), Failure Mode and Effects analysis (FMEA) and Design of
Experiment (DOE).
2.2 Advanced Product Quality Planning
Product quality planning is a structured method of defining and establishing the
6
steps necessary to assure that a product meets the expectations of the customer. The goal
of product quality planning is to facilitate communication with everyone involved to
assure that all required steps are completed in time (M. Bobrek a. et al., 2005). Later this
method has been developed in automotive industry with all necessary details and
processes. This logic is known as Advanced Product Quality Planning (see Figure 2.2).
Figure 2.2: APQP model (AIAG, APQP manual, 1998)
The main principles of implementing APQP plan are:
(a) Organize the team: the supplier’s first step in product quality planning is to define
responsibility to a cross functional team. Effective product quality planning requires the
involvement of more than just the quality department.
The initial team should contain representatives from engineering, manufacturing,
material control, purchasing, quality, sales, service, subcontractors and customers, as
appropriate.
7
(b) Define the scope: it is important for the product quality planning team in the earliest
stage of the product program to identify and clarify customer needs, expectations and
requirements.
(c) Team-to-team: the product quality planning team must establish lines of
communication with other customer and supplier teams. This may include regular
meetings with other teams. The extent of team-to-team (i.e. cross functional team) is
dependent upon the number of issues requiring a solution.
(d) Training: the success of a product quality plan is dependent upon an effective
training program that communicates all the requirements and development skills to
fulfill customer needs and expectations.
(e) Customer and supplier involvement: the primary customer may begin the quality
planning process with a supplier.
(f) Parallel engineering: it is a process where cross functional teams attempt for a
common goal. It replaces the sequential series of phases where results are forwarded to
the next area for execution. The purpose is to accelerate the introduction of quality
products sooner.
(g) Control plans: control plans are written descriptions of the systems for controlling
parts and processes. A separate control plan covers three distinct phases: prototype, pre-
launch and production.
(h) Concern resolution: during the planning process, the team will face the product
design and/or processing concerns. These concerns should be documented on a matrix
with assigned responsibility and timing plan. Disciplined problem-solving methods
(such as, 8D) are recommended in difficult situations.
(i) Product quality timing plan: the product quality planning team’s first task should be
8
the defining and development of a timing plan. The type of product, complexity and
customer expectations should be considered in selecting the timing elements that must
be planned and charted.
(j) Plans relative to the timing chart: the success of any program depends on meeting
customer needs and expectations in a timely manner at a cost that stand for value.
Concurrent engineering performed by product and manufacturing engineering activities
working concurrently is the driving force for error prevention.
Most of the application of APQP has been used for production process in the
manufacturing industry by many researches. First significant application of APQP in
integrated management system (IMS) design has been employed by M. Boberk et al.
(2005) (See Figure 2.3). This model as a procedure has been tested on over the thirty
certified quality management systems and two environmental management systems.
Moreover The ISO 9001 and ISO 14001 standards share parallel management
techniques and principles. Both of them require organizations to formulate policies, to
define roles and responsibilities, to appoint management representatives, and to train
personnel (Tan et al., 1999). Implementing both ISO 9001 and ISO 14001 demands
many duplicate management tasks. For instance, both ISO 9001 and ISO 14001 require
documentation control and auditing all working procedures and processes. Therefore,
two separate documentation systems are needed to meet their requirements which
involve a lot of documentation, written procedure, checking, control forms, and other
paper work. Hence, integrated management systems (IMS) have drawn the attentions of
both academics and practitioners (S.X. Zeng et al. 2005).
9
Figure 2.3: Considering the APQP model with the IMS (M. Bobrek, et al. 2005)
2.3 Six Sigma and APQP
Walter Shewhart introduced three sigma as a measurement of output variation in
1922, and argued that action on the process is needed when the output go beyond this
limit. The three sigma concept is related to a process yield of 99.973 percent and stand
for a defect rate of 2,600 per million, which was sufficient for most manufacturing
organizations until the early 1980s (Mahesh S. et al. 2005).
In response to the threat to American manufacturing from the Japanese, several
quality models were introduced starting in the 1980s to assist make domestic production
of goods and services more competitive. In this regard, Motorola found that they were
losing a large quantity of their business and productivity through the cost of non-quality.
This includes not only the 2,600 parts per million losses in manufacturing, but lost
business due to defective parts and support of systems in the field that were unreliable.
10
A Motorola engineer, Bill Smith, found that the quality level associated with a
measure of Six Sigma corresponds to a failure rate of two parts per billion and adopted
this as a standard. The program to achieve this lofty goal was developed at Motorola and
coined “Six Sigma”, which included many of the systematic and rigorous tools
associated with the Six Sigma programs of today (Mahesh S. et al. 2005).
The immediate objective of Six Sigma is defect reduction. Reduced defects lead
to output improvement; higher production rate and improved customer satisfaction. Six
Sigma defect reduction is intended to lead to cost reduction. It has a process focus and
aims to call attention to process improvement opportunities through systematic
measurement (Mahesh S. et al. 2005).
Stamatis (2000) argued that organizational culture needs to put quality into planning and
drive quality throughout the entire organization. He states that Six Sigma reformulates
the quality operating system introduced by Ford Motor Company in the early 1990s. The
APQP method alone, Stamatis believes to be superior to Six Sigma, with the policy of
the organization being the major contributing factor to success in lieu of the Six Sigma
method.
11
CHAPTER 3
MEASURMENT SYSTEM ANALYSIS
As a simple definition, measurement is a process of evaluating an unknown
quantity and expressing it into numbers. The traditional approach to the management of
the measurement process is calibration. In simple terms, calibration is a process of
matching up the measuring instrument scale against standards of known value, and
correcting the difference, if any. Calibration is done under controlled environment and
by specially trained personnel. On the shop floor, where these instruments are used, the
measurement process is affected by the factors like method of measurement; appraiser’s
influence, environment, and method of locating the work piece do induce variation in the
measured value. It is imperative to assess measure and document all the factors affecting
the measurement process, and try to minimize their effect on the measurement (Stamatis,
D.H., 2000)
Due to the fact that all measurements contain error, and in keeping with the basic
mathematical expression: Observed value = True value + Measurement Error,
understanding and managing "measurement error," generally called Measurement
Systems Analysis (MSA), is an extremely important function in process improvement
(Montgomery, 2005a). In the early 1990's, the Automotive Industry Action Group
formalized MSA in the automotive industry with its publication of Measurement
12
Systems Analysis, Reference Manual, now in its Third Edition, finally becoming a
de facto standard of the entire manufacturing industry (AIAG, 1992, 2002).
3.1 Variation: Common and special causes
Actually any process might contain many source of variability, no two products
or characteristic are exactly alike. The differences among products may be large, or they
may be immeasurably small. Any distribution (such as normal distribution) can be
categorized by:
• Location
• Spread
• Shape
Common causes refer to many sources of variation within a process that has a
stable and repeatable distribution over time. This is called in statistical control an in this
case the output of the process is predictable.
Special causes (often called assignable cause) refer to any factors causing
variation that are not always acting on the process. That is, when they occur, they make
the overall distribution change. Furthermore, the changes in the process distribution due
to special causes can either be detrimental or beneficial. When detrimental, they need to
be identified, and removed. When beneficial, they should be identified and made a
permanent part of the process.
13
3.2 The process improvement cycle
In applying the concept of continual improvement to process, SPC manual
propose three stage cycles which illustrated in Figure 3.1.
Figure 3.1: Process improvement cycle (AIAG, SPC, 1995)
Every process subject to improvement can be located somewhere in this cycle.
The objective of first stage is a basic understanding of the process among identifying
common and special causes and achieving a state of statistical control. When a better
understanding of the process has been achieved, the process must be maintained at an
appropriate level of capability. Due to processes are dynamic and will change, the
performance of the process must be monitored with effective measures. To covering the
customer needs which are sensitive to variation within engineering specification,
additional process analysis tools, including more advanced statistical method such as
PLAN DO
STUDY ACT
PLAN DO
STUDY ACT
PLAN DO
STUDY ACT
1- ANALYZE THE PROCESS • What should the process be
doing? • What can go wrong? • What is the process doing? • Achieve a state of statistical
control • Determine capability
2- MAINTAIN THE PROCESS • Monitor process
performance. • Detect special cause variation
and act upon it.
3- Improve the process • Change the process to better
Understand common cause variation.
• Reduce the common cause variation
14
designed for experiment (DOE) and advanced control charts (such as cumulative control
chart) may be useful in the third stage (i.e. improve the process).
3.3 Quality of measurement data
The quality of measurement data is defined by the statistical properties of
multiple measurements obtained from a measurement system operating under stable
conditions. For instance, suppose that a measurement system, operating under stable
conditions, is used to obtain several measurements of a certain characteristic. If the
measurements are all “close” to the master value for the characteristic, then the quality
of the data is said to be “high.” (AIAG, MSA, 2002).
Similarly, if some or all of the measurements are far away from the master value,
then the quality of the data is said to be “low”. The statistical properties most commonly
used to characterize the quality of data are the bias and variance of the measurement
system. The property called accuracy refers to the location of the data relative to a
reference (master) value, and the property called precision refers to the spread of the
data (see Figure 3.2).
It is important to realize that the accuracy (or bias) and precision (or variance)
are independent of each other. Then controlling one of these sources of error does not
guarantee the control of the other. Finally, the next section (3.5 and 3.6) will be focused
primarily on the precision of the gage, not its accuracy. Evaluating the accuracy of a
measurement system often requires the use of a standard, for which the true value of the
measured characteristic is known. Often the accuracy feature of an instrument can be
modified by making adjustment to the instrument (Montgomery, 2005a).
15
Figure 3.2: Relationships between precision and accuracy (MSA, 2002)
3.4 Capability index
A capability index relates the voice of the customer (specification limits) to the
voice of the process (process limits). Many customers ask their suppliers to record
capability index for all special product characteristic (or key characteristic). This
measure reflect any continual improvement afford as well and defined as the ratio of the
distance from the process center to the nearest specification limit divided by a measure
of the process variability. The idea is illustrated graphically in Figure 3.3. The figure
shows a histogram of process output along with the specification limits.
2dR
σ) is obtained from Table 3.4 and X is overall mean of data while USL and LSL
present the specification control limits.
