Faculty of Engineering, Blekinge Institute of Technology, 371 79 Karlskrona, Sweden
Feasibility study for geometry
assurance in low volume manufacturing of complex products
With application in the shipbuilding industry
Henrik Ehrenberg | Filip Malmenryd
Blekinge Institute of Technology, Karlskrona, Sweden 2020
Master of Science in Mechanical Engineering
June 2020
i
This thesis is submitted to the Faculty of Engineering at Blekinge Institute of Technology in partial fulfilment of the requirements for the degree of Master of Science in Mechanical Engineering. The thesis is equivalent to 20 weeks of full time studies.
The authors declare that they are the sole authors of this thesis and that they have not used any sources other than those listed in the bibliography and identified as references. They further declare that they have not submitted this thesis at any other institution to obtain a degree.
Contact information Authors:
Henrik Ehrenberg heeh15@student.bth.se Filip Malmenryd
fima15@student.bth.se
Institute advisor:
Professor Tobias Larsson Tobias.larsson@bth.se
Department of Mechanical Engineering
Faculty of Engineering
Blekinge Institute of Technology SE-371 79 Karlskrona, Sweden
Web: www.bth.se Phone: + 46 455 38 50 00 Fax: + 46 455 38 50 57 Company advisors:
Per Wahlsten
per.wahlsten@saabgroup.com Production Engineering Per Bergström
per.x.bergstrom@saabgroup.com
Manufacturing Engineering Outfitting
ii
Abstract
Geometrical variation is an unavoidable aspect in all types of manufacturing that may, unless managed, risk failure in fulfilling product requirements which may result in rework, delays and bad publicity. The term geometry assurance includes the tools, methods and processes that can be utilized to manage the effects of geometrical variation and to ensure fulfillment of esthetical, functional and assembly requirements. While state of the art research in geometry assurance is extensively applied within the automotive and aerospace industries with great success, its application in low volume manufacturing of complex products remains limited. The shipbuilding industry is an example of such an industry, often manufacturing large and complex products in low quantities. Further, the shipbuilding industry has historically been labor-intensive and relied on craftsmanship throughout the product realization process. However, studies indicate that a technology-intensive development is crucial for companies in order to maintain market
competitiveness. This transition places high demands on a well-established geometry assurance process in order to ensure successful assembly and fulfillment of product requirements.
In this thesis, a feasibility study is conducted on how geometry assurance may be applied in low volume manufacturing of complex products. By developing guidelines on how geometry assurance may be applied, the purpose is to improve geometrical quality throughout the product realization process and to reduce lead times, costs and increase assembly precision.
To explore the feasibility of geometry assurance in low volume manufacturing of complex products, a work structure consisting of three phases was established. In the first phase, a current state analysis of the collaboration partner Saab Kockums was conducted parallel to studying state of the art research in geometry assurance. In phase two, the state of practice of companies in the automotive and aerospace industries was studied in order to determine how they apply state of the art research.
By interviewing industry specialists and combining gained knowledge from the first two phases, guidelines on how geometry assurance may be applied in low volume manufacturing of complex products was developed. In phase three, based on these guidelines, suggestions on how the geometry assurance process in pipe manufacturing at Saab Kockums can be improved was developed.
The results of this study indicate that geometry assurance is applicable in low volume manufacturing of complex products. However, alternative methods may be required. Based on gained knowledge and insights from interviews with industry specialists, guidelines on how geometry assurance in low volume manufacturing of complex products may be applied are proposed. In order to improve the geometry assurance process in pipe manufacturing at Saab Kockums, this study proposes general guidelines for improvement along with a process and prototype measurement tool for the fitting- pipe methodology. The specially designed prototype measurement tool presents an alternative measurement method that can be used in cramped spaces where it is difficult to access with a 3D- measurement arm, the proposed primary measurement technique.
In conclusion, this study indicates that geometry assurance is applicable in low volume manufacturing of complex products and suggests three methods for how it may be achieved.
However, each of these methods needs to be further investigated in order to determine their
applicability in other low volume manufacturing industries. Further, the prototype measurement tool and process for the fitting-pipe methodology indicates potential for improving the geometry
assurance process in pipe manufacturing. However, further work is needed to complete the process for fitting-pipes and to finalize the prototype measurement tool for production use.
Keywords: Geometry assurance, dimensional engineering, low volume manufacturing, complex
products, quality assurance, robust design
iii
Sammanfattning
Geometrisk variation är en oundviklig aspekt i alla typer av tillverkningsprocesser som medför risken att produktkrav inte uppfylls om det inte hanteras, vilket kan innebära konsekvenser som omarbete, förseningar och negativ publicitet. Begreppet geometrisäkring innefattar de verktyg, metoder och processer som kan användas för att hantera effekten av geometrisk variation och därmed säkerställa att estetiska-, funktionella- och sammansättningskrav uppfylls. Medan ledande forskning inom geometrisäkring är framgångsrikt applicerad i stor utsträckning inom fordon- och flygindustrin är dess applikation inom fåstyckstillverkning av komplexa produkter begränsad. Skeppsindustrin är ett exempel på en fåstyckstillverkning som ofta producerar stora och komplexa produkter i få antal.
Vidare har skeppsindustrin historiskt sett varit en arbetskraftsintensiv industri som förlitat sig på varvsarbetares hantverk och erfarenhet genom produktrealiseringsförloppet. Studier tyder dock på att utveckling mot en teknologiintensiv industri är avgörande för att bibehålla konkurrenskraftighet.
Denna övergång innebär därmed att en väletablerad geometrisäkringsprocess är avgörande för att säkerställa lyckad sammansättning och uppfyllnad av produktkrav.
I detta examensarbete studeras genomförbarheten av geometrisäkring inom fåstyckstillverkning av komplexa produkter. Genom att utveckla riktlinjer för hur geometrisäkring kan appliceras är syftet att förbättra den geometriska kvalitén genom produktrealiseringsförloppet för att minska ledtider, kostnader samt öka precision i sammansättning.
