,
STOCKHOLM SWEDEN 2019
Flow Optimisation for Improved
Performance of a Multivariant
Manufacturing and Assembly Line
ALEXANDER SAEVAR GUDBJÖRNSSON
HAIDER MOHAMMED YASIN
KTH ROYAL INSTITUTE OF TECHNOLOGY
Multivariant Manufacturing and Assembly Line
Alexander Sævar Guðbjörnssonn Haider Mohammed Yassin
KTH
Royal Institute of Technology
Industrial Engineering and Management
Preface
This master thesis is the final part of the two year master program in Production En-gineering and Management at the KTH Royal Institute of Technology, Stockholm. The project is equivalent of 30 ECTS and was completed during the spring term of 2019.
Acknowledgements
We would like to thank our supervisor, professor Daniel Semere who gave us valu-able input and guidance during the process of our work. Our supervisor at Stoneridge
Electronics, Mikael Sterner receives our great appreciation for providing us with the
op-portunity to work on this research project and his support and assistance throughout the thesis work. Our co-workers, including management and operators, at Stoneridge
Electronics receive gratitude for all the help we have received for the duration of the project. They answered all questions we had and provided us with key points when it came to improving the manufacturing line. Special thanks go to Jerzy Mikler for his invaluable input while creating the ExtendSim model.
Abstract
Stoneridge, Inc. is an independent designer and manufacturer of highly engineered
electrical and electronic components, modules and systems principally for the auto-motive, commercial vehicle, motorcycle, agricultural and off-highway vehicle mar-kets. A subsidiary of Stoneridge, Inc. is the company Stoneridge Electronics. They spe-cialise in instrument clusters and tachographs which are manufactured in high quan-tity in their production plant in Örebro, Sweden. This master thesis focuses on the production line of an instrument cluster called Angela. In close collaboration with
Stoneridge Electronics, the goal was to find ways to improve the output of the Angela
line by at least 10% compared to the three best months in terms of output from the year before.
The Angela line was analyzed thoroughly and from different perspectives using lean tools such as value stream mapping, spaghetti diagram and continuous improve-ment. Finally, the simulation software ExtendSim was used in order to simulate and analyse different suggestions. The results show that various steps can be taken to im-prove the efficiency and output of the manufacturing line by as much as 16.3%.
Due to the fact that other production lines within the production are similar to the one that the project was carried out on, the project results could be applicable for the other lines as well.
Keywords: Value Stream Map, Spaghetti Diagram, Discrete Event Simulation, Waste,
Sammanfattning
Stoneridge, Inc. är en oberoende designer och tillverkare av högteknologiska
elek-triska och elektroniska komponenter, moduler och system huvudsakligen för fordon-smarknaderna. Ett dotterbolag till Stoneridge, Inc. är företaget Stoneridge Electronics. De är specialiserade på instrument kluster och färdskrivare som tillverkas i produk-tionsanläggning i Örebro. Denna examensarbete fokuserar på produktionslinje av ett instrumentkluster som heter Angela. I nära samarbete med Stoneridge Electronics, målet var att hitta sätt att förbättra produktionen av Angela linje med minst 10 % jäm-fört med de tre bästa månaderna när det gäller produktion från året innan.
Glossary xv
Acronyms xvi
1 Introduction 1
1.1 Background . . . 1
1.2 Problem Statement . . . 2
1.3 Aim & Objectives . . . 2
1.4 Delimitations . . . 2 1.5 Product . . . 3 1.6 Production Process . . . 4 1.6.1 Common IC . . . 4 1.6.2 Final Assembly . . . 7 1.7 Outline . . . 8 2 Methodology 9 2.1 Scientific Approach . . . 9 2.2 Case Study . . . 10
2.2.1 Information & Data Collection Methods . . . 10
2.2.2 Time . . . 10 2.2.3 AXXOS . . . 10 2.2.4 QlikView . . . 11 2.3 Literature Review . . . 11 2.4 Data Analysis . . . 12 2.5 Interviews . . . 12 3 Literature Review 13 3.1 History of Lean Manufacturing . . . 13
3.2 Principles of Lean Manufacturing . . . 14
3.4 Lean Tools . . . 18
3.4.1 5S . . . 18
3.4.2 Total Productive Maintenance (TPM) . . . 19
3.4.3 Just in Time (JIT) . . . 19
3.4.4 Poka-Yoke (Error Proofing) . . . 20
3.4.5 Kaizen (Continuous Improvement) . . . 20
3.4.6 Standardized Work . . . 20
3.4.7 Kanban . . . 20
3.4.8 PDCA . . . 21
3.4.9 Five Why’s . . . 21
3.4.10 SMED (Single-Minute Exchange of Dies) . . . 22
3.5 Time . . . 22 3.5.1 Process Time (PT) . . . 22 3.5.2 Cycle Time (CT) . . . 22 3.5.3 Throughput Time . . . 23 3.5.4 Takt Time . . . 23 3.5.5 Lead Time (LT) . . . 23 3.5.6 Set Up Time . . . 23 3.6 Assembly . . . 24 3.7 Bottleneck . . . 25 3.7.1 Bottleneck Identification . . . 25
3.8 Value Stream Mapping (VSM) . . . 25
3.8.1 Current state . . . 26
3.8.2 Internal mapping . . . 27
3.8.3 Waste . . . 27
3.8.4 Future state . . . 28
3.9 Spaghetti Diagram . . . 28
3.10 Assembly Line Balancing . . . 28
3.10.1 Assembly line . . . 28
3.10.2 Line balancing . . . 29
4 Discrete Event Simulation 30 4.1 Discrete Event Simulation . . . 30
4.2 ExtendSim . . . 30 4.2.1 Final Assembly . . . 31 4.2.2 Common IC . . . 36 5 Results 41 5.1 Time study . . . 41 5.1.1 Common IC Soldering . . . 41 5.1.2 Common IC Coating . . . 44 5.1.3 Final Assembly . . . 46 5.2 VSM . . . 48
5.2.2 Current State Final Assembly . . . 49 5.3 Spaghetti Diagram . . . 49 5.3.1 Common IC . . . 49 5.3.2 Final Assembly . . . 51 5.4 ExtendSim . . . 51 5.4.1 Common IC . . . 52 5.4.2 Final Assembly . . . 55
5.5 Current State Analysis . . . 58
5.5.1 Common IC Analysis . . . 58
5.5.2 Final Assembly Analysis . . . 62
5.6 Suggested Improvements & Analysis . . . 65
5.6.1 Common IC . . . 65
5.6.2 Final Assembly . . . 69
6 Recommendations & Conclusion 76 6.1 Recommendations . . . 76 6.1.1 Final Assembly . . . 77 6.1.2 Common IC . . . 79 6.2 Conclusion . . . 81 References 82 A Extra 85 A.1 Time Study . . . 85
A.1.1 Common IC . . . 85
A.1.2 Final Assembly . . . 89
B Extra 2 93 B.1 VSM . . . 93 B.1.1 Common IC . . . 93 B.1.2 Final Assembly . . . 93 C Extra 3 113 C.1 Spaghetti Diagram . . . 113 C.1.1 Common IC . . . 113 C.1.2 Final Assembly . . . 114 D Extra 4 115 D.1 ExtendSim . . . 115 D.1.1 Common IC . . . 115 D.1.2 Final Assembly . . . 115 E Extra 5 123 E.1 Photos . . . 123
1.1 Angela Instrument Cluster . . . 4
1.2 Common IC process flow . . . 5
1.3 Final assembly process flow . . . 7
4.1 First steps of simulation . . . 31
4.2 Random Number Block Properties . . . 32
4.3 Activity Block . . . 32
4.4 Lookup Table . . . 32
4.5 Lookup Table . . . 33
4.6 Operators . . . 33
4.7 Return Racks . . . 34
4.8 Bring front casing . . . 34
4.9 Batch and Unbatch for Transport . . . 34
4.10 Control area . . . 35