Not Accurate
Accurate
Not Precise Precise
Note: Some current literatures defines accuracy as the lack of bias
16
Fre
qu
ency
f
Figure 3.3: Process capability (Stefan steiner, et al., 2007)
Generally, if pkp CC = , the process is centered at the midpoint of the
specifications and otherwise the process is off-center.
3.5 Repeatability and Reproducibility
Repeatability is the variability of the measurements obtained by one person while
measuring the same item repeatedly. This is also known as the inherent precision of the
measurement equipment (see Figure 3.4).
Reproducibility is the variability of the measurement system caused by
differences in operator behavior. Mathematically, it is the variability of the average
values obtained by several operators while measuring the same item with the same
instrument.
Process variability (3 sigma)
Process mean to nearest
Specification limit
Upp
er specification
lim
it
Low
er specification
lim
it
2
6d
R
p
LSLUSLC
σ)−
=
−−=
22
3,
3min
dR
dR
pk
LSLXXUSLC
σσ ))
17
-4 -2 2 4
0.1
0.2
0.3
0.4
Figure 3.4: Repeatability (MSA Third Edition, 2002)
Figure 3.5 displays the probability density functions of the measurements for
four operators with the same item and same instrument. In this shape the variability of
the individual operators is the same, but because each operator has a different bias, the
total variability of the measurement system is higher when two or three operators are
used than when one operator is used.
Possible causes for poor repeatability include within part (form, position, surface,
taper, sample consistency), within instrument (repair; wear, equipment or fixture, poor
quality or maintenance), within method (variation in setup, technique, holding,
clamping), within appraiser (technique, position, lack of experience), within
environment (short cycle fluctuations in temperature, humidity, vibration, lighting,
cleanliness or wrong gage for the application
Potential sources of reproducibility error include between parts (average
difference when measuring type of parts A, B, C, etc. using the same instrument,
operators, and method), between instruments (average difference using instruments A,
Repeatability
18
B, C, etc., for the same parts, operators and environment), between appraisers (average
difference between appraisers A, B, C, etc., caused by training, technique, skill and
experience. this is the recommended study for product and process qualification and a
manual measuring instrument)
Figure 3.5: Reproducibility
3.5.1 Range & Average Method
The Range & Average Method computes the total measurement system
variability, and allows the total measurement system variability to be separated into
repeatability, reproducibility, and part variation. To quantify repeatability and
reproducibility using average and range method, multiple parts, appraisers, and trials are
required. The recommended method is to use 5 parts, 2 appraisers and 3 trials, for a total
19
of 30 measurements. The measurement system repeatability is:
Repeatability 2
15.5
d
R=
Where R is the average of the ranges for all appraisers and parts, and 2d is found in
Appendix B with Z = the number of parts × the number of appraisers, and W = the
number of trials. For instance, with 5 parts, two appraisers and 3 trial, Z is equal to 10
and W is equal to 3.
The measurement system reproducibility is:
Reproducibility = nr
ityrepeatabil
d
X range2
2
2
15.5−
Where rangeX is the average of the difference in the average measurements between the
appraiser with the highest average measurements, and the appraiser with the lowest
average measurements, for all appraisers and parts. d2 is found in Appendix B with Z =
1 and W = the number of appraisers, n is the number of parts, and r is the number of
trials.
The measurement system repeatability and reproducibility is
GR &R = 22 ilityreproducibityrepeatabil +
20
The part variability is
Where pR is the difference between the largest average part measurement and the
smallest average part measurement (then the Z is equal to 1), where the average is taken
for all appraisers and all trials, and 2d is found in Appendix B with Z = 1 and W = the
number of parts.
The total variability, measurement system variability and part variation combined is
22& pVRRTV +=
The percentage of measurement system repeatability and reproducibility is
%GR&R = 100
TV
RGR&
%GR&R ratio is used to evaluate whether a measurement system or gauge is
able to properly measure the quality characteristics of a product. AIAG, MSA (2002)
indicates that if %GR&R ratio is less than 10%, then the measurement system is
considered to be acceptable; if %GR&R ratio is larger than 30%, then the measurement
system is not acceptable and should be improved. When %GR&R ratio is laid in 10–
30%, the acceptance of measurement system depends on higher authorities in the
companies. When %GR&R ratio lie in between 10% and 30%, many companies
consider the measurement system is barely acceptable (Jeh-Nan Pan, 2006). Table 3.2
2
15.5
d
RV
p
p =
21
summarize the stated above.
GR&R Percentage Measurement system Less than 10% Acceptable
10% to 30% May be acceptable based on importance
Of application, gage cost, etc.
More than 30% Unacceptable-measurement System needs improvement
Table 3.1: Acceptance levels of GR&R (MSA, 2002)
Note that, the utilization of the range and average method, x chart shows the
ability of the gage to distinguish between units of product (Montgomery, 2005b).
Meaning, more than 50% of the readings should be lying outside of the control limits
(AIAG, MSA, 2002). If the data show this pattern, then the measurement system should
be adequate to detect part-to-part variation. The range chart can assist in determining:
• Statistical control with respect to repeatability
• Consistency of the measurement process between appraisers for each part.
The final step in the numerical analysis is to determine the number of distinct
categories that can be reliably distinguished by the measurement system. This is the
number of non-overlapping 97% confidence intervals that will span the expected product
variation. Note that the ndc is truncated to integer.
ndc = 1.41
RGR
Vp
&
If the instrument lacks discrimination (sensitivity or effective resolution) it may
not be an appropriate instrument to identify the process variation or quantify individual
22
part characteristic values. A general rule of thumb is the measuring instrument of
discrimination that is to be at least one-tenth of the range to be measured. Traditionally
this range has been taken to be the product specification. Recently the 10 to 1 rule is
being interpreted to mean that the measuring equipment is able to discriminate to at least
one-tenth of the process variation. For instance, a 1mm resolution ruler cannot be used
to satisfy the rule of thumb for a measurement design with 5mm+1/5mm-
1specification.instead an instrument with lesser resolution should be used (with .01).
Furthermore, from number of data category point of view (ndc) if the result is 5 or more,
the measurement system is interpreted as good, between 2 and 4 is conditional (i.e. the
final judgment is based on if the characteristic behavior define the safety or regular
requirements if not, it is not accepted), and less than 2 is unacceptable.
3.5.2 Analysis of the Variance Method application in MSA
The Analysis of Variance method (ANOVA) is the most accurate method for
quantifying repeatability and reproducibility. The ANOVA method allows the variability
of the interaction between the appraisers and the parts to be identified. The ANOVA
method for measurement assurance is the same statistical technique used to analyze the
effects of different factors in designed experiments (AIAG, MSA, 2002). The ANOVA
design used is a two-way, fixed effects model with replications. The ANOVA table is
shown in Table 3.2.
According to Tsai’s (1989) ANOVA model, it is a two-factor design of
experiment under the same condition of measurement, where one factor is the inspector,
the other factor is the product, and both are random effect. The model is:
23
µ : measurement mean (total mean).
iP : effect of product (random effect).
jO : effect of inspector (random effect).
ijPO)( : effect of interaction between product and inspector (random effect).
ijlR : Effect of replicate measurements (error term).
ijlijjiijl RPOOPy ++++= )(µ
=
=
=
kl
pj
ni
,....,2,1
,...,2,1
,...,2,1
Table 3.2: Two way effect ANOVA model (MSA third edition, 2002)
By using the four expected mean squares in Table 3.2, one can get the estimated
values of these sources of variation, which are shown below:
Source of Variation
Sum of Square
Degrees of
Freedom
Mean Square
F-statistic
Appraiser ASS a-1
)1( −=
a
SSMS A
A
B
A
MS
MSF =
Parts BSS b-1
)1( −=b
SSMS B
B
E
B
MS
MSF =
Interaction (Appraiser,
Parts) ABSS (a-1)(b-1)
)1)(1( −−=
ba
SSMS AB
AB
E
AB
MS
MSF =
Gage ESS ab (n-1)
)1( −=
nab
SSMS E
E
Total TSS abn-1
24
abn
Y
bn
YSS
a
i
iA
2
1
2.. )( •••
=
−=∑
abn
Y
an
YSS
b
j
iB
2
1
2.. )( •••
=
−=∑
BA
b
j
ija
i
AB SSSSabn
Y
n
YSS −−−= •••
==∑∑
2
1
2.
1
)(
abn
YYSS
b
j
n
k
ijk
a
i
T
2
1 1
2
1
•••
= ==
−= ∑∑∑
BAABTE SSSSSSSSSS −−−=
a = number of appraisers,
b = number parts,
n = the number of trials, and
Then the repeatability, reproducibility, and the variability of gage can be calculated
through the following:
EMSityrepeatabil 15.5=
bn
MSMSilityreproducib ABA −= 15.5
The interaction between the appraisers and the parts (i.e. appraiser differences depend on
the part being measured) is
n
MSMSI EAB −= 15.5
25
The measurement system repeatability and reproducibility is
222& IilityreproducibityrepeatabilRR ++=
The measurement system part variation is
an
MSMSV ABAP
−= 15.5
3.6 Stability
Stability is the total variation in the measurements obtained with a measurement
system on the same master or parts when measuring a single characteristic over an
extended time period. That is, stability is the change in bias over time.
Possible causes for instability include:
• Instrument needs calibration with reducing of calibration interval.
• Worn instrument, equipment or fixture.
• Poor maintenance- air, power, filters, rust.
• Worn or damaged master, error in master.
• Poor quality instrument.
Control limits are used to show the extent by which the subgroup averages and
ranges would vary if only common (random) causes of variation were present. The
formulas for the upper control limit (UCL) and lower control limit (LCL) for x and R
charts follows:
26
RDUCLR 4=
RDLCLR 3=
RAXUCL
X 2+=
RAXLCL
X 2−=
The factors 3D , 4D and 2A are constants based on subgroup size n taken from the
Table 3.3. When the size of the subgroup is lower than 6 then, lower control limit of the
range is defined as zero.
n 2 3 4 5 6 7 8 9 10
4D 3.27 2.57 2.28 2.11 2.00 1.92 1.86 1.82 1.78
3D 0 0 0 0 0 0.08 0.14 0.18 0.22
2d 1.13 1.69 2.06 2.33 2.53 2.7 2.85 2.97 3.08
2A 1.88 1.02 0.73 0.58 0.48 0.42 0.37 0.34 0.31
Table 3.3: Indices of control chart limits
Although, assumption in the development of the Xbar and R control charts is that
the primary distribution of the quality characteristic is normal. Several authors have
investigated the effect of departures from normality on control charts. Burr (1967)
comments that the usual normal theory control limits constants are very robust to the
normality assumption and can be employed unless the population is extremely non-
normal.