För att studera genomförbarheten av geometrisäkring inom fåstyckstillverkning av komplexa
produkter har en arbetsstruktur bestående av tre faser etablerats. I den första fasen genomfördes en nulägesanalys av samarbetspartnern Saab Kockums ur ett geometrisäkringsperspektiv. Parallellt studerades även forskning inom geometrisäkring. I fas två studerades företag inom fordon- och flygindustrin för att avgöra hur de applicerar ledande forskning inom geometrisäkring. Genom att intervjua industrispecialister och kombinera den samlade kunskapen från de föregående två faserna utvecklades riktlinjer för hur geometrisäkring kan appliceras inom fåstyckstillverkning av komplexa produkter. I fas tre användes de generella riktlinjerna för att utveckla förslag på hur
geometrisäkringsprocessen inom rörtillverkning på Saab Kockums kan förbättras.
Resultatet av denna studie indikerar att geometrisäkring är applicerbart inom fåstyckstillverkning av komplexa produkter. Dock kan det kräva alternativa arbetssätt och metoder i vissa aspekter. Baserat på den samlade kunskapen och insikter från industrispecialister föreslås riktlinjer för hur
geometrisäkring skulle kunna appliceras inom fåstyckstillverkning av komplexa produkter. För att förbättra geometrisäkringsprocessen inom rörtillverkning på Saab Kockums föreslår denna studie generella förbättringsriktlinjer och även en process samt en prototyp av ett mätverktyg med
tillämpning i passrörsmetodologin. Det specialdesignade prototyp-mätverktyg möjliggör en alternativ mätmetod som kan användas i trånga utrymmen där det är svårt att komma åt med en 3D-mätarm, som är tänkt att vara den primära mätmetoden.
Som slutsats indikerar denna studie att geometrisäkring är applicerbart inom fåstyckstillverkning av komplexa produkter och föreslår tre metoder för hur det kan åstadkommas. Dock måste var och en av metoderna vidare undersökas för att avgöra deras tillämpbarhet i andra industrier som tillverkar serier av få enheter. Vidare så visar prototyp-mätverktyget och processen för passrörsmetodologin potential för att förbättra geometrisäkringsprocessen inom rörtillverkning på Saab Kockums. Dock krävs fortsatt arbete för att färdigställa processen för passrör samt vidare utveckling av mätverktyget för tillämpning i produktion.
Nyckelord: Geometrisäkring, fåstyckstillverkning, komplexa produkter, kvalitetssäkring, robust design
iv
Acknowledgement
We would like to express our gratitude towards Saab Kockums for letting us conduct our master thesis in collaboration with them. We especially want to thank our supervisors Per Wahlsten and Per Bergström at Saab Kockums for their continuous guidance and support. Further, we would also like to thank everyone else at Saab Kockums that has supported us and patiently answered all our questions. We have always felt welcomed and appreciated.
We would also like to express our gratitude towards everyone that has agreed to be interviewed and provided us with valuable knowledge, or in any other way helped us during our thesis work. We especially want to thank professor Rikard Söderberg for providing valuable knowledge.
Lastly, we would like to thank our institute advisor professor Tobias Larsson for providing valuable insights and guidance throughout our thesis work.
Henrik Ehrenberg & Filip Malmenryd
2020-06-10
v
Table of contents
1. Introduction ... 1
Background: Geometry assurance ... 1
High volume & low volume manufacturing ... 1
Problem statement: Application of geometry assurance in low volume manufacturing of complex products ... 2
Company description: Saab Kockums ... 2
Purpose, objective & framing of research questions ... 3
Delimitations ... 3
2. Work structure ... 4
Phase 1: Analysis of current state at Saab Kockums and a study of state of the art in geometry assurance ... 4
Phase 2: State of practice in automotive and aerospace industry ... 5
Phase 3: Application at Saab Kockums ... 5
3. Methodology ... 6
Quantitative & qualitative research approach ... 6
Literature study ... 6
Design research methodology ... 7
Interviews ... 8
Prototyping ... 9
Evaluation of measurement techniques ... 9
4. Geometry assurance – Theoretical framework ... 10
Geometrical variation ... 10
Geometry assurance engineering ... 12
Concept phase ... 12
Verification (pre-production) phase ... 18
Production phase ... 18
5. Geometry assurance – State of practice ... 19
Status analysis of Saab Kockums ... 19
State of geometry assurance within the automotive and aerospace industry ... 20
Proactive mindset ... 20
Geometry assurance specialists ... 20
Measuring ... 20
6. Results ... 21
General guidelines for geometry assurance in low volume manufacturing of complex products ... 21
vi
Specifying requirements ... 21
Geometry assurance methods ... 21
Pre-production series ... 23
Bottom-up evaluation ... 23
Geometry assurance department ... 23
Application of geometry assurance in pipe manufacturing at Saab Kockums ... 24
General guidelines for pipe manufacturing ... 24
Cross-functional teams in concept phase ... 24
Computer aided tolerancing ... 24
Geometry assurance engineering specialist ... 25
Process for verification of fitting-pipe design ... 25
Prototype of measurement tool ... 28
7. Discussion ... 34
Work structure ... 34
Interviews ... 35
Geometry assurance methods ... 35
Fitting-pipe process ... 36
Prototype measurement tool ... 37
8. Conclusion & future work ... 38
General guidelines ... 38
Application in pipe manufacturing at Saab Kockums ... 39
Future work ... 40
References ... 41
Appendices ... 43
Appendix A ... 43
vii
List of figures
Figure 1: Work structure based on the design research framework [10]. ... 4
Figure 2: Design research methodology framework [10]. ... 7
Figure 3: Areas of relevance and contribution used during the literature study [10]. ... 7
Figure 4: Geometrical variation contributors [14]. ... 10
Figure 5: Example of a tolerance stack-up [15]. ... 10
Figure 6: Late discovery of problems related to quality increases costs [16]. ... 11
Figure 7: Geometry assurance process [1]. ... 12
Figure 8: Geometric tolerancing symbols [19]. ... 13
Figure 9: Example of drawing with GD&T [20]. ... 13
Figure 10: Example of tolerance allocation [1]. ... 14
Figure 11: Robust design example [15]. ... 14
Figure 12: Locating scheme using the 3-2-1 principle [1]. ... 15
Figure 13: Illustration of pipeline with a fitting-pipe, marked in green. ... 19
Figure 14: Process for verification of fitting-pipe design. ... 25
Figure 15: 3D measurement arm. ... 26
Figure 16: Measurement tool concept. ... 28
Figure 17: Physical prototype of measurement tool. ... 28
Figure 18: Chuck attaching to a premanufactured pipe. ... 29
Figure 19: Simplified 3D-printed chuck. ... 29
Figure 20: Step 1 of measurement procedure. ... 29
Figure 21: Step 2 of measurement procedure. ... 30
Figure 22: Step 3 of measurement procedure. ... 30
Figure 23: Step 4 of measurement procedure. ... 31
Figure 24: Step 5 of measurement procedure. ... 31
Figure 25: Step 6 of measurement procedure. ... 32
Figure 26: Telescopic arm enabling measurement of longer distances. ... 32
Figure 27: T-connection enabling a modular design with multiple arms. ... 33
Figure 28: Protocol of alignment between measurement data and nominal pipe design ... 43
viii
List of tables
Table 1. Each industry specialist’s relation to geometry assurance. ... 8 Table 2. Key questions used in interviews during current state analysis at Saab Kockums. ... 8 Table 3. Key questions used in interviews with industry specialists during analysis of state of practice.
