Reduction of non-value added work at
Essity Hygiene and Health AB
Opportunities for automation and digitalization in Baby products production
at Falkenberg Plant
Lisa Antonsen
Industrial Design Engineering, master's level
2018
Luleå University of Technology
ACKNOWLEDGEMENT
I would like to thank everyone that have been involved in this project and helped me with this master thesis in Industrial Design Engineering, at Essity Hygiene and Health AB in Falkenberg. At Essity I first and foremost want to thank Marcus Lehto, my supervisor for guiding me in the right direction and giving me good advices. A huge thanks to thank the operators at Fenix 2 who kindly participated in all my test, interviews, surveys and made me feel welcome to the team. I would also like to thank all the employees at Essity Hygiene and Health AB who participated in interviews and provided me with the information needed for the project. Thank you to Magnus Stenberg, my supervisor at Luleå University of Technology, for giving me great feedback and guid-ance in the right direction.
Last but not least, I would like to thank the most important people in my life. Mom and dad, thank you for always having my back, believing in me and supporting me. You are the best parents imag-inable.
Lisa Antonsen Luleå, June 2018
ABSTRACT
Today’s new technology provides endless opportunities to automate and digitize operations. An increasing num-ber of companies replace manual tasks with new technologies, aiming at increasing efficiency and productivity. This master thesis project was made in corporation with Essity Hygiene and Health AB at Falkenberg. In order for Essity to continue to be one of the leading manufacturers and maintain their financial ability, they need to take a step in the right direction towards a more modern production.
The purpose with the project is to propose ways to reduce the amount of manual non-value adding work per-formed by operators, with focus on time losses, ergonomic and safety. The project shall deliver realistic concepts that reduce manual tasks performed by operators through automation, and find new ways to quality-proof the material.
The project has worked according to an iterative work process where the cycle was repeated, until the outcome satisfied the needs. The project started with mapping the current state where Fenix 2, one of the machines, was selected for the mapping. Two weeks was spent in the production to gain understanding of the operators work tasks and working environment. The project did, among other things, observations, interviews, surveys, ergono-mic analysis, flowcharts and spaghetti diagrams. The result from the current state shows that the refill of material is the most time-consuming task for the operators. It was also made clear, based on step counters and spaghetti diagrams that the operators walk a lot. The result from the mapping formed the basis for the development areas. The project chose to proceed with the material refill, walking and the quality and made requirement specifica-tions for each area.
An ideation process was used to generate as many ideas as possible for the chosen areas. The ideas then became early concepts, which were evaluated with weighted matrixes and feedback from employees. The concepts were then further developed into three final concepts, Ground Floor, Second Floor and Integration. The concepts shows that it is possible to reduce the manual working time with 4,5 hour per shift and decrease the walking with better information system. Further recommendation is that Essity proceed with the results, both the mapping and the concepts. The recommendation is to do another evaluation and include the operators in the process and after that use the results in this thesis as the basis for future implementations.
KEYWORD: Production development, Ergonomic, Safety, Automation, Digitalization, Information system, Alarm, Mapping, Waste.
SAMMANFATTNING
Dagens teknik medför oändligt med möjligheter till att automatisera och digitalisera verksamheter. Allt fler företag ersätter manuellt arbete med ny teknik i syfte att öka effektiviteten och produktiviteten. Det här exa-mensarbetet har genomförts i samarbete med Essity Hygiene and Health AB i Falkenberg. För att fortsätta vara ledande inom deras bransch och behålla sin konkurrenskraftighet behöver Essity ta ett steg i rätt riktning och modernisera sin produktion.
Syftet med projektet är att föreslå sätt att minska mängden manuellt, icke värdeskapande arbete hos operatörerna med fokus på tid, ergonomi och säkerhet. Projektet ska leverera realistiska koncept som minskar mängden icke värdeskapande arbete hos operatörerna genom att automatisera manuella arbetsmoment och finna nya sätt att kvalitetssäkra materialet.
Projektet har arbetat efter en iterativ arbetsprocess där processen upprepats tills resultatet uppfyllde behoven. Pro-jektet startades med en kartläggning över nuläget där Fenix 2, en av maskinerna valdes ut. Två veckor spendera-des i produktionen för att få förståelse för operatörernas arbetsuppgifter och arbetsmiljö. Projektet gjorde bland annat observationer, intervjuer, enkäter, ergonomiska analyser, flödesscheman och spagetti diagram. Resultatet från nuläget visade på att det är påfyllnaden av material som tar mest tid för operatörerna under ett skift. Baserat på spagettidiaramen och stegräknarna blev det tydligt att operatörerna går mycket. Resultatet från kartläggningen utgjorde grunden för vilka områden som projektet vidareutvecklade. Projektet valde att gå vidare med material-påfyllnad, gående och kvalitet och för dessa områden så upprättades en kravspecifikation.
En idégenerering användes för att skapa så många idéer som möjligt för de olika områdena. Idéerna blev sedan till tidiga koncept som i sin tur utvärderades med hjälp av viktningsmatriser och feedback från anställda. Koncep-ten utvecklades ytterligare till tre slutliga koncept, Första Våningen, Andra Våningen och Integrations konceptet. Koncepten visar att det är möjligt att minska både arbetstiden med 4,5 timmar samt gåendet genom förbättrat informationssystem. Vidare rekommendationer är att Essity fortsätter arbeta med resultaten från både kartlägg-ningen och de slutliga koncepten. Rekommendationen är att göra ytterligare en utvärdering och inkludera operatörerna i utvecklingsprocessen. Efter det bör resultatet i den här rapporten användas som grund för vidare implementering och framtida projekt.
NYCKELORD: Produktionsutveckling, Ergonomi, Säkerhet, Automatisering, Digitalisering, Informationssystem, Kartläggning, Slöserier.