According to Montgomery 2005a, the role of theory and assumption such as normality
and independence is important to have reliable limits for the sake of monitoring the
27
process performance, but plays a much less important role in the application of Xbar and
R chart which is considered as trial control limits. They allow us to investigate whether
the process was in control when the n initial samples were selected.
3.6.1 Gage capability index
One of the popular ratio(s) which reflects the overall results of the stability study
is gage capability index. This ratio(s) is a numerical summary the behavior of a product
or process characteristic to engineering specification. This measure also often called
capability or performance indices or ratio and known as pC and pkC (AIAG, SPC,
1995).
3.7 Relationships between capability of the manufacturing process and
measurement system errors
As stated before, measurement system is the collection of instruments or gages,
standard, operations, methods, fixtures, software, personnel, environment and
assumptions used to quantify a unit of measure or fix assessment to the feature
characteristic being measured; the complete process used to obtain measurements. For
instance Figure 3.7 is illustrated the effect of the instrument capability (known as
accuracy of instrument- suppose the defined tolerance is equal to 5+0.01 and 5-0.01) on
the control chart. In this case source of variation come from unsuitable instrument (i.e.
the control charts present the false runs which appear as outlier points).
2
6d
R
g
LSLUSLC
σ)−
=
−−=
22
3,
3min
dR
dR
gk
LSLXXUSLC
σσ ))
28
Actual process variation
Observed process variation Production gage variation
Figure 3.6: the effect of Gage variation on the process capability
For the clarification of the relationship between GR&R and manufacturing process
capability index, in the case where the (higher order) measurement system used has a
GR&R of 10% and actual process Cp is 2.0, the observed process Cp will be 1.96. When
this process is studied with GR&R of 30% and the observed process Cp will be 1.71. A
worst case scenario would be if a production gage has not been qualified but is used.
Figure 3.7: The effect of the measurement on the results (MSA, 1998)
Average
Range
Measured to Nearest 0.001 mm Measured to Nearest 0.01 mm
29
If the measurement system GR&R is actually 60% (but that fact is not known)
then the observed Cp would be 1.20. The difference in the observed Cp of 1.96 versus
1.20 is due to different measurement system. Figure 3.6 and Table 3.1 Show the spoil
effect of gage variation on the process capability and summary of the stated results
respectively.
Table 3.4: Relationship between Cp & %GR&R
3.8 Advanced Product Quality Planning and Measurement System
Analysis
Obviously, planning is the key stage before designing and purchase of
measurement equipment or systems. Many decisions made during the planning stage
could affect the direction and selection of measurement equipment. In some cases due to
the risk involved in the component being measured or because of the cost and
complexity of the measurement device the OEM customer (Original Equipment
Manufacturer) may use the APQP process and committee to decide on the measurement
strategy at the supplier.
Before a measuring process request for quotation package can be supplied to a
potential supplier for formal proposals, a detailed engineering concept of the
measurement process needs to be developed. The teams of individuals that will employ
and be responsible for the maintenance and continual improvement of the measurement
process have direct responsibility for developing the detailed concept and this can be
%GR&R Actual process Cp Observed process Cp
10 2.0 1.96 30 2.0 1.71 60 2.0 1.2
30
part of the APQP team.
Furthermore, not all product and process characteristics require measurement
systems. A basic rule of thumb is whether the characteristic being measured on the
component has been identified in the control plan or is important in determining the
acceptance of the product or process.
31
CHAPTER 4
SYSTEM IDENTIFICATION AND PROBLEM STATEMENT
A case from a company (ABC Technology) located in south-eastern region of
Iran is presented in order to illustrate the problem and also to explore the application of
MSA in controlling and managing of quality planning in a better way. The ABC
Technology company is a tier 1 supplier (i.e. main supplier) to XYZ Company. The
company names are changed for the sake of confidentiality.
4.1 Overview of the XYZ and ABC company profiles
XYZ Company was born in 1999 and shares experience and technology with
European company to bring all engineering, manufacturing, contracting and services
activity in the power generation field. The company is based in Tehran and has
manufacturing facilities for twenty production of gas turbine in Karaj and all other
Turbine off base equipment in different places in Iran together with qualified sub-
supplier.
ABC Eng, Co. which is one of the main suppliers of XYZ, by employing skillful
experienced personnel and utilizing qualified laboratories and super alloy vacuum
casting machining workshop, as the first Iranian company achieved manufacturing
know-how of hot gas path components of gas turbines. This firm, located in Tehran/Iran
32
with 70 employees, and became registered to IMS (Integrated Management System–
ISO 9001, ISO 14001 and Occupational Health and Safety Assessment-OHSAS 18001).
For illustration of ABC organizational structure, definition of the project is
needed in advance. According to PMBOK (project management the body of knowledge,
2004) a project is temporary endeavor undertaken to create a unique product, service or
result (temporary means that every project has definite beginning and definite end).
Since this company is responsible for designing and modification of gas turbine blades,
activities of any new or modification in projects are dominated by the product design
requirements (i.e. defining the scope of the projects, making relation between tasks,
reviewing, verification and validation). This is why all of the project managers are
selected from the product design and engineering department. Table 4.1 illustrates the
structure of the ABC company (current cases for the company are underlined). It is
obvious that, the organization structure could easily be categorized as a projectized
structure.
4.2 Problem definition
In the following sectors the problem statement for the ABC Company is defined
as the quality management system challenges and the technical problem statement. In
general problem statement the effectiveness of current quality management system will
be considered whereas in technical problem statement the details of the product design
process will be analyzed.
33
4.2.1 Quality management system challenges
In order to investigate and analyze the existing system in ABC Company, one
should clarify the process and integration in the system. In this case, investigation has to
be done according to the result of external auditing reports by the Certification Body
(CB). Reports covered for three years are always used by the auditing certification
body.
Table 4.1: Categorization of company structure (PMBOK guide, third edition)
Organization
Structure
Project
Characteristic
Functional
Weak Matrix
Balanced Matrix
Strong Matrix
Projectized
Project Manager’s
Authority Little or None Limited
Low to
Moderate
Moderate
To High
High to
Almost Total
Resource
availability Little or None Limited
Low to
Moderate
Moderate
To High
High to
Almost Total
Who controls the
Project budget
Functional
Manager
Functional
Manager Mixed
Project
Manager
Project
Manager
Project Manger’s
Role Part-time Part-time Full-time Full-time Full-time
Project Management
Administrative Staff
Part-time Part-time Part-time Full-time Full-time
34
4.2.1.1 Maintenance of the Certificate
As a global approach, major company (XYZ) wants their partners (ABC
Company who is one of their main suppliers) to implement and register to the ISO 9001,
ISO 14001 and OHSAS 18001 and then, continual improvement of the related process
performance.
In the case of ABC, the company has been registered to ISO 9001 from 2002 and
IMS recently, the history of the related record such as the reports of the certification
body present that the number of non-conformities, still remains to be solved. Obviously,
due to this company relying heavily on engineering and product design activities, most
of the non-conformities are caused by the designing and engineering department.
Moreover any certification body after issuing the certificate should for at least three
successive years, audit the quality management system of that organization (in the
second year the prime attention is not only resolving the former non-conformities, if it
exist, but also improvement of the performance). On the other hand, if any minor non-
conformity, at least for two successive years is repeated, the minor non-conformity will
be considered as a major non-conformity. Following this, the certification is suspended
for three mounts and second auditing is necessary.
Some of the minor non-conformities of this company which have a potential to
be a major non-conformities are listed as bellow:
1) ISO 9001: Requirement 7.3.1 (see Appendix A). In some case, the stages of the
design and development review, verification and validation has not been covered
as consequently.
35
2) ISO 9001: Requirement 4.1 (see Appendix A). Defined measures for engineering
and design process are not sufficient for effective monitoring and measuring of
the quality parameters.
3) ISO 9001: Requirement 8.4 (see Appendix A). In some cases, the data records of
the ABC company have not been utilizing to identifying improvement
opportunity properly.
4) ISO 14001: Requirement 4.3.1 (see Appendix A). In some cases, the ABC
company has not been considering the environmental aspects and related
potential risk requirements in the product planning stage.
In ABC Company, two major departments making judgment about acceptance/
rejection of the products are product design and quality control departments. With
the growing of the company’s production, lack of consensus among the two
departments, have increased dramatically and need for harmonization between the
two departments is obvious. For instance, during the year 2007, two sets of delivered
products to customer has records of 60 parts, which was rejected by the quality
control but accepted after re-measurement by the product design department (this
statement has been obtained by the interview with the quality control and product
design department managers).
4.2.1.2 Process and integration
Since, the ISO 9001 is only a general model quality management system
standard, and there are no any guidelines or suggestions for the process integration and
performance indicator. Table 4.2 and Table 4.3 illustrate product design process and
36
indicators. Obviously, applying two indicators with long period could not satisfy the
needs of an effective controls and subsequently, assurance of process improvement.
As a prime requirement of the ISO 9001, the interaction and sub-sequence of the
process should be defined and implemented by the any clients of the ISO 9001
certificate. In this regard, the ABC company presents these requirements as Figure 4.1.
Indicator Target Graph
type
Reporting
period
Responsible of
calculation
Responsible of
action
Percentage of documents which is
ready 90%
Trend / Bar chart
Two mounts Scheduling
process owner
Project managers and Product design
manager
On time delivery 90% Bar chart Two mounts Scheduling
process owner g
Project managers and Product design
manager
Table 4.2: Product design process indicators in ABC
From Figure 4.1, the approch is to classify into nine defined processes P, D, C, A
cycle. (The cycle starts with Planning and Product design (Plan), which is implemented
and executed by the Marketing, Purchasing, Manufacturing (Do). Control is carried out
by the Processes tooling and Project management (C) and then, Analyzing and
improvement (A). Obviously, it is more general and relations have not been addressed
clearly. The question then arises, with this general approach how is the project stage
monitored and controlled by the project managers?