... 9
ix
Nomenclature
Premanufactured pipe – Pipes that are manufactured in advance and later assembled on site Fitting-pipe – Pipes that are intended to be manufactured and mounted once the premanufactured pipes have been assembled
Spool – An individual pipe in a pipeline system
Pipeline – Composition of several spools connected in series
Flange – The rim of a pipe that for example is used to attach another pipe
List of abbreviations
GD&T – Geometric dimensioning and tolerancing RD&T – Robust design & tolerancing
CAT – Computer aided tolerancing DOF – Degrees of freedom
3D – Three-dimensional
ARC diagram – Areas of contribution and relevance diagram
DRM – Design research methodology
1
1. Introduction
This chapter presents an introduction to geometry assurance and its relevance in low volume
manufacturing, specifically in the shipbuilding industry. Furthermore, the collaboration partner Saab Kockums is introduced, along with the scope of this thesis.
Background: Geometry assurance
Geometrical variation is an unavoidable aspect of every manufacturing process that, unless
managed, may affect the ability to fulfill product requirements as nominal dimensions never can be expected. The effect of geometrical variation is further complicated when multiple components are assembled, as their individual variation may propagate. This presents the challenge of specifying and fulfilling product requirements in order to manage the effects of geometrical variation. Failure in fulfilling these requirements, such as esthetical, functional and assembly requirements, may lead to consequences such as rework, delays and bad publicity [1].
Quality problems caused by geometrical variation are often discovered in late stages of the product realization process, often during assembly or pre-production. The consequences of unmanaged geometrical variation become more difficult to manage the later it is discovered [2]. This highlights the necessity of taking proactive measures in order to manage and control geometrical variation in early stages of the product realization process [1].
Tools, processes and methods that are utilized to manage and control the effects of geometrical variation are summarized as the practice of geometry assurance. The main purpose of geometry assurance is to provide the means to create a product design that is able to withstand the effects of geometrical variation and still fulfill product requirements. Being able to control the effects of geometrical variation is the foundation for developing a predictable, reliable and cost-effective manufacturing process. Therefore, geometry assurance is an important practice for manufacturing companies in order to stay competitive [1].
High volume & low volume manufacturing
High volume manufacturing is characterized by production of high quantities of products in short time intervals. The automotive industry is an example of such an industry, where the production generally is optimized to manufacture units in series according to a cycle time. The paced production heavily relies on fulfillment of defined geometrical requirements in order to ensure successful assembly at each station within given time. This stresses the importance of geometry assurance in high volume manufacturing [1].
Low volume manufacturing is characterized by production of low quantities of products and is in general not as constrained by a paced production in contrast to high volume manufacturing.
Industries such as shipbuilding have a historical past of relying on craftsmanship during the
production and assembly of complex products [3]. Therefore, the time needed for certain activities or tasks cannot always be precisely determined in advance as in high volume manufacturing.
2
Problem statement: Application of geometry assurance in low volume manufacturing of complex products
In high volume manufacturing, geometry assurance is a well-researched area that is constantly being developed and extensively used in both the automotive industry [1], as well as in the aerospace industry [4]. But in applying cost effective geometry assurance in low volume manufacturing of complex products, the research is limited. The shipbuilding industry is an example of such an industry, where large and complex products of low quantity are manufactured [3], [5].
In recent development, the shipbuilding industry is facing increased competitiveness due to the global market demanding for example large and cost-effective ships with high quality. In order to remain competitive on the global market, shipyards have to keep up with development and the latest technology [3].
The shipbuilding industry has historically, and still to some degree, often largely relied on shipyard- workers craftsmanship to solve manufacturing problems in production. This working method risk result in extended lead-times and may introduce changes in the product’s design and function.
Thereof, units in low volume manufacturing tends to become individuals even though they are based on the same nominal design. Further, this may also lead to vulnerability as the operator possesses the competence rather than the manufacturing process. Studies indicate that a technology-intensive direction rather than a labor-intensive direction is the long-term strategy to strive towards, due to increased competitiveness within the industry [6], [7], [8]. This places high demands on the process of design and manufacturing, where geometry assurance plays a crucial role in assuring a company’s competitiveness. However, the conditions for geometry assurance in low volume manufacturing of complex products differ from its application in high volume manufacturing. A prototype or a pre- series is not necessarily feasible from a time and cost perspective in low volume manufacturing, making it more difficult to determine manufacturing capabilities prior to production [9]. This, in addition to the fact that low volume manufacturing is in general not as constrained to a paced production as in high volume manufacturing, may require a different approach to geometry
assurance in low volume manufacturing. Therefore, research on how to apply geometry assurance in low volume manufacturing of complex products needs to be further developed in general, but especially with application in the shipbuilding industry, which is the subject of this thesis.
Company description: Saab Kockums
This thesis was done in collaboration with Saab Kockums. Saab Kockums is a company specializing in naval technology for military application. Saab Kockums is a business area within the Swedish defense group Saab, which is providing products, services and solutions for both military and civil applications.
Saab Kockums is currently developing and manufacturing their next generation submarine-class A26,
which provide cutting edge stealth technology, together with the sterling air-independent propulsion
system. In Karlskrona, Sweden, Saab Kockums are continuously working on improving their modern
shipyard where the new submarine class is manufactured. Therefore, Saab Kockums are interested in
developing the research on how geometry assurance is applied in low volume manufacturing of
complex products.