4.11 Shift Blocks . . . 35
4.12 Exit Blocks for Final Assembly . . . 36
4.13 Get and Set Blocks . . . 36
4.14 Random and Value Block Properties . . . 37
4.15 Get SMD rack . . . 37
4.16 Stepper Motor Assembly and Changeover . . . 37
4.17 After Stepper Batch and Unbatch . . . 38
4.18 Exit Blocks for Common IC Pieces . . . 38
4.19 Shift Blocks, Resource Pool Block and Stats . . . 39
4.20 Activities Linked to Stop in Production . . . 39
4.21 Lunch Schedule . . . 39
4.22 Shutdown Block . . . 40
5.1 Angela 1 Soldering PTs . . . 42
5.2 Angela 1 Soldering Further Analysis . . . 42
5.3 Solder Comparison . . . 43
5.4 Solder LT . . . 43
5.5 Angela Coating . . . 44
5.6 Angela Coating Further Analysis . . . 45
5.7 Coating Comparison . . . 45
5.9 PT for Angela 01 . . . 47
5.10 PT for Angela Buss variant . . . 47
5.11 Spaghetti Diagram Common IC . . . 50
5.12 Spaghetti Diagram Final Assembly . . . 51
5.13 Plotter for Current Common IC Output . . . 55
5.14 Production plotted . . . 58
5.15 Comparison of Solder LT . . . 58
5.16 Comparison of Coating LT . . . 59
5.17 Total LT for Common IC . . . 60
5.18 Automatic vs. Manual Soldering . . . 61
5.19 Automatic vs. Manual Soldering . . . 62
5.20 PT for all Angela variants . . . 63
5.21 PT for Angela truck variants . . . 63
5.22 Reduced PT . . . 70
5.23 Balanced line . . . 70
5.24 Meeting the demand . . . 75
6.1 Lean pick chart . . . 78
6.2 Pick Chart for Common IC . . . 80
A.1 Angela 9 Soldering Process Time . . . 85
A.2 Angela Bus Soldering Process Time . . . 86
A.3 Angela 9 Further Analysis . . . 86
A.4 Angela Bus Soldering . . . 87
A.5 Angela 1 Detailed Process Times . . . 87
A.6 Angela 9 Detailed Process Times . . . 87
A.7 Angela Bus Detailed Process Times . . . 87
A.8 Marianne Detailed Process Times . . . 88
A.9 Desiree Detailed Process Times . . . 88
A.10 Angela Coating . . . 88
A.11 Marianne Coating . . . 88
A.12 Desiree Coating . . . 89
A.13 Broken up process time 900597/09 Whit Out ODO . . . 89
A.14 Process time 900597/09 Whit Out ODO . . . 90
A.15 Broken up process time 900598/02 Buss . . . 90
A.16 Process time 900598/02 Buss . . . 91
A.17 Broken up process time 900597/01 WITH ODO . . . 91
A.18 Process time 900597/01 WITH ODO . . . 92
B.1 Common IC Current State Angela 1 . . . 94
B.2 Common IC Current State Angela 9 . . . 95
B.3 Common IC Current State Angela Bus . . . 96
B.4 Common IC Current State Desiree . . . 97
B.5 Common IC Current State Marianne . . . 98
B.7 Common IC Future State Angela 9 . . . 100
B.8 Common IC Future State Angela Bus . . . 101
B.9 Common IC Future State Marianne . . . 102
B.10 Common IC Future State Desiree . . . 103
B.11 Final Assembly current state Buss variant . . . 104
B.12 Final Assembly current state ODO variant . . . 105
B.13 Final Assembly current state without ODO variant . . . 106
B.14 Final Assembly Future state ODO variant IMP1 . . . 107
B.15 Final Assembly Future state With out ODO variant IMP1 . . . 108
B.16 Final Assembly Future state ODO variant IMP3 . . . 109
B.17 Final Assembly Future state With out ODO variant IMP3 . . . 110
B.18 Final Assembly Future state ODO variant IMP5 . . . 111
B.19 Final Assembly Future state With out ODO variant IMP5 . . . 112
C.1 Spaghetti Diagram Common IC . . . 113
C.2 Spaghetti Diagram Final assembly between 09:02-10:45 . . . 114
C.3 Spaghetti Diagram Final assembly between 11:57-14:00 . . . 114
D.1 First Part of Common IC ExtendSim model . . . 116
D.2 Second Part of Common IC ExtendSim model . . . 117
D.3 Third Part of Common IC ExtendSim model . . . 118
D.4 Fourth Part of Common IC ExtendSim model . . . 119
D.5 Fifth Part of Common IC ExtendSim model . . . 120
D.6 Part one Final assembly . . . 121
D.7 Part two Final assembly . . . 122
E.1 First station in the Final assembly line ODO-station . . . 123
E.2 Second station in the Final assembly line Display-station . . . 124
E.3 Third station in the Final assembly line Mech-station . . . 125
E.4 Fourth station in the Final assembly line Vibration-station . . . 126
E.5 Fifth station in the Final assembly line Calibration-station . . . 127
E.6 Sixth station in the Final assembly line Slutmontage-station . . . 128
E.7 Seventh station in the Final assembly line Screw-station . . . 129
E.8 Eighth station in the Final assembly line APPtest-station . . . 130
E.9 The Angela Final assembly line . . . 131
E.10 Angela Variant with ODO-screens . . . 132
E.11 Angela buss Variant . . . 133
E.12 Angela Variant without ODO-screens . . . 134
E.13 Pallet of finished products going to the warehouse . . . 135
E.14 Solder Load . . . 136
E.15 HMT . . . 137
E.16 Solder Inspection . . . 138
E.17 Coating Load . . . 139
E.18 Coating Inspection . . . 140
5.1 Number of HMT pieces attached to Angela variants . . . 44
5.2 Table that shows VA and NVA times . . . 48
5.3 Table that shows VA and NVA times . . . 49
5.4 Table that shows wasteful movement out of Common IC . . . 50
5.5 Table that shows ratio of production . . . 52
5.6 Batch size of each product . . . 52
5.7 Analysis of shutdown in Common IC . . . 53
5.8 Utilization of workstations in current state Common IC . . . 54
5.9 CT and Output . . . 54
5.10 Table that shows ratio of production in Final Assembly . . . 55
5.11 Analysis of shutdown in Final assembly . . . 56
5.12 Utilization of workstations in current state Final Assembly . . . 57
5.13 CT and Output . . . 57
5.14 CT if only one product at a time in soldering . . . 59
5.15 CT if only one product at a time in coating . . . 60
5.16 Comparison of output and CT . . . 60
5.17 Utilization of operators in current state . . . 62
5.18 Comparison of different states in Common IC with current shift plan . . 66
5.19 Comparison of different states in Common IC with one worker . . . 67
5.20 Comparison of current and future state . . . 67
5.21 Utilization of workstations in current state Common IC . . . 68
5.22 Table that shows comparison on VA and NVA times . . . 69
5.23 Current State CT and Output . . . 69
5.24 CT and Output after line balancing . . . 71
5.25 CT and Output after removing the vibration station . . . 71
5.26 CT and Output after balanced line with vibration station removed . . . . 71
5.27 CT and Output after introducing the stepper machine into the balanced line . . . 72
5.28 CT and Output after balancing stepper . . . 72
5.29 Station Utilization based on Improvement suggestions . . . 73
5.30 Improvement suggestions compared to each other and the current state 74 5.31 Table that shows VA and NVA times . . . 75
bottleneck Bottleneck can be described as a resource whose capacity is less than the
demands placed upon it, a process that limits throughput or any operation that limits output [1]. 10
cycle time The total time from the beginning to the end of a process. Cycle time
in-cludes process time, during which a unit is acted upon to bring it closer to an output, and delay time, during which a unit of work is spent waiting to take the next action. 3