CONTENT
1 INTRODUCTION
2
1.1 Background 2 1.2 Stakeholders 2 1.3 Objective and aims 2 1.4 Project scope 3 1.5 Thesis outline 3
2 COMPANY DESCRIPTION
5
3 THEORETICAL FRAMEWORK
7
3.1 Industrial design engineering 7 3.2 Lean manufacturing 7 3.3 Industry 4.0 9 3.4 Ergonomics 10 3.5 Human-machine interaction 11 3.5.1 Information 11 3.5.2 Alarm system 11 3.5.3 Design principles 12 3.6 User Experience 12
4 METHODE 14
4.1 Process 14 4.2 Planning 15 4.3 Mapping of the current state 15 4.3.1 Non-structured mapping 15 4.3.2 Structured mapping 16 4.3.3 Decision 18 4.4 Literature 18 4.5 Benchmarking 19 4.6 Ideation 19 4.7 Requirement Specification 20 4.8 Concept evaluation 20 4.9 Final concept development 20 4.10 Method discussion 215 CONTEXT
24
5.1 Material refill 24 5.1.1 Glue 24 5.1.2 Pulp 25 5.1.3 Plastic bags 25 5.1.4 Cardboard boxes 26 5.1.5 Second floor 26 5.2 Quality tests 27 5.3 Production stop 28 5.4 Working environment and ergonomics 29 5.4.1 Incidents rapports 29 5.5 Analysis of context 29 5.5.1 Walking 29 5.5.2 Alarm 31 5.5.3 Ergonomic analysis 31 5.6 Survey results 32 5.7 Benchmarking 32 5.8 Ongoing projects 33 5.9 Summary of context 33 5.10 Requirement specification 346 CONCEPT DEVELOPMENT
38
6.2 Early concepts 38 6.2.1 Glue 38 6.2.2 Plastic bags 38 6.2.3 Cardboard boxes 38 6.2.4 Second floor 39 6.2.5 Walking/Alarm/HMI 42 6.2.6 Quality testing 43 6.3 Concept evaluation 43 6.3.1 Analysis 45 6.4 Final Concept development 45 6.4.1 Ground floor 45 6.4.2 Second floor 50 6.4.3 Integration 507 DISCUSSION AND CONCLUSION
54
7.1 Discussing the final result 54 7.2 Relevance 55 7.3 Recommendation 55 7.4 Objective and aims 55 7.5 Conclusion 56 7.5.1 Research question 1 56 7.5.2 Research question 2 56
LIST OF FIGURES
Figure 1. Layout over one of the Baby machines, Fenix 2. 5 Figure 2. Factors that effect production development. 7 Figure 3. Distribution between direct value and waste, according to Sörqvist (2013). 8 Figure 4.Human motion according to Pascal (2007). 8 Figure 5. Colors meaning according to Osvalder and Ulfvengren (2011). 12 Figure 6. Project circle, based on Karlsson et al. (2013). 14 Figure 7. Overview over some of the areas in the context. 24 Figure 8. Glue tank, closed and opened. 24 Figure 9. Flowshart over the change of pulp. 25 Figure 10. The conveyor with plastic bags. 25 Figure 11. Material on the unwinder and storage on the second floor. 27 Figure 12. Flow chart of the work process. 26 Figure 13. Tensile strengths. 27 Figure 14. Photo taken during a RIS-stop with maintenance. 28 Figure 15. Mapping of an operator during 60min. 30 Figure 16. Mapping of an operator during 40min. 30 Figure 17. Operators movement during a 15 min production stop. 30 Figure 18. 30 min work on the second floor. 30 Figure 19. Histogram over the steps, converted to eight hours. 30 Figure 20. Histogram over steps recorded. 31 Figure 21. The yellow areas display where it is possible to directly see the alarm lamp. 31 Figure 22. Refilling glue in basket. 31 Figure 23. Emptying the basket of glue in the tank. 32 Figure 24. Concept Cardboard boxes 1. 39 Figure 25. Concept Cardboard boxes 2. 39 Figure 26. Concept Cardboard boxes 3. 39 Figure 27. The existing layout of Fenix 1 and Fenix 2. 40 Figure 28. The new layout with conveyor in the edges. 40 Figure 29. Different placements of the robot. 40 Figure 30. Layout for the early concept, with extended back and front. 41 Figure 31. Early concept Monitor. 42 Figure 32. Mobile device scanning a QR-code. 42 Figure 33. Example of the alarm lamps. 43 Figure 34. Function of the scissor lift. 46 Figure 35. Movement of the pallet in the automated system. 46 Figure 36. Layout of the second floor. 47 Figure 37. Visualization of the two lifts with small conveyors in both ends. 47 Figure 38. The two Gantry robots in the front of the second floor. 47 Figure 39. Closer view of the gripper. 47 Figure 40. Storage (blue box) to avoid contamination. 48 Figure 41. Material and operators flow on the second floor. 49 Figure 42. Example of control room 50 Figure 43. Three of the four glue tanks on the second floor. 50 Figure 44. Monitor on the ground floor 50 Figure 45. Monitor on the second floor. 51 Figure 46. Visualization of the placement of the monitors, and the visibility (yellow areas). 51 Figure 47. Alarm lamps in different colors. 51
APPENDICES
APPENDIX A. Gantt scheme 1p
APPENDIX B. Number of steps 2p
APPENDIX C. REBA analysis 2p
APPENDIX E. Survey questions 2p
1
INTRODUCTION
This chapter presents an introduction to this master thesis, including background,
stakeholders, objective and aims, project scope and the thesis outline.
With today’s higher demands on effective production systems not only the machines, but also operators, must work as efficiently as possible. This master thesis proj-ect was made in corporation with Essity Hygiene and Health AB at Falkenberg Plant. Essity is a leading global hygiene and health company that develops and produc-es solutions within different businproduc-ess areas. Falkenberg Plant produces incontinence products and baby care such as diapers under the brands Libero, Drypers and Pequenin.
Falkenberg Plant has eight highly automated production lines and three operators on each line that monitor, re-fill materials, handle production stops, etc. This project aims at mapping operators’ work with focus on non-val-ue adding work, ergonomic and safety. The project will also focus on material handling, material flow and waste management related to the Baby machines (machines producing diapers).
This master thesis project is part of the Industrial Design Engineering program at Luleå University of Technology. The course, A7009A covers 30 credits and is conducted over 20 weeks in the spring of 2018.