3
Purpose, objective & framing of research questions
By studying how geometry assurance can be applied in low volume manufacturing of complex products, the purpose of this thesis is to improve geometrical quality throughout the product realization process in order to reduce lead times, costs and increase assembly precision.
The objective of this thesis is to develop guidelines on how geometry assurance may be applied in low volume manufacturing of complex products. In addition, the objective is to develop methods, tools and processes that can improve the geometry assurance process in pipe manufacturing at Saab Kockums.
To help achieve the objective of this thesis, the following three research questions will be examined and analyzed:
• RQ1: What differences can be identified between low- and high volume manufacturing from a geometry assurance perspective?
• RQ2: How might geometry assurance be applied in low volume manufacturing of complex products?
• RQ2.1: How might the geometry assurance process in pipe manufacturing at Saab Kockums be improved?
Delimitations
• Due to this thesis being a feasibility study, the theoretical framework of geometry assurance is limited to the basis of the practice.
• The research on state of practice in other industries is limited to companies within the
automotive and aerospace industry due to these being prominent examples of how geometry assurance is applied.
• Application at Saab Kockums is limited to pipe manufacturing.
• Due to company restrictions, specific sections are restricted for internal use at Saab Kockums.
These sections are thus not included in this report version.
4
2. Work structure
This chapter presents the work structure of this thesis.
The work of this thesis is divided into three main phases, which is illustrated in Figure 1. In phase one, a current state analysis of Saab Kockums from a geometry assurance perspective is conducted, along with a study of state of the art in geometry assurance. Phase two investigates state of practice within companies in the automotive and aerospace industry. Lastly, in phase three, guidelines are developed on how the geometry assurance process may be improved in pipe manufacturing at Saab Kockums. In addition, an improved process and prototype measurement tool for verifying nominal model against measurement data for fitting-pipes is developed.
Figure 1: Work structure based on the design research framework [10].
Phase 1: Analysis of current state at Saab Kockums and a study of state of the art in geometry assurance
This thesis was initiated with an introduction to Saab Kockums. During this introduction, the aim was to produce a current state analysis of Saab Kockums work within geometry assurance and identify an area of focus. The current state analysis was conducted through interviews with personnel from different areas associated to geometry assurance, such as R&D and industrialization. During the current state analysis, study visits were conducted in the production facility and in a submarine.
Parallel to the current status analysis, a literature study was conducted in order to study state of the art in geometry assurance. Further, the purpose of the literature study was to gather knowledge on how different geometry assurance tools work and is applied.
5
Phase 2: State of practice in automotive and aerospace industry
As a complement to the literature study, companies within the automotive and aerospace industry were studied. This was done in order to understand how successful industries in geometry assurance are applying state of the art research. Further, the purpose was to investigate similarities and
differences between high- and low volume manufacturing from a geometry assurance perspective.
This was done through a study of how their processes for geometry assurance function, in
combination with interviews with industry specialists. With this data, guidelines on how geometry assurance may be applied in low volume manufacturing of complex products could be developed.
Phase 3: Application at Saab Kockums
In this phase, the general guidelines developed from the previous two phases were used to develop guidelines on how the geometry assurance process in pipe manufacturing can be improved at Saab Kockums. A process for how the production documentation of fitting-pipes can be verified and updated prior to manufacturing was developed. Further, a prototype measurement tool was
developed to enable measurement in cramped space of the relation between premanufactured pipes where a fitting-pipe later will be mounted.
The development of the process was done through discussions and interviews with engineers at Saab Kockums. The development of the prototype was done through workshops and brainstorming sessions with engineers at Saab Kockums, where specifications and requirements were established.
6
3. Methodology
This chapter presents the methodology of how this thesis was conducted, along with an explanation and motivation of the chosen methods.
Quantitative & qualitative research approach
A quantitative research approach is characterized by gathering structured and statistical data in large quantities. This approach can for example be used in a questionnaire containing yes or no questions, where the data is statistically analyzed to test or verify a hypothesis. A qualitative research approach is characterized by rather gathering data in form of words, describing the subject through opinions, thoughts and impressions. The data is then analyzed through summarizing and interpreting. This approach often involves a smaller number of samples [11].
In this thesis, interviews with industry specialists were conducted. During these interviews each industry specialists expressed their thoughts and opinions based on questions related to how
geometry assurance can be applied in low volume manufacturing of complex products. Due to this, a qualitative approach was chosen in this study.
Literature study
Relevant literature was studied in order to determine the state of the art in geometry assurance and to build an understanding of the practice. In this regard, relevant literature refers to literature needed to create the basis for studying the feasibility of applying geometry assurance in low volume manufacturing of complex products. To organize the search for relevant literature, guidelines from the design research methodology was used. This is further explained in Section 3.3.
Throughout the course of this thesis, Google Scholar and BTH Summon were used to gain access and study literature. The studied literature consists of research articles, books and thesis work.
7
Design research methodology
The design research methodology (DRM) proposed by Blessing and Chakrabarti intends to help researchers be more successful by providing suitable methods and guidelines for achieving a structured design research process. In the DRM framework, illustrated in Figure 2, an iterative process for doing design research is presented. This iterative process advocates a methodology where initially the research clarification is established. Further studies are then performed in order to gather more in-depth knowledge within the research area. With more in-depth knowledge, the research clarification can be elaborated more thoroughly. This process is then repeated iteratively throughout the research project as the knowledge within the research area gradually improves.
Finally, this will result in a clear definition of the research objective and increase the chance for successful results [10].
The basis of the DRM framework has been used to establish the work structure of this thesis, which is presented in Chapter 2. By continuously combining state of the art and state of practice with the current state analysis of Saab Kockums, the research clarification could be determined more thoroughly. This led to a clear vision of the research objective.
Figure 2: Design research methodology framework [10].
To support the search for literature, the design research methodology presents the use of areas of relevance and contribution (ARC) diagrams. The ARC-diagram is used to identify and organize
contributing areas from the literature that may be relevant, which helps establishing a foundation for the research [10]. In this thesis, the ARC-diagram was used in order to structure and organize
relevant areas reviewed during the literature study. The relevant areas that have been studied in order to build an understanding of geometry assurance and to determine state of the art is presented in Figure 3.
Figure 3: Areas of relevance and contribution used during the literature study [10].