just in time Just-in-time manufacturing system. In a full JIT system, the only parts
that enter a plant or move from process to process in a plant are those identified uniquely with a final product, no more or no less. Thus, every part being sup-plied and every part in the plant can be related directly to a bill of material of a product that is either in production or will shortly to be in production. 2
lead time Time that is required to fill an order or meet customer demand. 2
value added Denotes the "value" added to the materials received by a plant in the
plants operations. Value added can be a combination of true value and the non-value added work done in manufacturing a product. Best practice requires a plant to continually assess which of its activities is true value added and elimi-nate or reduce the non-value added activities. 15
work in progress Inventory consisting of products that are in a semi-finished state. 2
CT Cycle Time. xiv, 3, 6, 9, 10, 22, 27, 28, 41, 42, 44, 46, 48, 49, 54, 57–62, 65, 69, 71, 72 JIT Just-in-time. 2, 14, 17, 19, 27
LT Lead Time. xi, xii, 2, 18, 22, 23, 27, 43, 58–60 NVA Non Value Added. 11, 15, 16, 25, 28, 48
PCB Printed Circuit Board. 4, 6, 7, 32, 34, 37, 43, 45, 48, 50, 51, 58, 66
PT Process Time. xi, xii, 3, 12, 22, 31–33, 37, 41–49, 52, 54, 59, 60, 62, 63, 65, 70 VA Value Added. 11, 15, 25, 27, 48, 49
VSM Value Stream Map. 2, 3, 12, 17, 25–27, 48, 49, 64 WIP Work In Progress. 2, 18, 21, 27
Introduction
Most manufacturing companies, if not all, seek to minimize the time it takes to get their product to their customers with the highest possible quality. A big factor in reaching those goals is being as efficient as possible [2].
This master thesis is focused on improving the efficiency and consequently the output of a manufacturing line for Stoneridge Electronics. The manufacturing line that this project will focus on is the line producing one of their instrument clusters, Angela. Angela was chosen since it is one of the most important product in the plant and sold in high quantity.
An improvement of the manufacturing line would result in better service for the end customer which is what all successful companies strive for. Reduction in through-put time will also be discussed which would make the production more dynamic and quicker in adapting to changing customer demand. In this chapter, the background, aim, method and delimitation of this thesis are discussed.
1.1
Background
Stoneridge, Inc. is an independent designer and manufacturer of highly engineered
electrical and electronic components, modules and systems principally for the auto-motive, commercial vehicle, motorcycle, agricultural and off-highway vehicle mar-kets. The main focus of Stoneridge, Inc. solutions is to power vehicle intelligence sys-tems, provide dramatic increases in fuel efficiency, reduce emissions, and improve safety and security for everyone on the road.
Headquartered in Novi, Michigan, Stoneridge, Inc.’s global footprint encompasses 25 locations in 15 countries. One of those countries is Sweden. The subsidiary lo-cated there is called Stoneridge Electronics and it has offices in Solna, Stockholm and a manufacturing facility in Örebro which is where this master thesis was conducted.
Stoneridge Electronicscustomers include Daimler, Volvo, Scania and Man which makes the process highly regulated. The facility in Örebro is focused on making tachographs and instrument clusters, more commonly known as dashboards.
The manufacturing line that was analysed is 10 years old and it was built with various lean principles in mind. But a big part of lean is continuous improvement
[2]. As part of their continuous improvement process, Stoneridge Electronics wanted to identify key areas of improvement within the production process and an optimal solution for implementing the change.
1.2
Problem Statement
In the extremely competitive world of modern manufacturing, the importance of low time to market is imperative to a manufacturing companies success. Some of the rea-sons include the dynamic just in time (JIT) manufacturing concept and other lean ways of thinking, low work in progress (WIP) and small stock. By having short lead times (LT) and extremely efficient production, companies might be able to sell their products to a wider customer range [3].
The problem statement fabricated by Mikael Sterner, Production Manager at
Stoner-idge Electronics, was the following: What is the best way of improving the
productiv-ity of the Angela final assembly line by at least 10 percent?
1.3
Aim & Objectives
The main objective of this thesis was to find ways to improve efficiency and productiv-ity of the production of different Angela instrument clusters with the help of various lean manufacturing tools. The following objectives were set:
1. Create a current Value Stream Map (VSM). 2. Create a Spaghetti Diagram.
3. Analyse the VSM and find bottlenecks. 4. Break down and analyse bottleneck stations. 5. Analyse Spaghetti Diagram.
6. Introduce ExtendSim and analyse different options. 7. Improve the bottleneck status.
8. Develop a future VSM.
9. Formulate an action plan of improvement suggestions for management by using results from VSM, ExtendSim and Spaghetti Diagram.
1.4
Delimitations
variants go through the same process, the cycle time (CT) is not the same for each variant. How the operators move can also differ depending on the variant. Because of this, three categories were set and one variant in each category was analysed:
1. Angela 9 - Without ODO, variant 900597/09. This variant of the instrument clusters comes without small ODO screens but has a plastic front.
2. Angela 1 - With ODO, variant 900597/01. This variant has the ODO screens and a plastic front.
3. Angela Bus - With glass front, variant 900598/02. This variant is a bus variant and has a glass front.
These variants were chosen since they are the most sold variants in each of the cate-gories. The three different categories were picked since the difference in final assembly process times (PT) are the greatest between them.
Another decision that was made early in the project was to focus on the production after the SMD line. Subsequently, a VSM for the process was made focusing on sol-dering, coating and final assembly. Finally, there are countless tools that could be used to analyse and improve the process but only a few were used due to time restrictions. After careful consideration and discussions with the production manager and super-visor, VSM, process mapping, spaghetti diagrams and a simulation software were the tools used in this master thesis. A short analysis was done on batch size but none on production planning and production mix given time restraints.