1.1 Background
Today’s new technology provides endless opportuni-ties to automate and digitize operations. An increasing number of companies replace manual tasks with new technologies, aiming at increasing efficiency and pro-ductivity. In order for Essity to continue to be one of the leading manufacturers and maintain their financial ability, they need to take a step in the right direction. This project has been planned for a long time but has not been implemented due to time and resource short-age. Essity continuously performs improvement work in their production and they have a great habit of planning and implementing developments.
The operator’s work is sometimes very stressful and they do not have time to do quality checks and improvement work. It is primarily all those moments, non-value add-ed work that they wish to remove. In the future, they want to focus more on improvements and ensure that operators is provided with a less stressful work environ-ment where improveenviron-ment work is a part of everyday work.
1.2 Stakeholders
The stakeholders are divided into primary and second-ary stakeholders. The primsecond-ary stakeholders are those who will mostly, and directly be affected by the project. Operators are those who perform many of the manual tasks that hopefully can be automated and digitized in the future. New tasks require operators who need to adapt and probably learn new ways of working and new technologies.
Management will be affected, as they will need to make decisions on how to proceed with the project and possible implementations. They will also be needed to make economical decisions, for example, investment costs and education for the operators.
The secondary stakeholders will be indirectly affected by the project.
Maintenance staffs. New technology may require maintenance staff, like electricians and mechanics to learn new systems. New technology may require ex-tra maintenance and be an additional work task for the maintenance staff.
Suppliers. A new system/product may require suppliers to adapt their product or way of delivering the product.
1.3 Objective and aims
The purpose with the project is to propose ways to reduce the amount of manual non-value adding work performed by operators on the Baby machines with pri-ority on time losses, ergonomic and safety. The expect-ed outcomes are proposals around realistic actions that can be taken to reduce the amount of manual tasks per-formed by operators on the Baby machines and provide concepts for better material handling.
Aim 1: The project shall deliver realistic concepts that reduce manual non-value adding tasks performed by operators through automation, and find new ways to quality-proof the material.
Aim 2: The project will optimize material handling with focus on raw materials, material flows and wastes.
The project will work according to research questions, which will affect the project’s theory and methods. The questions are:
1. How can unsafe and non-ergonomic tasks be elimi-nated using automation and digital systems?
2. How can manual non-value added work be re-moved, and how much time would be saved per shift?
1.4 Project scope
The project is ongoing for a limited period of time, and the 20 weeks result in 800 hours of work. As this is a comprehensive survey of work tasks and material flows, delimitations need to be done in order to focus the proj-ect on what is important.
Baby machines. The project will only study the Baby machines and not the machines producing incontinence products. As there are eight different Baby machines, the project will mainly focus on one of them, Fenix 2. Implementation. The project will not implement any bigger solutions, mainly due to lack of time. Small im-provement proposals may be implemented and tested during the project.
Economy. The project will not make any financial de-cisions about the solutions, only recommendations to Essity. The project will also not take offer suggestions or focus on what the solutions may cost.
1.5 Thesis outline
The thesis is structured in the same chronologic order as the project was performed, from this introduction chap-ter to the final result.
Chapter 1 gives an introduction to the project where objectives, aims and project scope are established. The chapter also presents the background to the project as well as the stakeholders that will be affected.
Chapter 2 presents a company description of Essity with focus on Falkenberg plant. The chapter visualizes a lay-out of the Baby machine Fenix 2 together with an ex-planation of the different parts of the machine.
Chapter 3 presents the theoretical framework, which is the scientific foundation of this thesis. The chapter is introduced with a short description of industrial design
engineering followed by relevant theory related to the project.
Chapter 4 describes the process and the planning of the project, followed by the methods used for mapping the current state and relevant literature. The chapter tinues to describe the method used for developing con-cepts and evaluation.
Chapter 5 presents the context at Fenix 2, one of the Baby machines, with focus on non-value adding activi-ties. The chapter describes different work tasks followed by the time spent on each task. The chapter ends with an analysis of the context.
Chapter 6 presents the results of the concept develop-ment where ideas are transformed into several more de-tailed concepts. The concepts are then evaluated and the chosen concepts are developed in detail.
Chapter 7 includes a discussion of the final result, the relevance of the project and the recommendation. The chapter also includes a conclusion and how well the project achieved the objectives and aims stated in the beginning of the project.
2
COMPANY DESCRIPTION
This chapter presents a description of the company, Falkenberg plant
and the Baby machine Fenix 2.
Essity used to be a part of the SCA group, founded in 1929, but since June 2017 Essity is an independent company with focus on the essentials for everyday life. Essity is a leading global hygiene and health company with 48000 employees (in 2017). Sells are conducted in 150 countries under the brands Jobost, Leukoplast, Libero, Libresse, Lotus, Nosotras, Saba, Tempo, Vinda and Zewa (Essity, 2018). They develop and produce prod-ucts within their different business areas Personal Care, Costumer Tissue and Professional Hygiene. This project focus on Personal care that includes incontinence prod-ucts, feminine care, baby care and medical solutions. In 2016 Personal Care accounted for 33% of Essity’s net sales, which is approximately SEK 43bn (Essity, 2018). In the end of 2017 Essity had production at 27 plants in 21 countries (Essity, 2018).
Falkenberg plant produces Baby care on eight Baby machines and incontinence products on two inco ma-chines. This thesis focus on the operators work on the Baby machines and the following text will dive deeper into the work at the Baby machines. Falkenberg plant has around the clock production and the operators work in five different shift teams. The shift is based on a schedule where the operators work for seven days in a row followed by five vacant days. The shift varies in time from seven to nine hours, but during their working week they work mostly for eight hours. The around the clock production results in the operators taking turns for break, leaving only two operators at the Baby ma-chine during this time.
The Baby machines consist of a highly automated pro-duction line that assembles the raw material into a di-aper in a high speed. The didi-apers can be considered a simple product, but it contains twelve different
mate-rials. It is the operators’ responsibility to refill material, do quality checks, monitor, handle production stop and make sure that the machine produces the right order. The variety of work tasks requires a lot of experience, knowledge and good communication with each other. Figure 1 is a visualization of one of the Baby machines, Fenix 2, observe that the picture is not made to scale and just a visualization over the area.