Geometry assurance
Quality assurance Computer Aided
Tolerancing
Robust Design Dimensional
Engineering
8
Interviews
Semi-structured interviews can be used to give each interview participant the room and freedom to express personal viewpoints and reflections on predetermined questions which promotes a
discussion [12].
Semi-structured interviews were used when conducting the current state analysis at Saab Kockums.
This interview structure was also used in interviews with three industry specialists when studying how companies within the aerospace and automotive industry apply geometry assurance. During these interviews, their viewpoints on how geometry assurance may be applied in low volume manufacturing of complex products were discussed.
A set of predetermined questions was used as key questions during each interview. Questions used during the interviews at Saab Kockums are presented in table 2. Table 3 presents the questions used during interviews with industry specialists. During each interview with the industry specialists, the questions were chosen to guide the discussion depending on what industry and company the person had experience from. In table 1, each industry specialist’s relation to geometry assurance is
presented.
Table 1. Each industry specialist’s relation to geometry assurance.
Interviewee 1 Professor in product development with extensive experience of geometry assurance within the automotive industry
Interviewee 2 Geometry assurance engineer with extensive experience of geometry assurance within the automotive and aerospace industry, currently working with geometry assurance within the shipbuilding industry
Interviewee 3 Geometry assurance engineer currently working within the aerospace industry
Table 2. Key questions used in interviews during current state analysis at Saab Kockums.
What is your association with geometry assurance?
What challenges do you see from a geometry assurance perspective?
9
Table 3. Key questions used in interviews with industry specialists during analysis of state of practice.
What is your view on possible distinctions between low- and high volume manufacturing from a geometry assurance perspective?
How might geometry assurance be approached in low volume manufacturing compared to high volume manufacturing?
What is your view on challenges in low volume manufacturing from a geometry assurance perspective?
How is the geometry assurance process structured in companies you have experience from?
Do you have experience from geometry assurance in low volume manufacturing?
Prototyping
Development of a prototype measurement tool was done through an iterative process, where need finding was performed during the current state analysis of Saab Kockums in order to specify
requirements. Further, brainstorming sessions and workshops were conducted to explore ideas and thoughts together with engineers at Saab Kockums. 3D-printing was used in order to manufacture parts that, in combinations with bought components, were assembled. The different iterations of the prototype were evaluated through functional tests and discussions with people with experience in the application area. Tests that were performed involved usability, weight and the rigidness of the prototype. These tests were carried out by letting people with experience within the application area interact with the prototype and express their thoughts. The feedback was then taken into
consideration when developing the different iteration of the prototype.
Evaluation of measurement techniques
With focus on the application within Saab Kockums, an evaluation of measuring techniques for fitting-pipes was performed. This evaluation was done by consulting a company specializing in measuring equipment. During the visit of the company, the requirements for a measurement scenario of a fitting-pipe was presented. An optimal measuring technique was suggested based on the requirements of this scenario. This measuring technique was tested in order to discuss and evaluate its potential application.
10
4. Geometry assurance – Theoretical framework
This chapter presents a description of state of the art in the research of geometry assurance, along with methods, tools and processes that supports the practice.
Geometrical variation
All manufacturing will unavoidably result in geometrical variation in the product or component, which means that nominal dimensions never can be expected [13]. The total geometrical variation originates from three different types of variation contributors; part variation, assembly variation and the robustness of the design concept, which is presented in Figure 4. Part variation is summarized as the combination of machine precision and process variation in the manufacturing process. Assembly variation is summarized as the combination of assembly precision and process variation in the assembly process. Depending on the robustness of the design concept, the design will either amplify or suppress the variation originating from parts in the assembly [14].
Figure 4: Geometrical variation contributors [14].
Variation from different components can in assemblies result in tolerance stack-ups where variation from each component is amplified. These tolerance stack-ups can, unless managed, result in the final product not fulfilling its requirements, or risk unsuccessful assembly [15]. An example of a tolerance stack-up is illustrated in Figure 5, where each components variation can result in an unknown variation in the assembly.
Figure 5: Example of a tolerance stack-up [15].
11
Problems related to geometrical variation is often discovered late in the product realization process.
As presented in Figure 6, quality problems that require change, modification or rework is often related to high costs [16]. In general, the later quality problems related to geometrical variations are discovered, the more serious the consequences become. In addition to high costs, it also causes bad publicity and delays [1].
Figure 6: Late discovery of problems related to quality increases costs [16].
12
Geometry assurance engineering
Leading research of geometry assurance proposes a geometry assurance process divided into three phases; concept-, verification- (pre-production) and production phase, which is illustrated in Figure 7.
A combination of activities applied in each phase contributes to minimizing the effects of geometrical variation. These activities refer to tools, methods and processes [1].
Figure 7: Geometry assurance process [1].
Concept phase
In the geometry assurance process, the concept phase is a key factor in managing the effects of geometrical variation throughout the product realization process. The objective is to develop a concept design that is able to withstand the effects of geometrical variation in the manufacturing process. By focusing on geometry assurance early in the product realization process, requirements can be defined and optimized into a robust design to ensure that the final product will fulfill them.
Requirements are ideally defined on an overall product level and successively broken down into component level. This way of defining requirements is referred to as a top-down approach [1].
13 4.2.1.1 Geometric dimensioning & tolerancing
Geometric dimensioning and tolerancing (GD&T) describe acceptable intervals in which the
geometrical variation may deviate from the nominal value, while still fulfilling requirements. GD&T is used to describe the product or component and its requirements in relation to form, orientation, location and runout [17]. Datum references that are established by locating schemes is a key factor in mediating GD&T requirements. This enables a basis for repeatable verification of requirements later in the product realization process [18]. Locating schemes are further explained in Section 4.2.1.4.
Examples of symbols used to describe geometric tolerances are presented in Figure 8. In figure 9, an example of how these symbols in combinations with datum points and locating schemes describes a component is presented.
Figure 8: Geometric tolerancing symbols [19].
Figure 9: Example of drawing with GD&T [20].
14 4.2.1.2 Tolerance allocation
Tolerances are ideally defined according to the top down approach, where product requirements are divided into part tolerances, illustrated in Figure 10 [1]. When allocating tolerances, this is done with respect to requirements, cost and manufacturing capabilities. As tight tolerances generally are associated with increased costs, it is important to define good enough tolerances for each
component, while still fulfilling the requirements of the system or product. Different strategies can be used in order to achieve an optimized balance between cost-effectiveness and producibility when allocating tolerances [2]. Simulation tools can be used in order to aid tolerance allocation, further described in Section 4.2.1.5.