As a typical method, projects are frequently divided into more manageable
components i.e. sub-projects. In this regard, any project can be broken down into sub-
projects with the application of Work Breakdown Structure (WBS) method. The relation
between projects and sub-projects and activities are then addressed using Microsoft
37
planning
Product design
Marketing
Purchasing
Manufacturing
Financial management
tooling management
project management
analyzing &
improvement
Plan
Do Check
Action
Project software, critical paths are identified and regularly reported. The content of this
report indicates that the prime focus is on the project budget deviation. Then the related
indicators are measured and reported by the scheduling process manager to top manager.
This approch encourage the project managers to concentrate on the on time delivery
indicator and ignoring any request of applying potential risk assessment, reliability study
and other requirements which arise from Integrated Management System (IMS).
Figure 4.1: Process subsequence of ABC
4.2.1.3 Process planning and reliability
For the controlling of the product realization stages the product design
department of ABC Company define the process planning sheet, relevant check lists and
instruction. According to the subjected product model in this study, 14 inspection stages
38
with relevant checklists are requested by the process planning sheet.
Product design process
Process class: Customer
Oriented Process (COP) Process Owner: Engineering and design manager /
Project managers
Objective : producing Customer demands at the lowest cost with the highest reliability
Scope : All of the product
Procedures: Product design document control
Resources: Experts and trained staff / financial resources
Input From
process Output To process
Organizational goals/ management review/
action plans Planning
Purchasing data/ quality specifications / material
certificate / material quality planning
Purchasing
Primary samples / pre lunch /
product certificate Customer
Standards, requirements and needs of customer
and samples Customer
Drawings/ quality plan / acceptance level / instruction
Manufacturing
Result of customer satisfaction and complaints
Customer
Corrective and preventive actions status / measurement
of this process Planning
Non-conformity reports/ statistical analysis /
monitoring and measurement of this process
Analysis and improvement
Budget/Targeting cost Financial
management Request for service Purchasing
Purchased services / qualified suppliers
Purchasing Product part list and manufacturing process
Scheduling
Corrective and preventive actions / improvement projects
Planning Needed training Resource management
FPC, OPC Tooling management
Process Description: Receiving the requirements, data gathering, review of data adequacy, designing planning, designing executive, confirmation of prototype stage, resolving of problems and validation,
preparation of documents and product certificate
Table 4.3: Product design main process of ABC Company
Any measurement (attribute or numerical) should be recorded on this check list sheets.
For instance, in the dimensional control stage (which is before the polishing operation)
17 parameters are defined for measuring (such as dial, height, maximum feeler, twist,
39
thickness and etc.) and after the 100% checking, all of the data are recorded on the
checklists. This form format of checklists lacks the required and considered form
benchmarking control plan as suggested by the AIAG, APQP, 2002:
1- There is no identifying of the key characteristic (identifying of the key or critical
parameters help the control to focus the efforts on the essential demands like
safety, which can give rise to legal issues).
2- There is no any statistical referred method such as SPC and MSA.
3- There is no any reaction plan in the case when the operator will have to face
suspected parts and so on.
From reliability point of view, using of the highest reliability term in the
objective title of the product design process (see Table 4.3), led us to investigate the
application of the statistical tools in the product design stages. Unfortunately, a survey of
the action records indicates that the organization just focus on the specification limits
(USL and LSL). It is obvious this strategy (i.e. action on the output instead of action on
the process as a preventive strategy) followed by action only on the output is a poor
substitute for effective process management.
4.2.2 Technical problem statement
In order to analyze the current quality management system performance at ABC
company, data are collected to appraise the need for rehabilitation and improvement for
the associated product quality planning. The data are exploited and classified by making
use of gage R&R and stability methods (Note that in this case, the gage term is a
combination of a fixture with four micrometers).
40
The following assumptions related with GR&R are:
1. Two different operators, either for different set-ups or for a different time period, use
the gage to obtain replicate measurements on units.
2. In these types of studies, two components of the measurement system variability are
defined as repeatability and reproducibility.
3. Obtaining a 5 sample parts that represent the actual or expected range of process
variation. (i.e., sample parts have not been chosen in a flash from manufacturing
process)
4. Two candidate operators have been qualified for applying of the gage.
5. The set of gage has been calibrated by the authorized department before.
6. During the examination the identification numbers of parts are not visible to the
appraisers.
7. Part model selected randomly among 35 models.
8. According to product’s Checklist sheet 17 characteristics for controlling have been
defined which should be conducted by the QC operator .Obviously, doing of the
GR&R study for all of the defined parameters with 17*3*15 repetition is extremely
difficult. Then, in consultation with experts, 4 points were selected, this points
identified as D1, D2, D3 and D4 on the check list sheet and parts.
9. To reducing of environment effect during of study, GR&R started and continued for
two appraisers without any interruption.
10. The condition such as gage, method, examiner and sample parts are equal for two
appraisers.
11. During of the study, supervisor of the examiner has not authorized to butt in
41
measurement process.
12. During of the study, just one of the two appraisers is ready in the place.
13. The resolution of the gages is 0.01 (i.e., the equipment is capable to read a change of
0.01 it is mean the general rule of thumb which says “the measuring instrument
discrimination ought to be at least one-tenth of the range to be measured ” is
satisfied).
15. The tolerance of the stated characteristic (D1, D2, D3, D4) which defined by the
product design department are 5.6(+/-0.6), 5(+/-0.6), 5(+/-0.6), 4(+/-0.6)
respectively.
16. In this experiment the first error type is equal to 0.05%.
Assumptions related with Stability are:
1. The set of the gage has been calibrated by the authorized department and their
certificates exist.
2. Due to lack of existing of reference value(s) relative to a traceable standard, a
production part that falls in the mid-range of the production measurements has been
selected (this part is known as master part).
3. The experiments doing under controlled and protected environment.
Here, the long study results (i.e. average and range methods, which computes the
total measurement system variability and allows the total measurement system
variability to be separated into repeatability, reproducibility and part variation) are
illustrated in Table 5.1. Then, the analysis of the results is presented on the following
pages. Remember that Appendix D illustrates numerical results and data format in Mini-
Tab 14.1(2003) software spreadsheet while the Stability and ANOVA analysis is
42
presented in Appendix C and Appendix E. in addition appendix F and Appendix G may
be useful for the investigation of the measurement system capability to identifying part-
to-part variation and test of normality assumption on the D2 and D4 parameters.
Parameter
Title D1 D2 D3 D4
-Repeatability
- Reproducibility
Total GR&R
- Part-to-Part
Total Variation
29.93 24.25 38.52 92.28 100
45.38
35.49
58.23
81.30
100
59.55
0.00 59.55
80.33
100
49.03 37.52
51.79
78.52
100
ndc 3 2 2 2
Table 4.5: The results of the average and range method
As stated before, two operators (appraisers) and 5 parts were used for the long
study. Each operator inspected each of the 5 parts three different times. For the D1, D2,
D3 and D4 parameters, the GR&R percentage is greater than 30%, which don’t satisfy
the criteria for a good gage. For the three parameters D2, D3 and D4 the number of
distinct category is barely 2. Thus this is no better than a Go/Not Go gage.
Moreover for D1, D2 and D4 the repeatability (the variability of the
measurements obtained by one person while measuring the same item repeatedly) is
large compared to reproducibility (the variability of the measurement system caused by
differences in operator behavior), the reasons may be:
• The instrument needs maintenance.
• The gage may need to be redesigned to be more rigid.
43
Source DF SS MS F P
PART 4 5.5798 1.39495 3.31633 0.136*
OPERATOR 1 0.3371 0.33708 0.80137 0.421
PART * OPERATOR 4 1.6825 0.42063 1.62439 0.207*
Repeatability 20 5.1789 0.25895
Total 29 12.7783
* Significant at 05.0=α
• The clamping or location for gaging needs to be improved.
• There is excessive within-part variation.
One graphical method that is suggested is called an interaction plot. This plot
confirms the results of the F test on whether or not the interaction is significant. In this
particular interaction plot, the average measurement per appraiser per part vs. part
number (1, 2... etc.) are graphed in Figures 5.1, 5.2, 5.3, 5.4. Moreover, due to in the
average and range method, the interaction between operators and parts component could
not be estimated. For illustration of this interaction, the ANOVA fixed effect model can
be used. Appendix E presents ANOVA been used.
.
Table 4.6: ANOVA results for D1, fixed effects model
Figure 4.2: Operator * Part interaction for D1
PART
Average
54321
8.0
7.5
7.0
6.5
6.0
OPERA TO R
1
2
OPERATOR * PART Interaction
Gage R&R (Xbar/R) for D1
44
As an overall trend, the points for each appraiser average measurement per part
are connected to form k (number of appraisers) lines. The way to interpret the graph is if
the k lines are parallel there is no interaction term. When the lines are nonparallel, the
interaction can be significant. The larger the angle of intersection is, the greater is the
interaction. In this experiment in Figure 4.2 which is presented above, the lines are
widely divergent. In conclusion, appropriate measures should be taken to eliminate the
potential causes for the interactions.
Figure 4.3: Operator * Part interaction for D2
Table 4.7: ANOVA results for D2, fixed effects model
PART
Average
54321
6.2
6.0
5.8
5.6
5.4
5.2
5.0
4.8
OPERATOR
1
2
OPERATOR * PART Interaction
Gage R&R (Xbar/R) for D2
Source DF SS MS F P
PART 4 2.21949 0.554872 1.48804 0.355*
OPERATOR 1 0.34776 0.347763 0.93262 0.389*
PART * OPERATOR 4 1.49155 0.372888 1.28870 0.308*
Repeatability 20 5.78707 0.289353
Total 29 9.84587
* Significant at 05.0=α
45
Figure 4.4: Operator * Part interaction for D3
Table 4.8: ANOVA results for D3, fixed effects model
Figure 4.4 and Table 4.8 illustrate the worst condition of the interaction between
the parts and operators. Then, the ANOVA table is used to decompose the total variation
into four components: parts, appraisers, interaction of appraise and parts and
repeatability due to the instrument (see appendix E).
In this case (ANOVA results for D3) the reproducibility with 86.51% is large
compared to repeatability with 50.16%, and the possible causes could be:
• The appraiser needs to be better trained in how to use and read the gage
instrument.