The production line is where the diapers are assembled together. Over the production line is a second floor that holds eleven different materials placed on unwinders (a construction that allows the raw material rolls to ro-tate). The material on the second floor goes through the floor and down to the production line, where the diapers are assembled. After the assembly the diapers are cut into correct size and folded. The Bagfiller takes the folded diapers, sort them, compress them and pack the correct amount of diapers in plastic bags. The plastic bag closes by a welding and the bags leave the Bagfiller on a conveyor. The plastic bags then enters the Casepacker where the plastic bags are packed into cardboard boxes. The boxes then leaves the Casepacker on a conveyor that transports them to a warehouse. The test stations is where the operators perform quality tests.
Green = Container containing different kinds of rub-bish.
Red = The HMI (main computers), this is where it is possible to view and reset alarm.
Orange = Control panels over specific parts of the ma-chine.
Dark grey = Material that the operators refills. Blue = Glue tanks.
Yellow = Automatic truck (to pulp refill).
3
THEORETICAL FRAMEWORK
This chapter contains the theoretical information that the project concerns. The choice of
information is based on the areas of the project and literature in the field of industrial design
engineering. The information is presented to support and strength decisions and arguments.
3.1 Industrial design engineering
The past few decades have, according to Johannesson, Persson and Pettersson (2013), been characterized by new requirements for customization, sustainability, safe-ty and design. Global competition is getting tougher and products are becoming increasingly more complex. Al-ready during the industrialization, the ability to compe-te was important and companies tried to attract custo-mers using appealing design and quality. In the 1950s and 1960s, a new multidisciplinary science was develo-ped, the ergonomics (Johannesson, et al., 2013). Initially, the focus was mostly on human physical characteristics, followed by research on information intake and percep-tion (Johannesson, et al., 2013). Today, the industry has a market-oriented and customer-oriented view and de-velops its products and production system for reasons of competition. The challenge is to give each customer a custom made product with retained efficiency in manu-facturing and maintaining economy.
According to Johannesson et al. (2013) the industrial innovation process includes all activities that must be taken to realize a business opportunity by developing, manufacturing and launching a product on an identi-fied market. This can be a physical product, service or both. According to Bellgran and Säfsten (2005) there are several factors that affect the production system. They talk about external influences, actual options during the development and strategies and fundamental attitudes among the people involved in the development (figure 2).
Figure 2. Factors that effect production development.
The external influences that affect the production sys-tem could be history trends and globalization (Bellgran & Säfsten, 2005). Nationally, a number of important factors affected the transformation of production sys-tems in Sweden. Such factor is, according to Bellgran
and Säfsten (2005), the sociotechnical thoughts that characterize the work content for the operators. By actual options Bellgran and Säfsten (2005) means that the existing technology will affect the outcome of the production development. The company’s strategies and values for the people involved are also crucial for the development of the production system and how the end result turns out.
3.2 Lean manufacturing
Lean production is a widespread philosophy with the roots b from the Toyota Production System (Pascal, 2007). Lean is about developing effective flows and creating value for the customers, businesses and orga-nizations in several industries strive to be lean (Sörqvist, 2013). According to Sörqvist (2013), the goal is to achieve high efficiency and maintain satisfied customers. By eliminating non-value-adding work, time is liberated that can be used to create higher customer value. Those who succeed achieve, according to Sörqvist (2013), fine results, but far from everyone achieve success. At first glance, Lean can appear a simple concept, but according to Sörqvist (2013), it’s easy to be fooled, since lean re-quires a lot from the business and is a long-term work. So why Lean? In today’s business, there is a great-un-tapped potential, where substantial resources are put on things that do not create value for the business customers (Sörqvist, 2013). Lots of time and money are spent due to errors, shortcomings, dissatisfied customers, delays, bad practices and unnecessary work, just to name a few. Lean is a relevant area for this project as Essity strives to minimize waste and get a more lean production.
Waste
One fundamental part in lean manufacturing is to re-duce the “seven wastes” also called Muda in Japanese. The wastes are according to Bellgran and Säfsten (2005):
1. Overproduction means producing more than the costumer wants, producing ear-lier than required or producing overlarge batches.
2. Excess inventory means holding more buffers/inventory than required.
3. Waiting means that time is not used ef-fectively in the production and waiting ap-pears as an effect of not having the right conditions.
4. Transportation adds no value to the
tumers, except for transportation to the costumer.
5. Unnecessary motion means a motion that does not create value for the product. 6. Over processing means performing more
work than the costumer requires.
7. Defect product causes waste and what was not done right at start needs to be cor-rected.
An addition to the seven forms of waste mentioned above is untapped competence (Petersson, Johansson, Broman, Blucher, & Almsterman, 2010). By not using the competence of the employees, there is a risk of loos-ing improvements that would have been possible to im-plement. Muda is the opposite of value, and according to Pascal (2007) the costumers are not willing to pay for wait time, rework, extra inventory or other forms of Muda. According to Sörqvist (2013), a central part of lean work is to systematically review the processes, focusing on identifying non-value-adding work. The outcome of a value-flow analysis often leads to the fact that only a few percent are used for direct value added work (Sörqvist, 2013). Figure 3 illustrates how the time distribution between direct value added work, necessary but non-value added work and waste (Sörqvist, 2013). Sörqvist (2013) divides non-value added work into two parts, where necessary non-value-added activities are required for the value added work to be carried out, but as such does not add a direct value to the custom-er. Waste, on the other hand, neither adds value nor is necessary for other reasons (Sörqvist, 2013). This may include movements and movements that happen unnec-essarily. The necessities of non-value-adding activities are usually more difficult to eliminate because basically they are required for the business to be run. Primarily, it is therefore more common to focus on clean wastes (Sörqvist, 2013).
Human actions or in this case, operators work, can also be divided into three categories described by Pascal (2007), see figure 4.
* Actual work adds value to the product
* Auxiliary work supports actual work, usually occurs before or after actual work.
* Muda is the motion that creates no value.
Wasted human motion is linked to workplace ergo-nomics where poor ergonomic design negatively affects productivity, safety and quality (Pascal, 2007). Produc-tivity suffers when there is unnecessary walking, twist-ing and reachtwist-ing and the ergonomics has likely the big-gest impact on safety (Pascal, 2007). Pascal (2007) also claims that quality suffers when operators has to strain to process or check a work piece because of twisting and reaching.