1.5
Product
Figure 1.1: Angela Instrument Cluster
1.6
Production Process
The production process is the following for Angela:
1. Incoming material. The incoming material is distributed to workstations 2 (SMD), 3 (Common IC) and 4 (Final Assembly).
2. SMD D-Line. The SMD or surface-mount device produces electronic circuits with a method that involves mounting or placing components directly onto the surface of the Printed Circuit Board (PCB).
3. Common IC. See below. 4. Final Assembly. See Below. 5. Packing.
6. Dispatch.
As mentioned in 1.4, the focus was set on the Common IC and the Final Assembly.
1.6.1
Common IC
When the project work started, the Final Assembly was the only focus. But after vis-iting the manufacturing plant several times and observing the flow of products it be-came clear that the Common IC can play a big part in the production output of the product. As a consequence, Common IC was analysed in the same way as Final As-sembly.
The Common IC is a crucial part of the production since it serves four different final assembly lines. The products assembled at these lines are called Angela, Marianne, Desiree and Ebba. Even though the master thesis focuses on only one of these prod-ucts, Angela, the other products had to be analysed to be able to find a realistic output of the final assembly for Angela. After looking at the production plan, Ebba was re-moved from the analysis since that part of the production is negligible due to the vary Small weakly demand of 200 instrument clusters. There are numerous different pro-cesses related to the soldering and coating part of Common IC as seen by Figure1.2:
• 1 Load Board Manual: In this step, the worker removes a PCB board from a rack and places it in one of three slots. The slots differ in size and fit a certain product. After inserting the PCB, the worker turns two plastic hooks to secure the board. Finally, worker must press a button for the board to move (E.14).
• 1.1 Load Board Auto: After the manual operation, the board moves automati-cally to the next station.
• 1.2 HMT Assembly & Inspection: HMT stands for hand-mount technology and in this workstation, the operator mounts small electric components on the PCB. The number of components attached to the PCB board can vary from one to six depending on the product. After assembling, the operator presses a button to initialise an automatic quality check inspection (E.15).
• 1.4 Transfer 1 & 2: After HMT, the board is transferred to a waiting stop. The latter transfer happens after cooling when the board goes back in a circular fash-ion. An elevator lowers the board and from there it moves towards the first step of the process. Another elevator elevates the board and moves it to solder in-spection. Transfer 2 is the only operation within soldering that acts on multiple products at the same time. This leads to lower CT for that process.
• 1.5 Waiting Stop: A programmed stop for waiting.
• 1.6 Flux: Flux is sprayed on the board in order to reduce the oxides that tend to form when hot metals are in contact with the air.
• 1.7 Preheating 1, 2, 3 & 4: The board is preheated to be able to solder more quickly.
• 1.8 Solder 1, 2 & 3: Electric components are soldered to the PCB board.
• 1.9 Cooling: After preheating and soldering, the board is extremely hot so cool-ing is necessary.
• 2 Solder Inspection Manual: In this step the board is removed from the rack and inserted into a solder inspection machine.
• 2.1 Solder Inspection Auto: The automatic step of the inspection takes place in this process (E.16).
After going through soldering, the PCB boards must go through coating. There are several processes related to that as well:
• 3 Loading: After soldering inspection the operator loads the PCB board for coat-ing (E.17).
• 3.2 Coating 1 & 2: The coating adds a protective layer to the PCB boards.
• 3.3 UV: UV curing is the next step. It prevents damage from rough handling, installation, and reduction of mechanical and thermal stress. It can prolong the life of the product during its operation.
• 3.4 Inspection: The last step in the machine is manual inspection (E.18).
• 3.5 Stepper motor: A part of Common IC is attaching stepper motors to the PCB boards. The operator must walk out of the line to work at the stepper assembly machine (E.19).
1.6.2
Final Assembly
The main focus was put on the final assembly part of the production. When the work-piece reaches this stage of the production, all small electric components and a step-per motor have been mounted on the PCB. Furthermore they have been soldered and coated in the Common IC line. The processes in final assembly are the following:
Figure 1.3: Final assembly process flow
places two or no odometer (ODO) screens on the board with rubber and metal frame.
• 2 Display assembly: A large display is added to the board.
• 3 Mechanical assembly: The screen undergoes manual display inspection after an overlay assembly. Then the pointers are mounted and a pointer press snap fits them to the board.
• 4 Vibration test: Board is inserted into a vibration test machine for a 60 second test.
• 5 Test line: The test line is divided into three stations Electrical, Vision and AVT here Pointers are calibrated and a function test is run.
• 6 Back cover and front assembly: Back cover is assembled to the front. Operator does a visual inspection of the workpiece.
• 7 Screw station: Back cover is screwed in place with eight screws.
• 8 Appearance test: Last station in Final Assembly, protection film is added and product labelled.
1.7
Outline
The outline of this master thesis is as follows:
1. Introduction: The company, product and the production process is introduced. 2. Methodology: The methodology used in the thesis is introduced.
3. Literature review: The relevant literature is introduced and the reader is made familiar with all of the terms used in solving the problem statement.
4. Discrete Event Simulation: A short introduction of the theory in simulation is introduced. The functionality of the models is described in detail.
5. Results: The results are revealed and supporting data and analysis is provided. 6. Discussion & Recommendations: The problem statement is discussed with
Methodology
This chapter explains and supports the methodology used to address the research questions. The chapter describes the methods for information and data collection. Fur-thermore, the way observation and time recording was carried out will be described, as well as the evaluation methods for assessing the results of the simulations.
2.1
Scientific Approach
At the start of the thesis, the authors worked as operators within the production. Working in the production line helped in understanding the process from the perspec-tive of the operators. Furthermore, trust was established with the operators which made timing the processes under normal working conditions easier. It was impor-tant that the operators did not behave differently in the sense of rushing because they were being timed. Thus, by establishing trust, time measurements followed standard movement and assembly rate. Additionally, the operators provided valuable inputs on a daily basis and through interviews. By understanding how and why the instru-ment clusters were made in a certain way, improveinstru-ments could be suggested after collecting data. The focus was put on quantitative analysis.
This master thesis is based on empirical data that was recorded at Stoneridge Elec-tronics in Örebro. The data that was collected for the thesis was collected by following methods:
1. In person measurements. Actual process and CT were taken by using a simple mobile phone stopwatch.
2. Data from Stoneridge. QliqView data collected from Stoneridge’s database. 3. Various interviews. Unformal meetings and interviews were had with operators,
production planners, production technicians and production leaders.
In order to solve the project with the best results it was crucial to get as familiar as possible with Stoneridge’s production system (Common IC and Final Assembly) and identify the bottlenecks. That process was very time consuming but when access to Stoneridge’s various databases was established it was easier to analyse the data. The
time spent observing was highly valuable in the long term because it provided us with better understanding of general flow of production.
2.2
Case Study
According to [4], a case study is used to investigate a current phenomenon within its real-life context. When the boundaries between the phenomenon and context are not clearly evident, multiple sources of evidence are used. Furthermore, the focus is put on the process rather than the results in an effort to reach a holistic results. Finally, the aim of the case study is to discover rather than to prove [5].