PART
Average
54321
5.92
5.90
5.88
5.86
5.84
5.82
OPERATOR
1
2
OPERATOR * PART Interaction
Gage R&R (Xbar/R) for D3
Source DF SS MS F P
PART 4 0.0049200 0.0012300 0.18053 0.937
OPERATOR 1 0.0000300 0.0000300 0.00440 0.950
PART * OPERATOR 4 0.0272533 0.0068133 9.92233 0.000*
Repeatability 20 0.0137333 0.0006867
Total 29 0.0459367
* Significant at 05.0=α
46
Source DF SS MS F P
PART 4 1.00803 0.252008 0.98028 0.507
OPERATOR 1 0.16280 0.162803 0.63328 0.471
PART * OPERATOR 4 1.02831 0.257078 1.93321 0.144*
Repeatability 20 2.65960 0.132980
Total 29 4.85875
* Significant at 05.0=α
Figure 4.5: Operator * Part interaction for D4
Table 4.9: ANOVA results for D4, fixed effects model
• Calibrations on the gage dial are not clear.
• The modified fixture needed to help the appraiser use the gage more consistency (see
also Figure 4.7).
Figure 4.6 also displays the individual reading operators for all parts as Figure
4.7, Figure 4.8 and Figure 4.9. The main purposes of these figures are to gain insight
into:
• The effect of individual operators on variation consistency
• Indication of outlier readings (i.e., abnormal readings)
PART
Average
54321
5.0
4.8
4.6
4.4
4.2
4.0
OPERA TOR
1
2
OPERATOR * PART Interaction
Gage R&R (Xbar/R) for D4
47
OPERATOR
21
8
7
6
5
4
D1 by OPERATOR
Gage R&R (Xbar/R) for D1
Figure 4.6: Individual readings by operators for D1
The Figure 4.6, Figure 4.7 and Figure 4.9 indicate that the operator 2 has the main effect
on the excess variation and outlier readings.
OPERATOR
21
6.5
6.0
5.5
5.0
4.5
4.0
3.5
3.0
D2 by OPERATOR
Gage R&R (Xbar/R) for D2
Figure 4.7: Individual readings by operators for D2
The Figure 4.8 indicates that the operator 1 and operator 2 have the excess variation and
outlier readings. As stated before, this problem could be stem from the part, fixture or
both of them.
48
OPERATOR
21
5.975
5.950
5.925
5.900
5.875
5.850
5.825
5.800
D3 by OPERATOR
Gage R&R (Xbar/R) for D3
Figure 4.8: Individual readings by operators for D3
OPERATOR
21
6.5
6.0
5.5
5.0
4.5
4.0
D4 by OPERATOR
Gage R&R (Xbar/R) for D4
Figure 4.9: Individual readings by operators for D4
Furthermore, throughout the experiments we observed that fitting the parts into
the gage is associated with a great difficulty. Due to lack of automatic resting of parts
into the gage, both ends of parts (left end and right end) should be held by human
operators and in this case, locking the levers becomes only possible if one of these two
different method are used, a) using a chin or b) assistance provided by another human
operator.
49
With reference to stability results (for two characteristics of D2 and D4) which
are illustrated in Appendix C and summarized in Table 4.10, it is obvious that the special
cause(s) run (i.e. in D2 graph), and pattern (i.e. in D4 graph) are formed. In D2 graph
there was a shift from the target to the USL in location and the shape violates normality
assumption. (See appendix G which provided the test normality by ARENA 7.01
software). If there is no action taken to correct this, it has the potential of getting out of
control. Apart from the graphical detection, a major difference between the Cg and Cgk
is used to detect and indicates that there is a special cause present.
One of the main reasons for this instability could have occurred from the lack of
gage design or calibration performance. Then, the measurement system needs action
(revising the calibration interval) and improvement (organizing and defining the sub-
project under APQP management) to improve the controls among the quality planning.
Index D2 D4
gC 21.56 20.39
gkC 1.42 6.79
Tolerance 5
+/-0.6 4
+/-0.6
Target 5
4
Table 4.10: Capability results on the D2 and D4
50
CHAPTER 5
DISCUSSION OF THE RESULTS AND CONCLUSION
The purpose of this thesis is to find a solution for improvement of controls
among the product realization when designing and making the gas turbine blades by
applying the two Automotive Supply Chain tools known as Measurement System
Analysis (MSA) and Advanced Product Quality Planning (APQP). Note that most
project in the field of MSA application focused on the scrap analysis alone or with Six-
sigma methodology as a measuring tools. As sated earlier, this approach could not
guarantee the control of the potential risks. Therefore, the APQP and MSA (together)
should be implemented respectively.
During this study, APQP methodology has been proposed to provide the
concepts and guidelines for embedding the Plan, Do, Study (or Check) and Action
philosophy along the product life cycle. Then, the utilization of these two techniques in
Measurement System Analysis (i.e., Stability and Range average method) argued that,
whether the control of the process stages (or project stages) of the product design are
effective or need improvement.
Furthermore, the results of the %GR&R on the Dial parameters (D1, D2, D3 and
D4 readings) indicate that the measurement variation is too large as compared to total
variation. This shows an urgent need to improve the measurement system.
Due to the structure of the ABC Company (Projectized, see Chapter 4), the
formation of the Cross Functional Team (CFT) with the assistance of the project
51
managers was the most difficult part of this study. The other limitations of this study
were:
1. The analysis has not investigated other factors or procedures that influence the
quality of the final product and potential risk.
2. This analysis evaluates process stability and capability for two parameter D2 and D4,
GR&R for D1, D2, D3 and D4 among 17 parameters.
3. This analysis did not consider financial impacts of the company ABC.
Some suggestions have been offered to help the improvement of the current
quality planning and resolving of the non-conformities which arise from Integrated
Management System requirements at the ABC company.
1. To satisfy of the expectations and requirements of the current IMS as well as
establishing lines of communication with other internal and external customers and
suppliers, the APQP team member should be arranged. Then, these requirements should
be formed as checklist to assure that if the relevant needs (such as: ISO 14000
requirements) are not met then, the next stages will be limited. Moreover doe to, the
projects managers will contrast with the new forms, it is suggested that this requirement
merged in the currents quality planning and checklists.
2. Since the success of an advanced product quality plan is dependent upon an effective
training program, then some courses emphasizing on the variation, capability,
repeatability and reproducibility concepts should be defined and developed in ABC
Company.
3. During the planning and execution of the project, the team will encounter product
52
design and/or processing concerns (such as: excess GR&R, non-conformities which
arise from internal or external auditing based on the ISO9001 or ISO14001). These
concerns should be documented on a matrix with assigned responsibility and timing.
Disciplined Problem-solving methods (such as 8D) are recommended in difficult
situations (see Appendix I).
4. In order to evaluation of APQP team efforts, the %GR&R, gC and gkC trends should
be reported by this team. This result not only could be used as an indicator (ISO 9001:
4.1 requirement) but also as a statistical tool could be utilized for verification of design
stages. A sample of MSA plan which has been included to the Chapter H could be
considered as part of the proposed APQP.
5. In order to identify the potential risk of the designed or modified product and
improvement of the team working performance, the Failure Mode and Effect Analysis
(FMEA) is suggested. To illustration the relationship between the FMEA with other
tools and specific customer requirements (Q101 term indicate the specific requirement
of the Ford company.), Appendix H has been included to this thesis. Meanwhile, the
Figure H.1 could be considered as the sample of the procedures which describes the
APQP methodology clearly.
Base on the result of this study replicating this study in the same system or other
sector, using MSA, APQP and FMEA methodology may be added to ranking
(prioritizing) the potential risk. Argument which has been developed on the GR&R may
be helpful for future investigations / research.
53
REFERENCES [1] AIAG Editing Group. (1995). Statistical Process control. Automotive Industry
action Group, Detroit-MI, USA
[2] AIAG Editing Group. (2002). Measurement Systems Analysis. Automotive
Industry Action Group, Detroit-MI, USA
[3] APICS. (2001). Basic of supply chain management. CPIM Certification Review
course. Participant Guide, Version 2.
[4] An American national standard. (2004). A guide to the project management Body
of knowledge (PMBOK). Published by project management institute, Inc. ISBN:1-930699-50-6.
[5] Arena 7.01, input analyzer, (2000). Rockwell soft. Inc. [6] Bernd-Arno Behrens Æ Ingo Wilde Æ Manfred Hoffmann (2007). Complaint
management using the extended 8D-method along the automotive supply chain. German Academic Society for Production Engineering (WGP). 1(1), 91-95
[7] Carol J. Robinson, Manoj K. Malhotra. (2004). Defining the concept of supply
chain quality management and its relevance to academic and industrial practice, Department of Management Science, The Moore School of Business, University of South Carolina, Columbia, SC 29208, USA, production economics, 42(4), 833-851.
[8] Choi. T. Y. (1999). Comparison of Quality Management Practices: Across the
Supply Chain and Industries. The Journal of Supply Chain Management. 35(1), 20-27.
[9] DaimlerChrysler, Ford Motor, General Motors. (1994). Advanced Product
Quality Planning and Control Plan. Published by AIAG, USA.
[10] David Hoyle. (2005b). Automotive Quality System. Handbook: ISO/TS
16949:2002. Published by Butterworth-Heinemann.
[11] David Hoyle. (2005a). ISO9000 Quality Systems Handbook. Butterworth-
Heinemann. [12] Deming, W. E. (1989, revised 1991). Foundation for management of quality in
54
the Western World. Paper presented to the Institute of Management and Sciences, Osaka, Japan, July1989.
[13] D. H. Stamatis (2003). Failure Mode Effect Analysis: FMEA from theory to
Execution, second edition. Published by American Society for Quality. [14] Fu-Kwun Wang. (2007). Applying Principal Component Analysis to a GR&R
Study. Department of Industrial Management National Taiwan University of Science and Technology. Journal of the Chinese Institute of Industrial Engineers, 24(2), 182-189.
[15] Gonza´lez-Benito J. et. al. (2005). An analysis of the relationship between
environmental motivations and ISO 14001 certification. British Journal of Management. 16(2):133-148.
[16] IQNet. (2006). the International Certification Network, IQNet Association, Switzerland. [17] International Organization for Standardization (ISO) (2002). ISO 9001: Quality
Management Systems – Requirements. [18] International Organization for Standardization (ISO) (2004). ISO 14001:
Environmental Management Systems – Requirements.
[19] J. D. Booker et. al. (2001). Designing Capable and Reliable Products. Published
by Butterworth-Heinemann.
[20] Jacobs, D.G. (2003). Anatomy of a supply chain. Transportation & Distribution.
44 (6), 60–63.