Figure 10: Example of tolerance allocation [1].
4.2.1.3 Robust design
The effects of geometrical variation are optimally managed as early as possible in the product realization process by optimizing the robustness of the design concept. A robust design is characterized as being insensitive to the total variation originating from the different types of variation contributors, previously illustrated in Figure 4 [14].
A robust design, in comparison to a sensitive design suppresses the amplification of geometrical variation. An example of a robust design is illustrated in Figure 11. Depending on where the support (X) of the beam is placed, the output variation is either suppressed or amplified by the input
variation. Moving the stand to the left results in amplification of the output variation, whereas moving the stand to the right suppresses the output variation, illustrating a more robust design [15].
A robust design allows for wider input variation, which helps to balance cost-effectiveness with producibility in tolerance allocation by for example enabling a cheaper manufacturing method [14].
Figure 11: Robust design example [15].
15 4.2.1.4 Locating schemes
Locating schemes are used to lock a part or assembly in its six degrees of freedom (DOF). A locating scheme is created by positioning locating points, called locaters. Locaters may be strategically positioned in order to reduce the amplification between input and output variation. This means that part and assembly variation is suppressed in a way that allows for wider input variation, which improves the robustness and further helps balancing cost-effectiveness with producibility. The increased robustness is a key factor in managing how variation propagates between individual parts, in order to ensure successful assembly [21].
Generally, the robustness is increased as the locaters are spread out as much as possible. However, the difficulty of strategically positioning locaters increases with the complexity of the product design, as there are several ways of locking a part in its six degrees of freedom. The orthogonal 3-2-1 locating scheme is a commonly used method for locating rigid parts. However, there are different types of locating schemes depending on the design of the product [22]. The 3-2-1 orthogonal locating system is illustrated in Figure 12.
Figure 12: Locating scheme using the 3-2-1 principle [1].
16
The six locaters are positioned in the following order to lock the part using the 3-2-1 locating scheme principle [1]:
• The locaters A1, A2 and A3 are referred to as the primary set of locating points. These locaters are positioned to control the following three degrees of freedom;
- Translation in Z-axis
- Rotation around X and Y-axis
• The locaters B1 and B2 are referred to as the secondary set of locating points. These locaters are positioned to control the following two degrees of freedom;
- Translation in X-axis - Rotation around Z-axis
• The locater C1 is referred to as the tertiary locating point. This locater is positioned to control the last degree of freedom;
- Translation in Y-axis
The positioning of the three groups of locaters defines three orthogonal planes [1]:
• The positioning of the primary group of locaters define the primary locating plane, A
• The positioning of the secondary group of locaters define the secondary locating plane, B.
Plane B is oriented perpendicular to plane A
• The positioning of the tertiary locater defines the tertiary locating plane, C. This plane is oriented perpendicular to both plane A and B
Söderberg et al. [1] presents what is referred to as the main rule of geometry assurance. This rule states that the same locating schemes should be used in manufacturing, inspection and production to the greatest extent possible. This minimizes the risk of introducing additional variation.
17 4.2.1.5 Simulation tools
In the aim of achieving and optimizing a robust design concept, CAT software like RD&T can be a very useful tool. Software like RD&T can provide functions like locating schemes optimization, stability analysis, statistical simulation analysis, contribution analysis and inspection preparation. These simulation tools allow for the design concept to be optimized from a robustness perspective, even before making any physical prototype [23].
Locating scheme optimization can be used in order to optimize the positioning of the locators in the locating scheme. This reduces amplification from input variation to output variation. When
performing the analysis, the RD&T software will highlight areas were output variation is amplified though color coding, creating easy visualization of problem areas. Stability analysis can then be used to evaluate the amplification caused by the locators and component design. This illustrates the amplification of variation in X, Y and Z direction in a number of points on the component [2].
With statistical simulation analysis, geometrical variation in assemblies can be simulated and evaluated. By determining tolerance widths and expected distribution for every part, simulation is often performed using a Monte Carlo model that randomly generates input numbers. The robustness and required tolerances can then be evaluated. By performing the simulations for a certain amount of iterations, a normal distribution curve can be produced where estimations can be made whether how likely it is for the requirements to be met and achieve successful assembly. The data used for simulation can also be replaced or complemented with actual measurement data. This allows for even more accurate simulations, or assembly verification through virtual pre-assembly [2].
Inspection preparation aims at optimizing the amount of inspection points to be used in order to verify requirements during production. During pre-production a large amount of inspection point are gathered in order to detect problems and make adjustments [2].
18
Verification (pre-production) phase
During the verification phase, the developed design concept is produced and tested in order to evaluate its ability to withstand the effects of geometrical variation during the manufacturing process. The main focus of this phase is to verify the fulfillment of the product requirements, and if necessary, the product design and tolerances can be adjusted. Along with verifying fulfillment of product requirements, the production system is also tested and adjusted in preparation for full scale production. During the verification (pre-production) phase the inspections-strategies are established, evaluated and determined. The purpose of the inspection preparation is to verify that product requirements are met during the manufacturing and assembly process. In the inspection preparation, the aim is to determine optimal inspection points that verify if the product fulfills its requirements or not. These inspections point can also be used to collect information about the manufacturing process which is used to adjust the product if needed. A large amount of inspection points is typically
collected during the verification (pre-production) phase to collect as much information as possible.
The information gathered from the inspection points are then used to monitor the manufacturing process when full scale production is launched [2].
Production phase
When the design concept has been optimized in terms of its geometrical robustness and the production system is sufficiently adjusted, full scale production is ready to be launched. As the manufacturing process is launched, continuous monitoring and control of the process is prioritized in order to quickly detect and correct errors [2]. Detection of problems related to geometrical variation in this phase can result in serious consequences. This stresses the importance of extensive work in the concept- and verification (pre-production) phase to ensure fulfillment of product requirements [1].
19
5. Geometry assurance – State of practice
Below follows a presentation of Saab Kockums current status from a geometry assurance perspective.
Further follows a presentation of how geometry assurance is applied within the automotive and aerospace industry. Examples are presented from companies that are successful in implementation of state of the art research in geometry assurance.