As aforementioned, this thesis is based on measurement data recorded at Stoner-idge’s factory in Örebro. The data was collected by recording actual times, by open interviews with the operators, production technicians and production leaders and by gathering existing data at the plant. After gathering all the data, various lean tools and a simulation software were used to analyse the system and find the bottlenecks.
2.2.1
Information & Data Collection Methods
Information and data for this study was collected through literature review, observa-tions, ExtendSim simulations and company software such as QlikView and AXXOS OEE (Overall Equipment Effectiveness).
2.2.2
Time
A vital part of this master thesis was to collect sufficient amount of data to analyse the system and propose improvements. One of the ways this was done was through observations by timing the processes like described in Chapter 2.1. After data gather-ing, the bottlenecks were easily identified. Furthermore, by timing various processes ourselves, the time it takes for one workpiece to go through all the steps could also be measured easily. The timings were all done on the iPhone stopwatch through their Clock application. For the final assembly, QlikView provided us with CT between each workpiece.
2.2.3
AXXOS
One of the softwares used in this project is called AXXOS. AXXOS is a production monitoring solution with a strong focus on follow-up and optimization which helps give companies the full picture of how effectively their production is running. Accord-ing to the companies website, the software helps its users in the followAccord-ing way:
2. Improve productivity. Enables continuous improvement resulting in increased efficiency and productivity through a platform which gives access to detailed production data and visualization tools.
3. Increase and secure output. Facilitates your planning for precise delivery and increased total production output by securing full control over availability, per-formance and order status.
4. Minimize efforts to follow up production performance. Reduces reporting time and allows for more value-adding improvement work through advanced automatic data collection combined with efficient and intuitive support for man-ual reporting of causes for disturbances.
5. Improve quality. Enables analysis of data from quality related disturbances on machine, product or order level which in turn has a direct impact on OEE when reasons for quality issues can be quickly identified and fixed.
6. Get the facts. Ensures easy access to reliable data needed for decision-making on improvements, investments and changes in production flow.
The operators use the software to monitor breaks in work due to machine down-time or personal breaks. By using the collected data it was easy to analyse Non Value Added (NVA) time and Value Added (VA) time with regards to a whole shift.
2.2.4
QlikView
QlikView is a a monitoring software that is used by Stoneridge to collect data from the actual production. This software is connected to the machines within the production which enables it to gather all the data that is recorded. This includes test times, stop times, quality, yield of production and deviations within the production. It can also generate different reports depending on what data is gathered. QlikView was used to get a better understanding of the production and confirm various time measurements.
2.3
Literature Review
The literature review aims to provide general understanding of the topic as well as provide a detailed insight into previous studies, and counts for a big part of this thesis work. It was carried out via online academic databases and search engines (e.g. KTH Primo, Google Scholar, Science Direct). Furthermore, reference lists were used to find more literature on certain topics. The search was conducted in English.
2.4
Data Analysis
After collecting sufficient data, which consisted of between 15-30 time measurements, analysis was done. To start off, a VSM was created. By analysing the VSM it was easy to calculate total throughput time and identify bottlenecks. Subsequently, ExtendSim was used to simulate different scenarios within the production and calculate the high-est theoretical output. Due to similar operating times, the simulation was made by using the average process time (PT) of different operations.
2.5
Interviews
Literature Review
In this chapter, the study of existing literature on lean manufacturing is introduced. The aim of the literature review is to understand the current status of lean manufac-turing, discuss studies of similar nature and introduce lean manufacturing criteria and lean tools.
3.1
History of Lean Manufacturing
The Machine That Changed the World is a book written in 1991. It is based on the Massachusetts Institute of Technology’s $5 million, five-year study on the future of the automobile. Written by James P. Womack, Daniel T. Jones, and Daniel Roos, it made the term lean production known worldwide. The book provides examples on how Henry Ford revolutionised mass production and how the Japanese improved it by creating the Toyota Production System (TPS). The combination of both provide the basis of lean manufacturing [6].
Toyota Production System
One of the biggest success story of the lean concept comes from Toyota, Japan [6]. Now known as the Toyota Production System (TPS), the idea of just-in-time production was originated by Kiichiro Toyoda, founder of Toyota [7]. It is a comprehensive business system that produced remarkable results in quality, productivity, and continuous im-provement marked by Toyota’s products and services. Toyota bases the system on continuous reduction of waste, respect for their employees, and customer satisfaction [8]. Combined with the idea behind mass production through an assembly line, intro-duced by Henry Ford to lower manufacturing costs, the lean concept was born [6].
Henry Ford
Before Henry Ford invented the assembly line process which led to mass production of automobiles, the car industry was a craftsmanship. One person manufactured the whole car and all the parts manually. Many parts came out imperfect and demanded a lot of rework which in turn increased the cost of manufacturing [6].
When Henry Ford established Ford Motors in Detroit, USA, mass production of the T-model automobile was established according to the moving assembly line prin-ciple. This encouraged automotive manufacturers around the world to incorporate mass production into their own production. When TPS started to develop further, in-spiration was brought from mass production of cars and mixed with parts of the old methods [6].
Lean Production
Post World War II, Japanese manufacturers faced big problems and issues regarding financial security, shortage of material and workforce [9]. A big part of their industry had to be rebuilt so they turned their eyes to the Western world to try to incorporate what they had achieved. The Japaneses sought to get valuable knowledge and ideas on how to rebuild their industries so they could meet the new demands and the new way of mass production. But the Japanese faced some difficulties adopting to the Western way of mass producing. The reasons for that were fundamental differences in the different markets. The Japanese faced smaller production volumes with limited resources and as a result they needed to develop a manufacturing system that could utilize less resources while being flexible to changes in production volume [10]. Under these circumstances and under the leadership of Taiichi Ohno, Lean Production as we know it was born [11].
In the late 1980s, automotive industries in the Western world started noticing that the industry in Japan achieved higher productivity and quality using less resources [10]. This made Western organisations interested and an investigation to what differed was started by Womack, Jones and Roos at Massachusetts Institute of Technology. The result of the study showed that there was in fact a big gap between the productivity and quality of the Japanese automobile producers when compared to the western ones. The Japanese way of production was later called lean production [11].
The lean thinking originated from Toyota’s shop floor according to the book The Machine That Changed The World which then also gave growth to several tools that we use today like Just-in-time (JIT), pull production, kaizen, visual control, one piece flow, visual control and more [12].
3.2
Principles of Lean Manufacturing
There are five main principles for lean [2]. 1. Value.
2. Value stream. 3. Flow.
One of the main benefits of these principles is that they can be seen and act as a series of steps that need to be carried out to implement lean thinking [13].
3.2.1
Value
The first principle is value. It is of vital importance to identify the value from the customers perspective and eliminate the waste. The end product that reaches the cus-tomer has to meet and satisfy the cuscus-tomer’s requirements and demand with the spe-cific price and time [13]. An organization needs to think about the value for the end customer from the beginning of the process otherwise it will not work.
3.2.2
Value Stream
The second principle is value stream. In this step, the organisation must identify the entire value stream for a product or product family. All steps are recognised and mapped. This includes all activities that the material goes through on its way to the fi-nal product [13]. All value added (VA) and non value added (NVA) steps get mapped, and by using this tool it is possible to visualize the production flow and find the bot-tlenecks.