[21] James B. Ayers. (2001). Handbook of Supply Chain Management. CRC Press.
[22] Jeh-Nan Pan. (2006). Evaluating the Gauge Repeatability and Reproducibility for
different Industries. Department of Statistics, National Cheng-Kung University, University Road, Taiwan. Springer, 40(4), 499-518.
[23] Kinsella, B. (2003). The Wal-Mart factor. Industrial Engineer Vol. 31 No.10,
pp.529-536.
[24] Krisztina Demeter, Andrea Gelei, Istva´ n Jenei. (2006). The effect of strategy
on supply chain configuration and management practices on the basis of two
supply chains in the Hungarian automotive industry. Corvinus University of
55
Budapest. Vol. 104, PP. 557-570.
[25] Lee, HL. (2000). Creating value through supply chain integration. Supply Chain
Management Review, Vol. 4 No. 4, pp. 30-6.
[26] M. Bobrek a, M. Sokovic. (2005). Implementation of APQP-concept in design of
QMS. Journal of Materials Processing Technology, Vol. 162–163, pp. 718–724.
[27] Mahesh S. et al. (2005). Six Sigma: concepts, tools, and applications. Industrial
Management & Data Systems, Vol: 105, pp. 491-505.
[28] Minitab 14.1, (2003).MINITAB Inc.
[29] Mongomery, D.C. (2005b). Design and analysis of experiment. New Jersey Jon
Wily & Sons.
[30] Mongomery, D.C. (2005b). Introduction to statistical quality control. New Jersey
Jon Wily & Sons.
[31] Robert Sroufe et al. (2007). an examination of ISO 9000:2000 and supply chain quality assurance. Journal of Operations Management, 26.503–520.
[32] Stamatis, D.H. (2000). Who needs Six Sigma anyway?. Quality Digest e-Store,
available at:www.qualitydigest.com/may00/html/sixsigmacon.html (accessed 30 January 2004).
[33] S.X. Zeng and P. Tian. (2005). Implementing integration of ISO 9001 and
ISO14001 for construction. Managerial Auditing Journal. Vol. 20 No. 4, 2005 pp. 394-407.
[34] Keah-Choon Tan., Kannan, V.R., Handfield, R.B., Ghosh, S. (1999). Supply
chain management: An empirical study of its impact on performance. International Journal of Operations & Production Management, 19, 1034–1035.
[35] Tsai, P. (1989). Variable gauge repeatability and reproducibility study using the analysis of variance method. Quality Engineering 1(1): 107–115.
56
APPENDIX A
ISO 9001 AND ISO 14001
A.1 The ISO 9000 and principles
ISO9000 defines a general quality management system as a set of interrelated or
interaction process that achieve the quality policy and quality objective.
To date, over half a million organizations in over 150 countries have registered
to quality management system through ISO standards. Just in the United States, Over
50,000 companies have obtained the new ISO 9000:2000 registration (IQNet, 2006).
The new revision is based on a process model approach and structures 21
elements into four major sections: management responsibility, resource management,
product realization and measurement, analysis and improvement (see Figure A.1)
The eight quality management principles as defined by ISO which are illustrated as
bellow:
• Principle 1: Customer focus
Organizations depend on their customers and therefore, should understand
current and future customer needs, should meet customer requirements, and attempt to
exceed customer expectations.
• Principle 2: Leadership.
Leaders establish unity of purpose and direction of the organization. They should
create and maintain the internal environment in which people can become completely
involved in achieving the organization’s objectives.
57
• Principle 3: Involvement of people.
People at all levels are the essential of an organization and their full involvement
enables their abilities to be used for the organization’s benefit.
• Principle 4: process approach.
A desired result is achieved more efficiently when activities and related
resources are managed as a process.
• Principle 5: System approach to management.
Identifying, understanding and managing interrelated processes as a system gives
to the organization’s effectiveness and efficiency in achieving its objectives.
• Principle 6: Continual improvement.
Continual improvement of the organization’s overall performance should be an
everlasting objective of the organization.
• Principle 7: Factual approach to decision making.
Effective decisions are based on the analysis of data and information.
• Principle 8: Mutually beneficial supplier relationships.
An organization and its suppliers are interdependent and a mutually beneficial
relationship enhances the ability of both to create value.
• ISO 9000:2000 requirements
1 Scope
1.1 General
1.2 Permissible exclusions
2 Normative references
3 Terms and definitions
58
Continual improvement of the quality management system
Management
responsibility
Measurement, analysis, andimprovement
Resource management
Product
realization
Customers
RequiremeInp
Legend Value adding Informatio
Customers
Satisfacti
Outpu Produc
4 Quality management system
Figure A.1: General ISO 9001 model
4.1 General requirements
The organization shall establish, document, implement and maintain a quality
management system and continually improve its effectiveness in accordance with the
requirements of this International Standard.
The organization shall
a) Identify the processes needed for the quality management system and their
application throughout the organization,
b) Determine the sequence and interaction of these processes,
59
c) Determine criteria and methods needed to ensure that both the operation and control
of these processes are effective,
d) Ensure the availability of resources and information necessary to support the
operation and monitoring of these processes,
e) Monitor, measure and analyze these processes, and
f) Implement actions necessary to achieve planned results and continual improvement of
these processes.
These processes shall be managed by the organization in accordance with the
requirements of this International Standard.
Where an organization chooses to outsource any process that affects product
conformity with requirements, the organization shall ensure control over such
processes. Control of such outsourced processes shall be identified within the quality
management system.
NOTE Processes needed for the quality management system referred to above should
include processes for management activities, provision of resources, product realization
and measurement.
4.2 General documentation requirements
5. Management responsibility
5.1 Management commitment
5.2 Customer focus
5.3 Quality policy
5.4 Planning
5.4.1 Quality objectives
5.4.2 Quality planning
60
5.5 Administration
5.5.1 General
5.5.2 Responsibility and authority
5.5.3 Management representative
5.5.4 Internal communication
5.5.5 Quality manual
5.5.6 Control of documents
5.5.7 Control of quality records
5.6 Management review
5.6.1 Review input
5.6.2 Review output
6. Resource management
6.1 Provision of resources
6.2 Human resources
6.2.1 Assignment of personnel
6.2.2 Training, awareness and competency
6.3 Facilities
6.4 Work environment
7. Product realization
7.1 Planning of realization processes
7.2 Customer-related processes
7.2.1 Identification of customer requirements
7.2.2 Review of product requirements
7.2.3 Customer communication
61
7.3 Design and/or development
7.3.1 Design and/or development planning
The organization shall plan and control the design and development of product.
During the design and development planning, the organization shall determine
a) The design and development stages,
b) The review, verification and validation that are appropriate to each design and
development stage, and
c) The responsibilities and authorities for design and development.
The organization shall manage the interfaces between different groups involved
in design and development to ensure effective communication and clear assignment of
responsibility. Planning output shall be updated, as appropriate, as the design and
development progresses.
7.3.2 Design and/or development inputs
7.3.3 Design and/or development outputs
7.3.4 Design and/or development review
7.3.5 Design and/or development verification
7.3.6 Design and/or development validation
7.3.7 Control of design and/or development changes
7.4 Purchasing
7.4.1 Purchasing control
7.4.2 Purchasing information
7.4.3 Verification of purchased products
7.5 Production and service operations
7.5.1 Operations control
62
7.5.2 Identification and traceability
7.5.3 Customer property
7.5.4 Preservation of product
7.5.5 Validation of processes
7.6 Control of measuring and monitoring devices
8. Measurement, analysis and improvement
8.1 Planning
8.2 Measurement and monitoring
8.2.1 Customer satisfaction
8.2.2 Internal audit
8.2.3 Measurement and monitoring of processes
8.2.4 Measurement and monitoring of product
8.3 Control of nonconformity
8.4 Analysis of data
The organization shall determine, collect and analyze appropriate data to
demonstrate the suitability and effectiveness of the quality management system and to
evaluate where continual improvement of the effectiveness of the quality management
system can be made. This shall include data generated as a result of monitoring and
measurement and from other relevant sources. The analysis of data shall provide
information relating to
a) Customer satisfaction (see 8.2.1),
b) Conformity to product requirements,
c) Characteristics and trends of processes and products including opportunities for
preventive action, and
63
d) Suppliers.
8.5 Improvement
8.5.1 Planning for continual improvement
8.5.2 Corrective action
8.5.3 Preventive action
A.2 Environmental Management Systems, ISO 14001
The ISO 14001 family of standards establishes a reference model for the
implementation of company environmental management systems (EMS), defined as
those parts of global management systems that gives details of the organizational
structure, planning activities, responsibilities, practices, procedures, processes and
resources for preparing, applying, reviewing and maintaining company environmental
policies. It contains standards that include guidelines for matters such as environmental
management, environmental auditing, environmental labeling or life cycle assessment.
The ISO 14001 standard is divided into five main sections: (4.2) environmental
policy, which involves making a statement of environmental intentions and principles;
(4.3) planning, which requires the company to specify the processes it uses to identify
the environmental problems that must be geared and to define specific objectives and
targets; (4.3.1 demand the organization shall establish, implement and maintain a
procedure to identify the environmental aspects of its activities and determining those
aspects that have or can have significant impact(s) on the environment.) (4.4)
implementation and operation, which involves both defining responsibilities for the
system and assuring the identification of training needs, the internal and external
knowledge of the system, the control of documents and operations, and the preparedness
64
for and response to emergencies; (4.5) checking and corrective action, which involve
procedures to monitor operations and to prevent and mitigate any non-compliance with
objectives and targets; and (4.6) management review, which implies setting up processes
through which senior managers review the suitability and effectiveness of the system
and define appropriate changes Gonza´lez-Benito J. et al. (2005) (See Figure A.2).
Figure A.2: General model of ISO 14001
A.3 Process and system approach in ISO9001:
Of particular importance among the eight quality management principles are
system approch to management and process approch. The ‘‘process-approach’’ to
achieving quality and ultimately customer satisfaction is the premise of the ISO 9001.
This principle is expressed as follows:
65
A desired result is achieved more efficiently when related resources and
activities are managed as process. The process approch involve managing the
interrelated activates and associated resources together to achieve a particular output.
See Figure A.3.
Figure A.3: Process definition
By the process, AIAG mean, the whole combination of supplier, producer,
people, equipment, input materials, methods, and environment, that work together to
produce output and the customer who use the output(see Figure A.4). The total
performance of the process depends upon communication between supplier and
customer.