Status analysis of Saab Kockums
Saab Kockums are currently manufacturing their new submarine class A26 in their modern shipyard.
In conjunction to this, the aim is to increase the number of premanufactured parts. In pipe manufacturing, the vision is to premanufacture pipes to the greatest possible extent, the non- premanufactured pipes are referred to as fitting-pipes. These fitting-pipes are intended to be manufactured and mounted once the premanufactured pipes have been assembled. This enables a process for managing eventual geometrical variation in pipe systems by correcting the fitting-pipe prior to manufacturing and thereby ensuring successful assembly. In Figure 13, a pipeline is illustrated where the chosen fitting-pipe is marked in green. The process for fitting-pipes and the definition of what pipes that are most suitable to function as fitting-pipes is continuously being optimized. The current guidelines for fitting-pipes is a simple geometry, with preferably only two connections and not more than one bend.
Figure 13: Illustration of pipeline with a fitting-pipe, marked in green.
Premanufacturing of pipes places high demands on the geometry assurance process in order to ensure successful assembly, as there are a lot of pipes in a submarine. Manufacturing pipes for a submarine with military application also possesses challenges due to very cramped spaces in
combination with strict requirements. Therefore, Saab Kockums is continuously trying to improve the geometry assurance process to ensure a high quality product.
20
State of geometry assurance within the automotive and aerospace industry
Proactive mindset
Geometry assurance has in general come a long way within the automotive and aerospace industry.
Companies like Volvo Cars, GKN Aerospace, Scania and Saab Aeronautics are examples of companies that are working extensively with geometry assurance, but also pushing the research forward. A clear link between these companies is their proactive mindset, where they aim to prevent problems before they occur and make sure to do right from the start. As a result of this mindset, they are working with a top-down work structure when specifying requirements in the product development process [4], [24]. This means that functional, esthetical and assembly requirements are specified for the final product early in the concept phase, which are then broken down into system and
component requirements. Tolerance allocation is performed, where GD&T are determined based on the requirements, cost and producibility. Great focus is also on developing a robust design which aims to be as non-sensitive to variation as possible.
Geometry assurance specialists
A common working method within these companies is that they have geometry assurance specialists, although in different constellations, to support the geometry assurance process. Volvo Cars for instance, have so-called GAE (Geometry Assurance Engineer) [2], [4]. The purpose of these specialist roles is to support work regarding geometry assurance in every phase of the product realization process. Saab Aeronautics are working in a similar way, where they have a department that supports with competence regarding geometry assurance, such as specifying GD&T, tolerance allocation and performing statistical variation simulations [4].
Measuring
Measurement is another key success factor within the leading companies, where measurement data is continuously collected and saved. The reasons for collecting measuring data are several. The measurements are partly done to verify that the product or component fulfill its requirements. To do this, GD&T are specified in the concept phase, and from which the requirements can be measured and verified repeatedly. The measurement data is then stored to be used in variation simulations in order to further enhance the design, but also to optimize locators to achieve a more robust design.
As a conclusion from observing these companies in the automotive and aerospace industry it can be determined that they share a similar mindset on how to work with geometry assurance and that they are applying state of the art methods, tools and processes for doing so, which has proven to be of great success.
21
6. Results
This chapter presents general guidelines for how geometry assurance may be applied in low volume manufacturing of complex products, along with application in pipe manufacturing at Saab Kockums.
General guidelines for geometry assurance in low volume manufacturing of complex products
Specifying requirements
A product should satisfy customer needs. Customer needs should thereby determine the product requirements. Within geometry assurance, these requirements form the basis for all other
requirements on the product [1]. By using the top-down approach, product requirements are broken down into sub-systems and part requirements, where every stakeholder agrees on what
requirements are needed. Geometric dimensions and tolerances are then allocated, specifying the allowed variation while still achieving the requirements. When specifying requirements, GD&T are crucial in order to mediate the geometric tolerances, but also to ensure repeatable measurement and verification of the requirements [18]. Therefore, parallel to specifying the requirements, reference system for the final product and parts are defined. This method is used in both the automotive and aerospace industry and has proven to be of great success [1], [2].
Geometrical variation in manufacturing processes is unavoidable, regardless of production volume, and the purpose of satisfying customer needs remain the same. Therefore, the results of this study indicate that working proactively is equally important and applicable in low volume manufacturing of complex products. This means for example specifying requirements according to the top-down approach, defining reference systems and applying geometric dimensions and tolerances on drawings early in the concept phase.
Geometry assurance methods
With a robust design, it is possible to achieve a stable design insensitive to input variation [14]. This becomes crucial in high volume manufacturing, where there is little room for modification or correction during production. Everything has to be ready for assembly and made sure to fit on the first try. This places high demands on the geometry assurance process, where for example extensive statistical variation simulations previous to production are required. This differs from low volume manufacturing, where in some cases it is possible to adjust and modify in production, which thereby enables different methods for geometry assurance. Simulation, although a useful tool in low volume manufacturing, may not be economically feasible to the same extent as in high volume
manufacturing. If a robust design is not possible to achieve in low volume manufacturing, it is crucial to know how much geometrical variation that might occur. This is especially important in large complex systems where the tolerance stack-ups may cause severe problems in fulfilling
requirements. Therefore, each design element needs to be evaluated based on the consequences of
the outcome due to geometrical variation. This will then determine the course of action in order to
geometry assure the design.
22
The result of this study proposes three methods on how geometry assurance can be applied in low volume manufacturing of complex products. The first method is based on using simulation tools to ensure a high probability of fulfillment of requirements. In this method, it is essential to have good knowledge about the manufacturing capabilities and what tolerances and requirements that are possible to maintain. However, as mentioned earlier, this method may not be economically feasible for the entire products, but rather for specific systems or certain interfaces where for example the consequences of not fulfilling the requirements are serious.
Another possible method is by purposely awaiting manufacturing of different components in order to measure and modify the design based on the turnouts of other components. This method can for example be appropriate when the knowledge of either the manufacturing process or the assembly variation is insufficient. When using this method, it is crucial to establish a plan for how components will be manufactured depending on the different turnouts of other components. This is necessary in order to ensure fulfillment of the requirements and to achieve predictability throughout the
manufacturing process. When developing this plan, it is important to involve the craftsmanship and experience that is often found in low volume manufacturing in order to verify the feasibility of the plan.