Jikoda
Jikoda is a Japanese term ensuring that everything is done correctly from the start of the process, and if some-thing goes wrong the process should stop (Petersson, et al., 2010). Build in quality can be applied in automatic processes where there is a possibility to automatically correct and ensure high quality. According to Petersson
et al. (2010) the most important condition to achieve a high-quality process is to use employees with appro-priate competence that complies with agreed ways of working. An automatic process on the other hand often provides the possibility to correct the process automati-cally in order to ensure high quality.
Poka-yoke is another Japanese term referring to mis-take-proofing which are devices that makes it near-ly impossible for operators to make a mistake/error (Liker, 2004). Poka-yoke means implementing simple and low-cost devises to find abnormal situations be-fore they occur, this reduces the worker’s physical and mental burden (Pascal, 2007). Example of errors could be missing process steps, process errors, missing parts, wrong parts, adjustment errors and equipment not set up properly just to mention a few (Pascal, 2007). A good poka-yoke on the other hand is simple, with long life and low maintenance, high reliability, low cost and de-sign for the workplace conditions (Pascal, 2007). When a poka-yoke notices an error it should according to Pas-cal (2007) either deliver a warning or shut down the machine. Poka-yokes alert abnormalities by activating a warning signal or light or shutting down the machine when operations fails, pieces are incorrectly positioned, sensors fails and so on (Pascal, 2007).
5S
5S is a visual management tool designed to create a work environment that is self-explaining, self-oriented and self-improving (Pascal, 2007). 5S stands for sort, set in order, shine, standardize and sustain (Pascal, 2007). This method is often the staring point when a company strives for a lean production since the method is consid-ered simple to use (Petersson, et al., 2010). A well-orga-nized workplace is the basis for eliminating waste and standardizes work (Petersson, et al., 2010). The first step of visual order is to sort and make sure that the work place does not get overrun with stuff since this impairs the flow of work (Pascal, 2007).
Criticism
According to Sörqvist (2013), two main reasons can be given to the fact that lean is criticized and questioned. First of all, many lean initiatives do not meet the expec-tations from the companies and therefore the concept is considered to be less successful. Sörqvist (2013) believes that this is due to lack of management support, less well implemented plans, excessive cost focus, poor planning and insufficient resources. Secondly, criticism of lean
is directed to the fact that investments are claimed to have adverse side effects in businesses using the con-cept’s methodology. According to Sörqvist (2013), these effects can be stress, occupational injuries, monotonous work and negative gender perspectives. Chay, Xu, Ti-wari and Chay (2015) on the other hand claims that one of the most critical elements to succeed with a lean transformation is to have a people-orientated produc-tion, something that often fails.
Common barriers that hinder lean transformation are issues surrounding shop floor workers, where companies fail to engage the workers into lean activities. Chay et al. (2015) claims that top-down approaches is the most common approach, and botton-up approaches that em-phasis more of the shop floor workers initiatives often is neglected. The top-down approach often used is insuf-ficient when transforming conventional manufacturer to become lean manufacturer (Chay, et al., 2015). The bottom-up approach is therefore important to succeed and sustain a lean manufacturing/organization. Manag-ers should be more aware of the impacts of Lean pro-duction on complex socio-technical systems and appre-ciate the upside of complexity (Marlon, Tarcisio Abreu, & Michel Jose, 2018). According to Marlon et al. (2018) their study showed that successful lean implementations relied on both increasing and taking advantage of com-plexity, instead of reducing it.
3.3 Industry 4.0
Already in the 1970s, information and communications technology were implemented in industries. Industry 4.0 was presented for the first time in 2011, and be-came a German strategic initiative. Since then, similar strategies have been proposed in other countries under the names like “Factories of the Future”, “Industrial Internet” and “Internet + ” (Mrugalska & K. Wyrwic-ka, 2017). Despite that Industry 4.0 is a widely spread concept there is no formally respected definition for it (Mrugalska & K. Wyrwicka, 2017).
New technology is required in order to convert the ex-isting production system to Industry 4.0.Wang, Wan, Li, and Zhang (2016) states that Internet of Things (IoT), big data, artificial intelligence (AI) technologies and cloud solutions are enabling factors for industry 4.0. Us-ing AI technologies together with powerful micropro-cessors results in smart machines and products, meaning that they communicate, control and compound with each other (Wang, et al., 2016). These smart artifacts are
interconnected with each other and the industrial net-work/internet. Cloud computing technology can, to-gether with the server network be used as a resource pool that can provide scalable computing ability to stor-age space on demand for big data analytics (Wang, et al., 2016).
Mrugalska and K. Wyrwicka (2017) describes that In-dustry 4.0 can be defined by three ideas, the Smart Product, the Smart Machine and the Augmented Oper-ator. The Augmented Operator is an interesting area, as the operators at Fenix 2 will be affected by future im-plementations. An Augmented Operator is about auto-mating knowledge and information in order to provide a flexible and adaptive production system. The operators will then be faced with a variety of tasks such as speci-fication, monitoring and verification of the production system. To ease the work the operators could be provid-ed with a mobile user focus assistance system (Mrugal-ska & K. Wyrwicka, 2017). Kolberg and Zühlke (2015) states the same thing as Mrugalska and K. Wyrwicka (2017), who also say that the operators can be equipped with smart watches that receive error messages and error location. Compared to signal lamps, recognizing failure does not depend on the location of the operator (Kol-berg & Zühlke, 2015).
As the technology becomes more complex the impor-tance of good user interfaces increases. Smartphones, Bluetooth, navigations systems are just a few of the system that already exist in our everyday life. This will according to Schmitt, Meixner, Goreck, Seissler and Lo-skyll (2013) change the traditional way of working in the industry. They introduce several different platforms and methods that can be used to improve the interaction as the complexity of the automation increases.
• Smartphones can be used in different ways in the production system. The phone can control diverse field devices via Bluetooth. Information is presented clearly and completed by the respective operators signa-ture. It is also possible to use QR-codes to provide in-formation. By scanning the codes operators are provided with needed information of a specific area, the QR-code contains passive data with needed information and can be decrypted by the phone.