Value-Adding (VA) & Non-Value Adding (NVA)
The three different kind of operation that are done in the processing of an unit are: 1. Non-value adding (NVA)
Defined as pure waste and involves actions that are pure waste and completely unnecessary from the customer perspective [14]. It can include double handling, over processing and making redundant quality checks for example. Waiting time is one of the biggest non value adding (NVA) activity in a processing plant [15]. 2. Necessary but non-value adding
Wasteful operations that are necessary nevertheless. Examples of these include walking to get parts, transferring tools between hands and opening delivery boxes. The NVA adding activities usually go under the seven wastes [12]. Elimi-nating these wastes would require a big change in production layout or the way supplies are delivered. It might be hard or impossible to implement some of those changes [16].
3. Value-adding (VA)
3.2.3
Flow
The third principle is having a continuous flow throughout the whole production which means that production goes from working with batches to moving one work unit at a time between the different steps of the production process. When and if im-plemented correctly this can result in reduction of waste, save money and more [2].
3.2.4
Pull
The fourth principle is to have a pull system in place which is achievable through the aforementioned principles if they are applied and followed correctly. The main aims and ideas behind the pull system is that you only start work on a new product when you have customer demand for it. By following this principle, an organisation should be able to reduce costs on storage for example [2].
3.2.5
Perfection
The last principle of lean is to work towards perfection. That means that the organisa-tion needs to eliminate all the NVA adding activities which in turn adds value to the end customer product [2].
3.3
Manufacturing Wastes
For a production to be as effective as possible the production flow needs to be formed in a way so that the wastes can be easily identified which makes it easier to eliminate [17]. Within lean there are seven forms of wastes that must be recognized so they can either be minimized or eliminated.
1. Overproduction
Overproduction is usually seen as the worst one of the seven wastes. This stems from the fact that it can easily lead to the rest of the wastes. Overproduction relates to a few different things. It can mean that a company produces more than the customer wants. It can also mean that the company produces faster on certain stages of production which leads to buildup of material when some processes are slower. It can also be related to big batch sizes [17].
It will lead to surplus in inventory which in turn ties up a lot of capital. Produc-ing to fast for a customer can lead to overproduction. A scenario relatProduc-ing to that issue is if the manufacturing company receives an order that is to be delivered at the end of the month but management decides to make the order early and store it for most of the month. It will lead to materials being tied up in inven-tory and customer order may change which will result in wrong products being made [17].
be to produce according to customer order and demand instead. Lean philoso-phy says that machine and humans should only be utilized when there is a useful task at hand [9].
Producing solely on customer orders and demand will prove to be troublesome for all manufacturing companies were throughput times are high. The customer will not wait for so long. So forecasting has to be utilized in the correct manner. A lean philosophy that can be adapted to reduce the waste of overproduction is Just-in-time (JIT). That philosophy means that we produce and deliver at the right time, but usually companies tend to work with a philosophy of their own, just in case [17].
2. Waiting
The form of waste related to waiting means unused time due to waiting for nec-essary prerequisites [17]. Waste due to waiting can be associated with waiting for material to arrive, waiting for information, products waiting to be processed, machine waiting due to the operators and operators waiting for the machine. Products in inventories are also seen as waste due to the fact that they are wait-ing in the warehouse [17].
A tool that can be used to visualize all waste due to waiting is aforementioned Value Stream Map (VSM). The VSM can be used to see and analyze the product-, material- and information-flow through the production. Setup-product-, changeover-product-, cycle- and process-times are mapped out to help visualise the production flow and to reduce waste.
3. Transport
Transport does not add value. The only transport that the customer is commonly prepared to pay for is shipping of final product. The internal transport within the factory can be categorised as total waste [17].
The unnecessary movements and motions of the product and operators within the factory can also be seen as waste. This will then result in wasteful transport within the factory that can lead to damaged products and or lost parts [18]. A tool that can be used to visualize and reduce waste due to transport is a spaghetti diagram [17]. The spaghetti diagram visualizes the physical flow of the material and the operators within the factory. The creator can also chose a specific area which reveals all the movements within that area. It can help greatly with visualising the waste that comes from unnecessary motion within the fac-tory [17].
4. Over Processing.
pay for the better quality [17]. Over processing can also occur when organisa-tions do not check the exact customer requirements and rather work with what they think the customer wants.
5. Inventory.
Inventory is the amount of material we have in storage and WIP. The material that is kept in inventory becomes waste when unused. This is due to the fact that it takes up space and may become unusable if something changes to the orders [18].
There are many different ways organisations can deal with inventory waste. Re-ducing the number of variants of some products could lead to a reduction in inventory and buffers. Doing that would decrease the amount of inventory for a product that is not necessarily sold in high volumes. Another solution could be to reduce the batch sizes and subsequently WIP. By reducing the inventory levels an organisation can achieve better production flow, lower LTs and a reduction in tied up capital due to large inventories [17].
6. Movement.
Movement or motion that does not add value is considered as waste. Examples of this include walking to get tools or material. Another example is when a workstation is designed in a way that an operator needs to stretch to get the tools [17].
7. Defects.
Producing parts that have defects will lead to waste through rework. Most of the time, parts get defected because of bad manufacturing processes that are either caused by human- or machine-errors [17].
This can be reduced by introducing a one piece flow production. By using one piece flow, the operator can notice a defect immediately and handle it with as much efficiency as possible with regards to rework [17].
3.4
Lean Tools
In this sub chapter, various lean tools will be introduced. There are numerous tools available but only the ones used in this master thesis will be described.
3.4.1
5S
1. Seiri (Sort)
When applying Seiri to the organisations, every object in the work area should be looked at and a decision should be made on keeping it or throwing it away. The object that are rarely or never used should be thrown away. This includes tools, materials and information.
2. Seiton (Set in order)
Seiton highlights the importance of having a set place for every object used in the production. Applying this principle should help the workflow go smoother and easier.
3. Seiso (Shine)
With Seiso, it is important to keep a clean workplace. This involves sweeping or cleaning and inspecting the workplace, tools and machinery on a regular basis. Everything must be kept in order.
4. Seiketsu (Standardize)
When the three first steps are done we can then move on to standardizing the work. It means that the organisation must organize and optimize the sequence of the work process so it has a repeatable work process for the operations that the operators can fallow.
5. Shitsuke (Sustain)
The last principle means making all employees follow the set standard without falling back to the old pattern of working. Requires self-disicpline of the workers [17].
3.4.2
Total Productive Maintenance (TPM)
This is a tool that is used within production to drive down the downtime of machines and equipment [19]. TPM focuses on proactive and preventive maintenance to opti-mize and maxiopti-mize operational time.
3.4.3
Just in Time (JIT)
3.4.4
Poka-Yoke (Error Proofing)
Poka-Yoke is a Japanese concept for solutions that should make it extremely hard or impossible to make a mistake in production. The concept is used to ensure that prod-ucts are done correctly from the beginning [17]. An example of Poka-Yoke is when a manufacturing system makes many different products while using the same screws, eliminating the possibility of making a mistake by picking the wrong screw.