A third method may be where simulation is used to ensure a certain probability of fulfilling
requirements and ensuring successful assembly. This method enables a balance between cost and a certain probability of successful assembly. This means that the cost of for example tighter geometric dimensions and tolerances needed for fulling a certain level of probability is balanced with the cost of reworking or modifying the components that falls outside of the statistical likelihood, and thereby not fulfilling the requirements. However, in this method it is still important to have a plan for how to solve the statistical exceptions. This method is therefore more likely suited for components where the consequences of not fulfilling requirements are relatively mild, and where the cost of rework or modification does not exceed the cost of higher probability of fulfilling requirements and thereby for example tighter tolerances.
23
Pre-production series
The automotive and aerospace industries are applying the verification (pre-production) phase as an important activity to evaluate the design concept before launching to full scale production. This is typically done with a pre-series, where for example verification of inspections strategies and further optimization of locating schemes are performed.
However, this type of verification might not be possible to the same extent in low volume
manufacturing, as a pre-series production often is not possible from a time and cost perspective. This presents a challenge in low volume manufacturing, especially when manufacturing a product that share little to none similarities to previously produced products. Based on this knowledge, and discussion with industry specialists, a geometry assurance process for low volume manufacturing is dependent on gaining knowledge of what requirements are possible to achieve prior to transitioning to production phase. This data should therefore be achieved through testing, estimation or be gathered from similar previously produced products or components. If neither of these options is possible, nor doesn’t provide sufficient data, this motivates the need to measure and learn from the first product or products in order to facilitate production of future units. This may result in the first unit being more expensive and time consuming, but if the basis for learning and gathering knowledge isn’t established early, this risk result in higher costs overall as mistakes may be repeated.
Bottom-up evaluation
A benefit from measuring manufactured components is the possibility to pre-assemble components in order to determine the geometrical variations, which is referred to as a bottom-up evaluation. This can help discover problems related to geometrical variation before final assembly and thereby minimize the consequences. This method can also be used in order to evaluate if components, either manufactured in-house or by sub-contractors, that has fallen out of tolerance still can be used.
Geometry assurance department
Looking at companies within the automotive and aerospace industry, it can be concluded that companies that are successful in their work with geometry assurance often have a specific
department or specialist focusing on geometry assurance [1], [4]. The purpose of this department is to support the product realization process in aspects regarding geometry assurance. Examples of this can be to support work within specification of requirements, setting and optimizing locating
schemes, specifying GD&T and performing statistical variation simulations. The department holds an important role in the concept phase and in the early stages of the product realization process, due to this being the most crucial part in the geometry assurance process. The department will further support later in the product realization process with additional optimization of the design and for example performing bottom-up evaluations.
Even if it may differentiate in some aspects on how geometry assurance is applied in low volume manufacturing of complex products, the result of discussions with industry specialists indicate that this type of specialist competence is necessary regardless of production volume. The difference from high volume manufacturing may therefore be how many specialists that are needed, not the
question of their existence.
24
Application of geometry assurance in pipe manufacturing at Saab Kockums
General guidelines for pipe manufacturing
As mentioned in Section 6.1.1, a key factor for geometry assurance is the top-down approach, where requirements and further tolerances are specified based on the requirements of the final product, requirements such as functional, assembly or esthetical. In pipe manufacturing at Saab Kockums, this translates to specifying the requirements of the pipelines and also the individual spools, based on the final requirements of the submarine and its sub-systems. These requirements are then broken down into geometric dimensions and tolerances for each pipeline and individual spool that are required for fulfilling the requirements. These requirements are then communicated through GD&T on each drawing of the spools.
Cross-functional teams in concept phase
Along with specifying the requirements and GD&T, all involved stakeholders for the pipes need to be involved early in the concept phase in cross-functional teams. Departments like production
engineering, quality control and manufacturing engineering outfitting should sign off on a design in order to achieve predictability throughout the product realization process of the pipes. These stakeholders need to make sure that the design is able to be produced, assembled and verified, based on the requirements.
Production engineering will have to verify that the design is possible to manufacture based on the specified requirements. If there are required geometric dimensions or tolerances that cannot be met, the design must be modified. If production is not able to manufacture within the geometric
dimensions and tolerances, or does not know what can be achieved, this must be tested or estimated.
If there is insufficiency in what requirements regarding what geometric dimensions and tolerances that is possible to produce, it becomes important to measure what is being manufactured. By collecting measuring data, Saab Kockums can further extend the knowledge about what geometric dimensions and tolerances that are possible to achieve. Further, through the process for fitting-pipe, that is presented in Section 6.2.5, it can be optimized what types of pipes that are the most suitable to function as fitting-pipes, and how much geometrical variation the fitting-pipe has to manage.
Computer aided tolerancing
In the process of tolerance allocation and designing the pipelines, CAT software should be used in order to optimize the robustness of the pipeline and its tolerances. By using CAT software, simulation can be performed in order to evaluate the tolerance needed to fulfill specified requirements, but also on how sensitive the design is to variation and get feedback on how to increase the robustness of the pipelines. This type of simulation may not be possible to perform on every pipeline from a time and cost perspective but should initially be done on examples that represent a majority of cases in order to gain knowledge of the obtained geometrical variation in pipelines.
25
Geometry assurance engineering specialist
Design engineers cannot be asked to know everything regarding geometry assurance, as they already have a lot to take into account when designing. Therefore, it would be beneficial to establish a team that focuses on supervising and supporting the geometry assurance process. This team, consisting of one or more geometry assurance engineers, would play a crucial role in ensuring a robust and geometry assured design. The geometry assurance team would help specify GD&T, create library of reference systems and support with for example statistical variation simulations and evaluation.
Process for verification of fitting-pipe design
According to the fitting-pipe methodology, described in Section 5.1, the space where the fitting-pipe is to be mounted needs to be measured and compared to the nominal design prior to manufacturing of the fitting-pipe. In Figure 14, a developed proposal for how this process can function is presented.
When the premanufactured pipes are assembled, the purpose of this process is to measure the relation between the flanges and compare this to the nominal model of the fitting-pipe. It is then determined whether the nominal design of the fitting-pipe can be used, or if it needs to be modified.
Below follows a detailed description of this process, how it is thought to be performed and what is left to be determined. This process is a further development of the process for fitting-pipes.
Figure 14: Process for verification of fitting-pipe design.