• Seamless Navigation Application displays dif-ferent maps appropriate to the situation. The transition between the different maps happens seamlessly as the
applications continuously provide the current situation of the maintenance to the worker. Locality information can also be used to create innovative maintenance func-tions decreasing the time-consuming search of defective areas.
• Augmented Reality (AR) has a great potential for presenting information and knowledge to support employees. AR uses an extension of the perception of reality by overlaying of virtual objects. This provides task-based information that can be projected interac-tively and immediately to the employee.
Even though the complexity of Industry 4.0 is groov-ing, the system has many advantages and huge potential. According to Mrugalska and K. Wyrwicka (2017) it in-creases competiveness and flexibility, optimizes decision making, increases resource productivity and keeping productive workers for longer as the systems provide flexible and diverse career paths, just to mention a few. Mrugalska and K. Wyrwicka (2017), believes that Indus-try 4.0 can be linked to Lean production. Augmented Operators can apply Jikoda, a Lean method used for quality controls as they reduce time between failures with help from notifications. The notification should contain information concerning both the error message and error location. The information can then be logged for further analysis as a part of continuously improve-ments (Mrugalska & K. Wyrwicka, 2017).
3.4 Ergonomics
Ergonomics is usually considered an important part in the design process (Magrab, Gupta, McCluskey, & Sandborn, 2009). According to Magrab et al. (2009) the product/system should be easy to use and effective-ly communicate its function, which results in a shorter learning time. When designing a workplace it is import-ant to consider the physiology of the human that are going to use the workplace (Hägg, Ericson & Odenrick, 2011). The relationship between a workplace design and the dimensions of the body determines the position of the body that is taken at work. How much your body position is limited depends on how much contact there is between the person and the environment. According to Hägg et al. (2011), the contact may be both physical in the form of furniture or visual in the form of moni-tors. Ergonomic consideration increases the chances of success amongst the users and facilitates maintenance for the end users. Ergonomics play an important role
regarding safety and minimizing injury (Magrab, et al., 2009).
Physical loads may be of different character and depend on amplitude, frequency and duration. The human body needs physical loads and rest to maintain the various functions of the body, even varied mental stress and so-cial contacts are needed to maintain normal function-ing and health (Hägg, et al., 2011). Physical loads can in some cases cause damage to the body’s tissues and organs (Hägg, et al., 2011). The work must therefore be carried out to provide load variations with varying light and heavy loads and with sufficient breaks.
3.5 Human-machine interaction
The cognitive ability and limitations of human beings in different situations need to be taken into account in the development of interfaces. This applies for both audio and visual information. Design aims at providing a prod-uct with interpretable information using form and color (Johannesson, et al., 2013). Understanding characters is primarily about using our visual sense so that we cor-rectly understand both the basic characters, archetypes and symbols (Johannesson, et al., 2013).
3.5.1 Information
Operators get their information in different ways, through information on displays, sounds, lights, indica-tors, etc. When designing user interfaces, the informa-tion presented should enable the operator to make deci-sions and sort information in an easy way. According to Osvalder and Ulfvengren (2011), the operator’s physical and mental properties must be the basis for the design of the user interface. Technical systems are becoming in-creasingly complex and it is through the user interface that the operators communicate with the technology. Osvalder and Ulfvengren (2011) also state that lack of understanding of technology can lead to human error and incidents or, at worst, accidents. According to Os-valder and Ulfvengren (2011), there are data indicating that 60-80% of all accidents in complex technical sys-tems are affected by the human factor.
According to Pascal (2007) there are four types of visual systems and their purpose is to share information at a glance. The most effective way of sharing information is to use a visual indicator for example a street sign, the second most effective way is to use a visual signal for ex-ample a traffic light. On third place is the visual control that limits behavior, for example parking lot lines and on
forth place is, according to Pascal (2007) guarantee that allows correct response only.
According to Osvalder and Ulfvengren (2011), there are three different categories of information, qualitative, quantitative and representative information. Qualitative information devices are suitable for use when only a few different states are displayed. When designing, it is important that the indicators differ as much as possi-ble from each other. Audio indicators can be used as an addition and the advantage is that sound is picked up even though the information device cannot be ob-served. Qualitative information devices are used when operators need numerical and accurate data, the infor-mation should be displayed in digital or analog form. Analog reading is faster to read, but according to Os-valder and Ulfvengren (2011), digital display is preferred when showing exact numbers. In design, it is important that the information device is easy to read and the scales should be as clear as possible. Representative informa-tion displays a picture of reality, such as a machine or process. It is important that symbols are logical and easy to read since representative, alphanumeric, qualitative and quantitative information is often displayed simul-taneously.
3.5.2 Alarm system
According to Osvalder and Ulfvengren (2011), the pri-mary purpose of an alarm system is to alert operators to deviations in the system so they can correct the er-rors and handle the interference. According to Thunberg and Osvalder (2008), an optimal alarm system will lead the operator to find the right information and enable early detection of deviations. Osvalder and Ulfvengren (2011), also say that the alarm must give the operators time to respond before it is too late and confirm actions from the operators. The alarm should be prioritized ac-cording to time and consequence.
According to Thunberg and Osvalder (2008), sever-al criteria’s needs to be considered when modernizing alarm systems. Future systems must be flexible and in-formation flow should be adapted to different operat-ing modes. Operators’ cognitive and physical conditions should also not be overloaded. According to Osvalder and Ulfvengren (2011), it is also important to consid-er how the old system looks when designing the new interface, so that both old and new are perceived as a whole.
3.5.3 Design principles
When designing displays it is important to have knowl-edge of the human cognitive abilities so that the design supports the user with relevant information.
Design principles for attention
To increase strengths and reduce weaknesses in human attention, the designer can, according to Osvalder and Ulfvengren (2011), minimize the time it takes to find information. Searching information is time consuming and a waste of resources and therefore should the infor-mation used most often be shown first. Inforinfor-mation be-longing to each other should be grouped and found in the same place on all displays (Osvalder & Ulfvengren, 2011). When two or more screens are integrated, atten-tion must be shared and it is not always advantageous to apply closeness between different objects. Instead, Os-valder and Ulfvengren (2011) suggest that putting the same color on things that belong together and linking them with arrows and lines should solve this. For in-formation from multiple sources of inin-formation, using both visual and auditory stimuli can facilitate the acqui-sition of information.