3.4.5
Kaizen (Continuous Improvement)
Kaizen is a tool used within organizations for continuous improvement. Standardiza-tion is a prerequisite for kaizen due to the fact that everyone in the company needs to be involved in the improvement process [17]. It is a process in which teams attack a manufacturing operation to make a series of quick, small steps in an effort to improve the process. It is also the process by which such small improvements are continued [8].
3.4.6
Standardized Work
Standardized work is a set of agreed upon operations that is established to be the best method and sequence for a process. This is a powerful tool that creates an efficient workflow that helps to minimize variations in a process, helps to maintain quality and helps operators meet the customer demand [19].
3.4.7
Kanban
The Kanban system was created to deal with the stream of work units in and out of supermarkets and work areas. It is an extraordinary method that is used to deliver the required sum of material or products exactly when it is needed [19]. In the system, the type and quantity of the units needed for production are written on a tag-like card. The card is called a "Kanban" [16]. There are three types of kanban:
1. Production-Ordering Kanban
The production Kanban shows the workers the amount of products that need to be made. This type of Kanban is often called in-process Kanban or simply a production Kanban [16].
2. Signal Kanban
Tells workers the amount of products that need to be produced in a batch. 3. Withdrawal Kanban
Tells the worker the amount of product that he can remove from the supermar-ket.
1. Rule 1. A process should withdraw the necessary products from the process just before in necessary quantity at exactly the right time.
2. Rule 2. A process should produce in the same quantities as the preceding pro-cess.
3. Rule 3. Defected product that fail to go through a quality check should never continue downstream activity before getting reworked.
4. Rule 4. The number of Kanbans must be minimized in an effort to decrease WIP. 5. Rule 5. Kanban should be used to accomodate small flucutations in demand. An organisation should be able to fine tune its production by using the Kanban system.
3.4.8
PDCA
PDCA or plan, do, check, act, is an effective, simple and popular method that was originally designed to analyze and develop processes but has since been applied in many other areas. This method was invented by W.Edwards Deming and is described in detail below [17].
1. Plan: In this phase the work is not only about planning the project and what needs to be done. Here it also includes defining the customers (Internal or Exter-nal) needs, collect data, distinguish and analyze the problems and identify the root cause of the problem [17].
2. Do: If the planning phase is thoroughly done then this phase is quite easy to perform. Implement the changes proposed in the previous phase. Also perform measurements of the result and carry out any training [17].
3. Check: Evaluate the measurements and the implementation work and analyze the results. Report the results to any decision makers that are involved in the project [17].
4. Act: If the improvement project is a success, the new improved level must be secured. The improvement has taken the business from one level to a new more improved level. This new level needs to be secured in a new standard [17].
3.4.9
Five Why’s
3.4.10
SMED (Single-Minute Exchange of Dies)
Setup cost can prove to be high and ignoring it can lead to many forms of waste [20]. SMED focuses on all methods that can be used to reduce changeover and set up, it should not be longer than 10 minutes [21]. Four steps are taken which involve identi-fying and improving internal and outer set-ups and preparation [22].
3.5
Time
Taking the times of each operation is necessary when analysing and balancing the manufacturing line. It is of utmost importance that the times are as accurate as possible so appropriate steps can be taken when balancing the line and making it more efficient in an effort to reach the desired state [16].
3.5.1
Process Time (PT)
Two different times were taken during observations. First, the focus was on process time (PT). A few definitions of PT exist in literature but in this project, PT was taken as the time from when the operator picks up the workpiece until he stops working on it. After this definition was set, it was important to only measure the time when an operator was actually working on the workpiece or when an automatic operation was taking place. In these timings, waiting time was not taken into account.
Due to the fact that the analysing was done for three operators working on 8 differ-ent workstations simultaneously, the stopwatch was stopped if the operator stepped away from the workstation and no automatic process was running. In literature it is defined as the amount of time a part spends being modified into a new, more valuable form [23].
In some literature, PT is called operation time. It is then referred to as LT for carry-ing out one manufacturcarry-ing step. It represents one part of the throughput time [24].
3.5.2
Cycle Time (CT)
The second time that was taken was Cycle Time (CT). Even though some numbers were accessible through company software, observations were necessary to under-stand the flow perfectly. CT was defined as task cycle. There are numerous different definitions in literature:
• The time a worker spends on a certain workstation before repeating the same operation [6].
• The time it takes to manufacture one individual product [24]. • Time it takes for the operator to finish all of the work tasks [25].
3.5.3
Throughput Time
Throughput time is defined as the time it takes to manufacture a product from the first step in production until the last. The throughput time is a part of the LT and includes transport times, queuing time, set-up time and producing time [24].
3.5.4
Takt Time
Takt time is defined as the rate the production needs to be at to complete the pro-duction of a product in order to meet the customer demand [17]. An example is if a manufacturing plant has a weekly demand of 2400 products. That means that if the production is working 40 hours a week the plant needs to produce a new product every 1 minute to meet the demand. Therefore, the takt time will be 1 minute.
3.5.5
Lead Time (LT)
When we talk about Lead Time (LT) we need to know that there are three different LT definitions within the frame of LT.
1. Order LT: The time from when the customer order is received by the company until the order i delivered. Order LT is the longest of the three types when a customer order is received the clock starts and it stops when the order has been received by the customer.
2. Manufacturing LT: Manufacturing LT is the time from when the received order from the customer is put into the ERP system and it can be seen by the production until the order is ready for delivery.
3. Production LT: Production LT is the time from when the start of actual produc-tion of the first part until the end of producproduc-tion ie when the order is ready for delivery.
3.5.6
Set Up Time
Set up time is the time it takes to prepare a machine or a group of machines for the up-coming task. There are two different types of set-up times [24], internal and external.
• Internal set up: Tasks that can only be performed when the machine is stopped. Involves changing or adjusting something in the machine that would be impos-sible to do with the machine running.
3.6
Assembly
A big part of this thesis was analysing the assembly in the manufacturing line. It is de-fined as the fitting together of manufactured parts into a complete machine, structure or unit of machine [17]. In other words it involves putting components together in or-der to build the final product. The parts can either be assembled by putting individual components together or sub assemblies. Sub assembly is a unit assembled separately but designed to be incorporated with other units into a larger manufactured product [26]. Assembly can be split in three groups [27]:
1. Manual Assembly: Assembly done by an operator with or without tools. 2. Mechanized or Automated Assembly: A completely automatic operation.
3. Robotic Assembly: Combination of manual and automatic operations, where handling and composing operations are performed by one or more robots [27]. There are numerous reasons for the importance of manual assembly in a produc-tion facility. These include [28]:
• Shorter product life cycles. With rapid technological advancements, new prod-ucts are pushed to market at an ever increasing rate. Furthermore, batch sizes are getting smaller and number of variants increasing. This makes investing in automated assembly machinery unviable for many organisations.