The system approach motivates organizations to link inputs to the system of
interrelated value-adding process of the organization. Taking system approach to
management means managing the organization as a system of processes so that all
process fit together, the input and output are connected and resources feed the process.
Activity
Activity
Activity
Process
In Out
66
Figure A.4: Process concept of AIAG point of view
Performance is monitored and sensors transmit information which causes
changes in performance and all parts work together to achieve the organization
objective.
THE WAY WE WORK/
BLENDING OF RESOURCES
CUSTOMERS
IDENTIFYING CHANGING NEEDS AND EXPECTIONS
STATISTICAL METHOD
VOICE OF CUSTOMER
VOICE OF THE PROCESS
PEOPLE
EQUIPMENTMATERIALS
METHODS
ENVRONMENT
INPUTS PROCESS/SYSTEM OUTPUTS
67
APPENDIX B
Value of 2d
Z
W
2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 1.41 1.91 2.24 2.48 2.67 2.83 2.96 3.08 3.18 3.27 3.35 3.42 3.49 3.55
2 1.28 1.81 2.15 2.40 2.60 2.77 2.91 3.02 3.13 3.2 3.30 3.38 3.45 3.51
3 1.23 1.77 2.12 2.38 2.58 2.75 2.89 3.01 3.11 3.21 3.29 3.37 3.43 3.50
4 1.21 1.75 2.11 2.37 2.57 2.74 2.88 3.00 3.10 3.20 3.28 3.36 3.43 3.49
5 1.19 1.74 2.10 2.36 2.56 2.78 2.87 2.99 3.10 3.19 3.28 3.36 3.42 3.49
6 1.18 1.73 2.09 2.35 2.56 2.73 2.87 2.99 3.10 3.19 3.27 3.35 3.42 3.49
7 1.17 1.73 2.09 2.35 2.55 2.72 2.87 2.99 3.10 3.19 3.27 3.35 3.42 3.48
8 1.17 1.72 2.08 2.35 2.55 2.72 2.87 2.98 3.09 3.19 3.27 3.35 3.42 3.48
9 1.16 1.72 2.08 2.34 2.55 2.72 2.86 2.98 3.09 3.19 3.27 3.35 3.42 3.48
10 1.16 1.72 2.08 2.34 2.55 2.72 2.86 2.98 3.09 3.18 3.27 3.34 3.42 3.48
11 1.15 1.71 2.08 2.34 2.55 2.72 2.86 2.98 3.09 3.18 3.27 3.34 3.41 3.48
12 1.15 1.71 2.07 2.34 2.55 2.72 2.85 2.98 3.09 3.18 3.27 3.34 3.41 3.48
13 1.15 1.71 2.07 2.34 2.55 2.71 2.85 2.98 3.09 3.18 3.27 3.34 3.41 3.48
14 1.15 1.71 2.07 2.34 2.54 2.71 2.85 2.98 3.09 3.18 3.27 3.34 3.41 3.48
15 1.15 1.7 2.07 2.34 2.54 2.71 2.85 2.98 3.08 3.18 3.26 3.34 3.41 3.48
>15 1.28 1.693 2.059 2.326 2.534 2.704 2.847 2.97 3.078 3.173 3.258 3.336 3.407 3.472
Table B.1: Value of 2d
68
APPENDIX C
STABILITY STUDY AND RESULTS
Sub-Group#
1 2 3 4 5 6 7 8 9 10
5.56 5.54 5.56 5.56 5.56 5.55 5.57 5.56 5.57 5.56
Data from 5.56 5.57 5.56 5.56 5.56 5.56 5.57 5.52 5.56 5.56
Each
Sub-Group 5.56 5.57 5.55 5.54 5.57 5.56 5.57 5.56 5.56 5.56
5.55 5.55 5.55 5.56 5.56 5.57 5.57 5.56 5.56 5.57
D2
5.56 5.56 5.57 5.56 5.56 5.56 5.57 5.57 5.57 5.57
Total 27.79 27.79 27.79 27.78 27.81 27.8 27.85 27.77 27.82 27.82
Average 5.558 5.558 5.558 5.556 5.562 5.56 5.57 5.554 5.564 5.564
(Xbar)
Range(R) 0.01 0.03 0.02 0.02 0.01 0.02 0 0.05 0.01 0.01
Table C. 2: Stability results for D2
Sample Mean
10987654321
5.57
5.56
5.55
__X=5.5604
UCL=5.57285
LCL=5.54795
Sample Range
10987654321
0.04
0.02
0.00
_R=0.02158
UCL=0.04563
LCL=0
Sample
Values
108642
5.56
5.54
5.52
5.585.575.565.555.545.535.52
5.585.565.545.52
Within
Overall
Specs
Within
S tDev 0.00928
C p 21.56
C pk 1.42
C C pk 21.56
O v erall
S tDev 0.00952
Pp 21.02
Ppk 1.39
C pm *
1
Process Capability Sixpack of master D2
Xbar Char t
R Chart
Last 1 0 Subgroups
Capability H istogram
Normal P rob P lot
A D: 4.342, P : < 0.005
Capability P lot
Figure C.1: Six pack report of master D2
69
Sub-Group#
1 2 3 4 5 6 7 8 9 10
4.4 4.42 4.4 4.41 4.4 4.42 4.38 4.42 4.4 4.39
Data from 4.39 4.38 4.41 4.4 4.4 4.41 4.39 4.39 4.41 4.39
Each
Sub-Group 4.39 4.38 4.41 4.42 4.39 4.41 4.4 4.41 4.41 4.4
4.41 4.41 4.41 4.4 4.4 4.39 4.39 4.41 4.41 4.39
D4
4.4 4.4 4.39 4.4 4.4 4.4 4.4 4.4 4.4 4.39
Total 21.99 21.99 22.02 22.03 21.99 22.03 21.96 27.77 27.82 21.96
Average 4.398 4.398 4.404 4.406 4.398 4.406 4.392 5.554 5.564 4.392
(Xbar)
Range(R) 0.02 0.04 0.02 0.02 0.01 0.03 0.02 0.05 0.01 0.01
Table C. 3: Stability results for D4
Sample Mean
10987654321
4.41
4.40
4.39
__X=4.4006
UCL=4.41376
LCL=4.38744
Sample Range
10987654321
0.04
0.02
0.00
_R=0.02281
UCL=0.04824
LCL=0
Sample
Values
108642
4.42
4.40
4.38
4.424.414.404.394.38
4.4254.4104.3954.380
Within
Overall
Specs
Within
StDev 0.00981
Cp 20.39
C pk 6.78
CC pk 20.39
O v erall
StDev 0.01044
Pp 19.17
Ppk 6.37
C pm *
Process Capability Sixpack of master D4
Xbar Chart
R Chart
Last 10 Subgroups
Capability Histogram
Normal Prob PlotAD: 1.893, P: < 0.005
Capability P lot
Figure C.2: Six pack report of master D4
70
APPENDIX D
NUMERICAL RESULTS AND DATA FORMAT
Table D.1: Numerical result of 4 characteristics
OPERATOR PART D1 D2 D3 D4
A 1 7.05 5.88 5.85 4.02
1 7.08 5.89 5.85 4
1 7.05 5.89 5.88 4.02
2 7.87 5.15 5.88 3.93
2 7.87 5.15 5.88 3.94
2 7.87 5.15 5.87 3.94
3 7.22 5.85 5.92 3.94
3 7.22 5.85 5.92 3.94
3 7.21 5.85 5.92 3.95
4 7.35 5.1 5.84 3.99
4 7.35 5.1 5.84 3.99
4 7.35 5.1 5.84 3.99
5 5.95 5.84 5.82 4.17
5 5.95 5.85 5.82 4.17
5 5.95 5.84 5.82 4.18
B 1 4.12 2.85 5.95 5.2
1 5.91 5.81 5.82 4.2
1 5.9 5.8 5.82 4.21
2 7.87 5.15 5.87 3.94
2 7.85 5.15 5.87 3.94
2 7.9 5.17 5.87 3.88
3 7.34 5.09 5.83 4
3 7.32 5.09 5.83 4.01
3 7.24 5.01 5.83 4.07
4 7.04 5.87 5.84 4.03
4 7.04 5.87 5.84 4.04
4 7.05 5.88 5.84 4.02
5 7.22 5.85 5.92 3.94
5 7.21 5.85 5.92 3.94
5 7.19 5.84 5.92 3.95
71
PARTS OPER. D1 D2 D3 D4
1 1 7.05 5.88 5.85 4.02 1 1 7.08 5.89 5.85 4.00 1 1 7.05 5.89 5.88 4.02 2 1 7.87 5.15 5.88 3.93 2 1 7.87 5.15 5.88 3.94 2 1 7.87 5.15 5.87 3.94 3 1 7.22 5.85 5.92 3.94 3 1 7.22 5.85 5.92 3.94 3 1 7.21 5.85 5.92 3.95 4 1 7.35 5.10 5.84 3.99 4 1 7.35 5.10 5.84 3.99 4 1 7.35 5.10 5.84 3.99 5 1 5.95 5.84 5.82 4.17 5 1 5.95 5.85 5.82 4.17 5 1 5.95 5.84 5.82 4.18 1 2 4.12 2.85 5.95 5.20 1 2 5.91 5.81 5.82 4.20 1 2 5.90 5.80 5.82 4.21 2 2 7.87 5.15 5.87 3.94 2 2 7.85 5.15 5.87 3.94 2 2 7.90 5.17 5.87 3.88 3 2 7.34 5.09 5.83 4.00 3 2 7.32 5.09 5.83 4.01 3 2 7.24 5.01 5.83 4.07 4 2 7.04 5.87 5.84 4.03 4 2 7.04 5.87 5.84 4.04 4 2 7.05 5.88 5.84 4.02 5 2 7.22 5.85 5.92 3.94 5 2 7.21 5.85 5.92 3.94 5 2 7.19 5.84 5.92 3.95
Table D.2: Data format in Minitab
72
APPENDIX E
GAGE R&R STUDY - ANOVA METHOD
Two-Way ANOVA Table with Interaction
Table E.1: ANOVA results for D1, fixed effects model
Table E.2: ANOVA results for D2, fixed effects model
Gage R&R for D1
Study Var %Study Var
Source StdDev (SD) (6 * SD) (%SV)
Total Gage R&R 0.559322 3.35593 81.14
Repeatability 0.508868 3.05321 73.82
Reproducibility 0.232152 1.39291 33.68
OPERATOR 0.000000 0.00000 0.00
OPERATOR*PART 0.232152 1.39291 33.68
Part-To-Part 0.402972 2.41783 58.46
Total Variation 0.689367 4.13620 100.00
Number of Distinct Categories = 1
Gage R&R for D2
Study Var %Study Var
Source StdDev (SD) (6 * SD) (%SV)
Total Gage R&R 0.553391 3.32035 93.79
Repeatability 0.550705 3.30423 93.33
Reproducibility 0.054459 0.32676 9.23
OPERATOR 0.054459 0.32676 9.23
Part-To-Part 0.204775 1.22865 34.70
Total Variation 0.590063 3.54038 100.00
Number of Distinct Categories = 1
73
Table E.3: ANOVA results for D3, fixed effects model
Table E.4: ANOVA results for D4, fixed effects model
Gage R&R for D3
Study Var %Study Var
Source StdDev (SD) (6 * SD) (%SV)
Total Gage R&R 0.0522388 0.313433 100.00
Repeatability 0.0262043 0.157226 50.16
Reproducibility 0.0451910 0.271146 86.51
OPERATOR 0.0000000 0.000000 0.00
OPERATOR*PART 0.0451910 0.271146 86.51
Part-To-Part 0.0000000 0.000000 0.00
Total Variation 0.0522388 0.313433 100.00
Number of Distinct Categories = 1
Gage R&R for D4
Study Var %Study Var
Source StdDev (SD) (6 * SD) (%SV)
Total Gage R&R 0.417548 2.50529 100.00
Repeatability 0.364664 2.18799 87.33
Reproducibility 0.203387 1.22032 48.71
OPERATOR 0.000000 0.00000 0.00
OPERATOR*PART 0.203387 1.22032 48.71
Part-To-Part 0.000000 0.00000 0.00
Total Variation 0.417548 2.50529 100.00
Number of Distinct Categories = 1
74
APPENDIX F
Xbar and R CHARTS
Figure F.1: Xbar and R chart on D1
Figure F.2: Xbar and R chart on D2
Sample Range
3
2
1
0
_R=0.304
UCL=0.783
LCL=0
1 2
Sample Mean 7.5
7.0
6.5
6.0
__X=7.186
UCL=7.497
LCL=6.875
1 2
R Chart by OPERATOR
Xbar Chart by OPERATOR
Gage R&R (Xbar/R) for D1
Sample Range
3
2
1
0
_R=0.311
UCL=0.801
LCL=0
1 2
Sample Mean
6.0
5.5
5.0
__X=5.861
UCL=6.179
LCL=5.543
1 2
R Chart by OPERATOR
Xbar Chart by OPERATOR