Design principles supporting perception
Good readability is the basis of good information where high contrast, good lighting and the right viewing angle should be used to succeed. According to Osvalder and Ulfvengren (2011), it is difficult to judge the size, thick-ness or color if the number of levels that the variable can assume exceeds five. There is a greater chance that the information will be interpreted correctly if the same message is presented in several different ways, for exam-ple, messages can be presented both visually and audibly. Osvalder and Ulfvengren (2011) states that similar ob-jects should be avoided since they easily create confu-sion, the objects shown should be clearly distinguishable. Design principles supporting memory
Osvalder and Ulfvengren (2011), states that short-term memory has a limited capacity and very little informa-tion can be kept active and processed at the same time. Furthermore the design of displays should not burden the short or long-term memory for operators, as this makes it harder to solve problems. According to Os-valder and Ulfvengren (2011), current information must be shown on displays so that the short-term memory can focus on problem solving instead. To predict events and think proactively is a difficult cognitive task, if one has to think about the current status of the system. Since
proactive behavior is more effective than retrospective, the interface should support the operator with clear in-formation about expected status (Osvalder & Ulfven-gren, 2011). When designing new interfaces, it should be taken into consideration, for example, the same col-ors should be used throughout the system (Osvalder & Ulfvengren, 2011).
3.6 User Experience
According to Osvalder and Ulfvengren (2011), colors should be used carefully and consistently in an interface. Colors carry different messages and how they accrue depends on culture and experience. They believe that an interface should not contain more than four differ-ent colors, where red and green combinations should be avoided due to colorblindness. Pettersson et al. (2004) agrees with Osvalder and Ulfvengren (2011), that a combination of red and green should be avoided. Os-valder and Ulfvengren (2011), writes that in the western world, the following stereotypic color coding is usually used, see figure 5. Pettersson et al. (2004) on the other hand claims that red stands for love, passion, heat, blood, revolution and yellow represent heat, sun and Easter.
Figure 5. Colors meaning according to Osvalder and Ulfvengren (2011).
Symbols can be used with advantage in the design of in-terfaces as long as using well-known and clear symbols. Symbols can either be representative which graphically resembles reality or abstract. According to Osvalder and Ulfvengren (2011), symbols have some advantages over text because symbols are easier to find far away, infor-mation is detected faster, less error margin, and they are not dependent on a language.
COLD, WATER &CALM
DRIVE, CONTINUE & ON
WARNING, SLOW &
TESTING
STOP, DANGER, HEAT &
FIRE
4
METHODE
This chapter describes an iterative work process, from the first to the last step
based on theoretical approaches. Mostly of the work was performed at Essity with
continuous meetings and feedback from the supervisor at Essity.
4.1 Process
The product development process is according to Jo-hannesson et al. (2013) characterized by I3, which are
iteration, integration and innovation. This project uses an iterative work process, which means that the project does not follow a linear order. According to Wikberg Nilsson, Ericson, & Törlind (2015), iteration is used to ensure that user needs are met, which means continually returning to the cause of the problem and increasing the understanding of the users situation.
The advantage of the iterative process is that the devel-oper is always ready to rethink and take a step back in the process (Wikberg Nilsson, et al., 2015). According to Karlsson, Osvalder, Rose, Eklund and Odenrick (2011) the workflow for a development work can be seen as a learning cyclical process and can be used for developing products as well as for developing of production systems. They agree with Wikberg Nilsson et al. (2015) that a cyclical way of working leads to a good overall approach in a short period of time and the chance of ending right increases due to the iterative method. The development process can be initiated in a number of steps where each step leads closer to the target. The iteration within each step continues until the evaluation shows that the re-sult is satisfactory. According to Karlsson et al. (2011),
the development work is continuous and parallel, the longer the development work is going on, the more the knowledge increases and the project approaches a more concrete solution.
The project circle described by Karlsson et al. (2011) includes eight steps, from planning and identifying user needs to evaluation and development (see figure 6). Due to the iterative process the project cycle are re-peated, usually two to three times, until the outcome are satisfying the needs. Ulrick and Eppinger (2012) on the other hand describe the design development process in six stages involving planning phase, concept develop-ment, system developdevelop-ment, detail developdevelop-ment, testing and further development as well as production uptake. Ulrick and Eppinger (2012) describes that the design process can be used in different ways depending on what is being developed. This project will work according to the process described by Karlsson et al. (2011). The six first steps described in their process fits well with the development of this thesis. The reason for not fulfilling the last two steps is mainly due to the limitations in time. The circle was repeated as new information was gained and the project entered different phases. The first cycle was done during the planning and mapping of the
cur-4 METHODE
Figure 6. Project circle, based on Karlsson et al. (2013).
PLAN ! MAKE A DISGNOSIS! FORMULATE GOALS AND REQUIREMENTS! SEARCH OPTIONS! EVALUATE! DEVELOP! CONDUCT PROGRESSIVELY! FOLLOW UP AND VALUE!
rent state. The circle was then repeated once the project entered the concept development phase and once again had to go back to square one. The circle was repeated a third time during the final concept development to make sure that the user needs were in focus. The itera-tive work enabled the project to be open to new infor-mation and requirements, from the start until the end.
4.2 Planning
During the first week of the project a project plan was established. The plan fills several different purposes, part-ly as a basis for timing, while ensuring that all stake-holders agree on the purpose and goals of the project. The plan included background and cause to the project. The project raised research questions together with Es-sity’s supervisor and charted stakeholders for the project. Skrårbratt (2009) states the importance of good com-munication early in the project. In order to focus the project; delimitations, objectives and goals were clari-fied. There are two types of delimitations, delineations that separate the project from ongoing operations and delineations that separate the project with respect to the subject area, area of activities, target group, geographic area etc. (Nordberg, 2008).
According to Wikberg Nilsson, Ericson and Törnlind (2013), the project plan is a living document that is con-tinuously updated on events in the project. The project plan established a Gantt scheme to visualize the given time for the project, the schedule showed the different phases, presentations and submissions of the project. The Gantt scheme is especially useful when you want to see how different types of activities relate to each other (Nordberg, 2008). According to Nordberg (2008), the Gantt scheme clarifies the intentions of the project plan in a very conspicuous and clear way. The projects Gantt scheme can be displayed in appendix A. The different steps in the project were planned and various milestones were presented and deliverable was established. Mile-stones are an important part of the project and it helps to facilitate good communication (Skårbratt, 2009). The project plan also determined how communication should be done between supervisors and how often.