• Any oversight in the design, planning or development phase is more easily iden-tified when a product goes through assembly-oriented product development. When an organisation is making a decision on whether to have manual or auto-matic steps there are a few factors to consider. Both methods have their advantages and disadvantages. In general, manual assembly is less risky since it does not require a big investment for machine producing a product that could be a failure. This is due to the low flexibility of machinery. Most automatic systems are product specific, con-sisting of design and manufacturing of special applications, and cannot be reused for other products [17]. However, when dealing with a high volume production, auto-mated systems are a big advantage. The reasons are numerous but include [17]:
• Fluctuating: It is easy to deal with fluctuating demand if the system is auto-mated. In manual assembly, small fluctuations can be handled with overtime but big fluctuations require more employees leading to less efficiency when or-ders are low.
• Motivation: When an operator is asked to do the same tedious work thousands of times with little to no movement, motivation and the attention to details might drop. This can lead to errors and lower productivity.
3.7
Bottleneck
A bottleneck is considered as the process in a series of processes that is a constraint. In a manufacturing line this means that the bottleneck is the operation that takes the longest time. Consequently, in a one piece flow ,it also sets the takt time of a line. By identifying and eliminating the bottleneck, it is possible to increase the output.
There are two types of resources in a plant, either a bottleneck or non-bottleneck [23]. "A bottleneck is any resource whose capacity is equal to or less than the demand placed upon it. And a non-bottleneck is any resource whose capacity is greater than the demand placed on it" [23].
Another important thing to remember is that there will always be a bottleneck within a chain. When bottleneck status is improved within the chain, another opera-tion will become the bottleneck. Improving bottleneck status is thus a part of continu-ous improvement which is a key point of lean thinking.
3.7.1
Bottleneck Identification
In the case that the bottleneck is not the first operation or station a buffer could accu-mulate before the operation or station with the longest processing time. This can be revealed through observation or when drawing up a VSM. Another way of identifying bottlenecks is making a bar chart that shows the capacity of each station or operation. This method is one of the most common lean tools and it is used to show how balanced or imbalanced a process is. In this step it is key to calculate the capacity per process and not per person or machine due to the fact that there could be several persons on the machine [23].
3.8
Value Stream Mapping (VSM)
A value stream is the flow of information and materials to produce a value for a cus-tomer. It includes everything from NVA activities to communication along the supply chain and the network of processes which materials and information flow through [19].
A Value Stream Map (VSM) is a map used to outline the current and future state of a production system. By creating one, the user can get a visual overview of the process flow and discover sources of waste and consequently which actions take the most amount of time [29]. The term value stream got it´s name from the fact that the process flow through the system should be like a stream or a flow through a river. The customer is farthest downstream and the product starts farthest upstream. Then it should flow to the customer as smoothly as possible [19].
Finally, the overall goal of a VSM is to move from a batch and push process to one-piece flow and pull through the entire value stream. The main goal is introducing a lean value stream that optimizes the flow of the entire system [29]. This is the way it should be implemented [19]:
1. Commit to Lean
2. Choose the Value Stream 3. Learn about Lean
4. Map the Current State 5. Identify Lean Metrics
6. Map the Future State (using the demand, flow, and leveling concepts). 7. Create Kaizen Plans
8. Implement Kaizen Plans
But when reading through the list, we should also keep in mind some important lean management principles:
1. Define value from your customer‘s perspective 2. Identify the value stream
3. Eliminate the seven deadly wastes 4. Make the work flow
5. Pull work, don‘t push it 6. Pursue to perfection 7. Continue to improve
3.8.1
Current state
When making the current state map, the user should start off by observing the system to understand it. He should then observe the general material flow. The information flow is also important. To represent different steps and different flows, the user uses various icons. There are three distinct types: material flow, information lines and general icons [19].
When drawing the information lines, it is important to include everything related to the information flow. Information can be aquired with kanban cards, meetings (manual), electronically, go-see scheduling and load levelling. Finally, before the user starts mapping, he has to decide which product he will analyse [19].
production information. Since the value stream should represent the flow from raw material to shipment it is important to start with the customer because he is the only individual that can define the true value of the commodity at hand [19].
The next step is adding the supplier production process. It can be hard to represent all of the suppliers so a rule of thumb is to include the one that supplies the most im-portant components or supplies the most by dollar amount. Lastly, the user connects all icons with information flow icons [19].
3.8.2
Internal mapping
After mapping the external processes of the production the user must focus on the in-ternal aspects. The user should record number of machines, processes, WIP inventory (raw or finished) and all CT in each process. When making the VSM, it is important to be honest and include all small details. Furthermore, it should be made by using data from a visit but not data on how it should run in theory [19].
3.8.3
Waste
After mapping the process it is important to identify and eliminate waste in an effort to shorten LT and improve the VA percentage. As mentioned in chapter 3.3, there are seven different types of waste. They have various effects on the value stream and are represented in the map in a certain way:
1. Overproducing
Producing too much of something or producing before it is required. It is pos-sible to find signs of overproduction by looking at inventory triangles. If there is any inventory at all, it is overproduction. Producing to much or too soon de-creases efficiency.
2. Inventory
Keeping inventory is not part of lean but in many cases perfect one-piece flow is hard to achieve. We represent it in the VSM by using a triangle with an “I“ in the middle.
3. Transportation
Represented with a truck icon. In most cases it should be possible to decrease it by using pull systems, one-piece flow, JIT supply chains and proper plant layout. 4. Waiting
5. Motion
Same as for waiting. 6. Over-processing
This occurs when complex solutions are applied to simple situations. Buying a highly inflexible machine for example could lead to overproduction which would increase inventory. All in the hopes of justifying the cost of the machine. Another example is requiring an excessive amount of signatures or paperwork for a unit.
7. Correction
Finally, the correction should be represented by an inventory icon. It is a truly NVA adding step [19].
3.8.4
Future state
After making the current state map, the user should brainstorm and come up with ideas on how to improve the production flow and decrease or eliminate waste. The first step lies in calculating takt time. Takt time is the demand from the customer per unit of time. It is important to note that takt time is not the same as CT. CT is the time between the same operation and not related to the customer. Still, CT plus a small efficiency factor should be as close to takt time as possible in a one piece flow lean system. After calculating the takt time, the user knows how fast the process should be in the ideal state. Redesigning the stations and layout with that in mind will make the process more lean.
The last thing the user must take into account is levelling of the production. After considering all of the above, a future state map can be generated and the way produc-tion could be is introduced. After the future state map is completed, an acproduc-tion plan can be devised that introduces how to transition from current state to future state [19].
3.9
Spaghetti Diagram
The spaghetti chart or spaghetti diagram is a visual aid that shows the movement of a workpiece or workers on the factory floor [22]. It is intuitive and helpful for identifying waste in movement and interruptions in flow.
3.10
Assembly Line Balancing
3.10.1
Assembly line
• Single model line: One product that is produced and assembled and all the parts are identical.
• Mixed model line: Products of different models are produced in a mixed se-quence [31].
• Multi-model line: Series of batches of products are produced where each batch only contains one or a group of remarkably similar models with moderate setup activities.
3.10.2
Line balancing
Assembly line balancing (LB), involves the challenge of allocating operations to work-stations along the assembly line so that the assignment are in some logical sense more efficient. When Henry Ford made the introduction of the assembly line, the study of assembly line balancing became a necessity. There is a difference in efficiency between a balanced line and an unbalanced line which can either result in loss of economics or gain [32].