Gage R&R (Xbar/R) for D2
75
Figure F.3: Xbar and R chart on D3
Figure F.4: Xbar and R chart on D4
Sample Range
1.0
0.5
0.0
_R=0.122
UCL=0.314
LCL=0
1 2
Sample Mean
4.6
4.4
4.2
4.0
__X=4.0517
UCL=4.1765
LCL=3.9269
1 2
R Chart by OPERATOR
Xbar Chart by OPERATOR
Gage R&R (Xbar/R) for D4
Sample Range
0.15
0.10
0.05
0.00
_R=0.018
UCL=0.0463
LCL=0
1 2
Sample Mean
5.91
5.88
5.85
5.82
__X=5.8643
UCL=5.8827
LCL=5.8459
1 2
R Chart by OPERATOR
Xbar Chart by OPERATOR
Gage R&R (Xbar/R) for D3
76
APPENDIX G
NORMALITY TESTING BY ARENA INPUT ANALYSER
Table G.1: output of Arena analyzer on D2
Table G.2: output of Arena analyzer on D4
Distribution Summary for D4 Distribution: Normal Expression: NORM(4.4, 0.0103) Square Error: 0.003163 Chi Square Test Number of intervals = 4 Degrees of freedom = 1 Test Statistic = 0.828 Corresponding p-value = 0.393 Kolmogorov-Smirnov Test Test Statistic = 0.174 Corresponding p-value = 0.0884 Data Summary Number of Data Points = 50 Min Data Value = 4.38 Max Data Value = 4.42 Sample Mean = 4.4 Sample Std Dev = 0.0104
Distribution Summary for D2 Distribution Summary Distribution: Normal Expression: NORM(5.56, 0.00937) Square Error: 0.127289 Chi Square Test Number of intervals = 4 Degrees of freedom = 1 Test Statistic = 25.3 Corresponding p-value < 0.005 Kolmogorov-Smirnov Test Test Statistic = 0.323 Corresponding p-value < 0.01 Data Summary Number of Data Points = 50 Min Data Value = 5.52 Max Data Value = 5.57 Sample Mean = 5.56 Sample Std Dev = 0.00947
77
APPENDIX H
FMEA INTERRELATIONSHIPS
78
Figure H.1: FMEA interrelationships, D. H. Stamatis (2003)
Quality planning team
Supplier quality assistance
Supplier or customer product engineering
Warranty responsible activity
Does customer feedback suggest control plan changes?
Are operating And SPC procedures sufficient to make control plan work?
Are preventive process actions identified?
Is the quality system Ford approved?
Have characteristics for sensitive process been identified for SPC?
Can control charts for variables be Used on all key characteristics?
Is the process ready for sign-off ?
Is 100% inspection required?
Does the plan have customer concurrency?
Can product be manufactured, Assembled, and tested?
Are engineering changes required?
- Repair rate objective - Repair cost objective - Field concerns - Plant concerns
Q101 quality system
Feasibility analysis - process/inspection flowchart - process FMEA - Floor plan - Historical warranty quality analysis - New equipment list - Previous statistical studies - Design of experiments - Cause-and-effect diagram
Can causes of field plant concerns be monitored?
Manufacturing control plan - Quality system/procedures - Key process/product characteristics - Sample size/frequency - Inspection methods - Reaction plan - Statistical methods - Problem-solving discipline
Process potential study - Statistical training - Implementation - Results
Process sign-off - Process sheets - Inspection instructions - Test equipment/gauges - Initial samples - Packing
Are process changes needed to improve feasibility?
Job # 1 Ford Motor Company
Never-ending improvement
Has a launch Team been Identified?
Was the process FMEA used to develop Process sheets?
79
Characteristic Specification Ranking
&
Source
Instrument
&
Resolution
Study date GR&R Action plan #
Thickness 4+,- 0.1 A
FMEA#035
Micrometer
0.01 26.2.2009 60% 02
Diameter 6+,- 0.5 B
SPC(2.2.2009)
Caliper
0.1 26.2.2009 29% 03
Table: H.1 MSA plan sample
80
APPENDIX I
PROBLEM SOLVING METHOD
The U.S. Government first standardized the 8D process during the Second World
War, referring to it as Military Standard 1520: Corrective action and disposition system
for nonconforming material. It was later popularized by the Ford Motor Company in the
1960’s and 1970’s. 8D has become a standard in the automotive supply chain. The 8D
Problem Solving Process is used to identify, correct and eliminate problems (i.e.
corrective action). The methodology is useful in product and process improvement. It
establishes a standard practice, with an emphasis on facts. It focuses on the origin of the
problem by establishing Root Cause. The extended 8D-method can be applied to an
unlimited number of customer-supplier-relations along the supply chain Bernd-Areno et
al. (2007).
8D procedure
As stated before, the automotive industry has agreed on a common method to
deal with complaints and to communicate these to the suppliers. This method is called
8D-Report (D for disciplines).It is a standardized procedure to handle fault complaints
and their corrective action plans. Within this method, the filed complaint is sent to the
supplier, who sets up a team to deal with the complaint (Figure I.1 shows a proposed
structure for the identifying and eliminating of the current problem root cause(s) on the
surface of the blade tile)
81
The procedure is illustrated as bellow:
Figure I.1: 8D procedure, Bern-Areno et al. (2007)
D.1 Form the Cross Functional Team
This is the first step of the 8D process and the first part of the 8D report and
defines the composition of the 8D team. The team should be cross-functional and
should include as members the process owner, a member from QA, and others who will
be involved in the containment, analysis, correction and prevention of the problem (in
the automotive supply chain this team is known as Cross Functional Team).
D.2 Describe the Problem This step involves a detailed assessment of the problem highlighted by the
customer. Under this step, the 8D report provides background information on and a
clear statement of the problem being highlighted by the customer.
1-Build Core Team
2-Describe Problem
3-Containment Action
4-Root Cause Analysis
5- Plan Corrective Action
6- Take Corrective Action
7- Stop Re-occurrence
8-Report Closure
82
D. 3 Contain the Problem
This discipline explains the extent of the problem and bounds it. Based on initial
problem investigation, all lots that are potentially affected by the same problem must be
identified and their locations addressed.
If the problem has an extremely high reliability risk and the application of the
product is critical (e.g., failure of the product is life-threatening), lots already in the field
may need to be recalled. However, recall must only be done under extreme cases
wherein the impact of reliability risk is greater than the impact of recall.
D.4 Identify the Root Cause
This 8D process step consists of performing the failure analysis and investigation
needed to determine the root cause of the problem. The corresponding portion in the 8D
report documents the details of the root cause analysis conducted. A detailed description
of the actual failure mechanism must be given, to show that the failure has been fully
understood.
D. 5 Formulate and Verify Corrective Actions
This next discipline identifies all possible corrective actions to address the root
cause of the problem. The owners of the corrective actions and the target dates of
completion shall be enumerated in this section of the report. It is also suggested that the
reasons behind each corrective action be explained in relation to the root cause.
83
D. 6 Correct the Problem and Confirm the Effects
The sixth discipline of the 8D process involves the actual implementation of the
identified corrective actions, details of which must be documented in the conforming
portion of the 8D report.
D.7 Prevent the Problem
Actions necessary to prevent these from being affected by a similar problem in
the future are called preventive actions. All preventive actions must be listed, along with
their owners and target dates of completion.
Note that some format of the 8D which started by the customer is needed to
identify the level of the changes requirements in documents such as control plan,
FMEA, instruction and so on.
D.8 Congratulate the Team
The last step of the 8D process consists of an acknowledgement from
management of the good work done by the 8D team. Approvals for the 8D report are
also presented in this last discipline.