4.3 Mapping of the current state
Case studies have, according to Martin and Hanington (2012) been proposed as a useful tool for designers. The researcher needs to make initial hypotheses, conduct re-search through interviews, observations and other forms of information (Martin & Hanington, 2012). In order
to find improvement areas and understand the opera-tors everyday lives, it was decided that the project would work with the operators for two weeks. No similar proj-ect has been performed at Essity Falkenberg and the operator’s work had never been analyzed as a whole. During the two weeks, the project worked the morning shift from 6:00 to 14:00 and was given the opportuni-ty to work with all five shift teams. The reason for this was to see differences and variations in working meth-ods between different shifts and operators. According to Martin and Hanington (2012), there are two different mapping methods called place-centered mapping and individual-centered mapping. In this project a combi-nation of the two have been used. The mapping was done at Fenix 2 (place-centered) by following operators (individual-centered). Below is a description of how the mapping was carried out and how the operators work tasks were analyzed.
4.3.1 Non-structured mapping
According to Kraft (2012), it is complex to finding users needs and many different methods is necessary to cover all the different needs. To find the immediate user needs the designer needs to talk to the user, and for example do an interview. Kraft (2012) describes a method called “A-day-in-the-life”, where the designer chooses to map a typical day in the users life, which was done at Fenix 2. This is, according to Kraft (2012), a great method to find latent needs as the designer can follow the user, take notes and photos and see how the user interacts with a product/system.
In the beginning of the two weeks the project focused a lot on understanding the variety of work task among the operators. Notes and photos were taken and the op-erators described the different tasks. By talking to the operators and showing an interest for their opinions a good communication was established and the operators started to share information and be open-minded. The overall and less structured mapping was done during the first week. This was done mainly to keep an open mind and not jump to conclusions about what needs to be investigated.
4.3.2 Structured mapping
Based on the result from the non-structured mapping a plan was made for the structured mapping. Mapping is a method that can be used for different aims and ap-proaches, but it is important to be aware of what is
in-tended to describe (Sörqvist, 2013). The plan, created by the project, contained what and how the different work tasks should be mapped in order to get sufficient mate-rial to base decisions on.
A flowchart can, according to Sörqvist (2013), be used to describe and visualize a process. A flowchart consists of different symbols that symbolize work tasks, where the symbols are connected with arrows to show the flow of the work. The best way to get a reality-based descrip-tion of the work, is to involve the employee in the pro-cess (Sörqvist, 2013). The project decided to create flow charts over the tasks that involved many different steps, to be able to get better understanding of the process-es. According to Sörqvist (2013), it is good to combine flow charts with the time it takes to perform a certain task. This was done by categorizing all the different tasks observed into material refill, quality tests and production stop. The material refill was then divided into glue, pulp, plastic bags, cardboard boxes and the second floor. As the second floor involves many different tasks, the tasks were divided once again to make it easier to take time. The flow chart that was created of the second floor be-came the basis for the timekeeping. The project choose to divide the task “changing material” on the second floor as all the tasks were not performed at once and all operators have their own way of working. The han-dling off collets and waste is usually just a few seconds at a time and for these areas it seemed more efficient to clock how much the operators are handling the material during a certain time. Many of the tasks, like refilling glue, pulp, plastic bags and cardboard boxes only occurs a few times per shift and therefore the project was able to study several tasks at the same time.
Interviews are one of the most basic methods of gath-ering information about human’s opinions (Osvalder, Rose & Karlsson, 2011). Martin and Hanington (2012) state that interviews are a fundamental research meth-od to verify and humanize data collected from other methods. Through interviews, understanding of people’s experiences, values, opinions, and how people reason. Interviews gather user opinions, which means that the result provides subjective data. This subjective meth-od can be seen as a complement to objective observa-tions. Osvalder et al. (2011) categorizes interviews in three different categories, structured, semi-structured and unstructured interviews. According to Osvalder et al. (2011), an unstructured interview is preferred if
you want qualitative data as open questions are asked. Combining interviews with observations gives informa-tion about the operator’s opinions and the way they act. Observations are an objective method used to gather information about how people behave in different situ-ations or in an interesting event (Osvalder, et al., 2011). The method is used to see how the user, in this case the operator acts in a certain situation, thus providing knowledge of how a task is performed. According to Osvalder et al. (2011), the purpose of observations is to get an understanding of the incident situation in the natural environment. A part of the observation was to find non-ergonomic postures and make an evaluation of the risks.
According to Martin and Hanington (2012), ergonomic analysis can be used as a predesign analysis of the work-ing environment. One of the best-known methods for analyzing work-related body postural risks is, according to Osvalder et al. (2011), REBA (Rapid Entire Body Analysis). The REBA analysis consists of four steps start-ing with choosstart-ing which postures are to be analyzed. This can be done in three different ways, analyzing bad postures, analyzing from time samples (for example, take a posture every 10th second) or by doing a tasks analysis, which can be done by breaking down the work into smaller tasks. The second step is to collect data from dif-ferent postures which is easiest when using a video cam-era or taking photos of the chosen postures. The third step is to assess the postures and then put together the results in step four.
Not only does the project focus on reducing non-value added work, it also aims at providing the operators with a better working environment. And in order to do so, it is important to listen to the operators and get their point of view. Unstructured interviews were used as the project constantly asked the operators about their opin-ions and encouraged them to talk about things that they wished to improve. The operators are the ones with the greatest knowledge about their ways of working and therefore the project tried to gain as much information as possible from them.
By observing the operators the project could see how much time the operators spent on different work tasks and find non ergonomic work postures. The project took time using a phone, the time was started when the operators began to preform a task, paused if they got interrupted, and then stopped when the operators were
finished. The time it takes to perform a task depends partly on the operator, but also on the work environ-ment that might causes the operators to perform a task faster due to stress. In order to get reliable data, each work tasks were timed several times performed by sever-al operators at different shifts. How many measurements that was done on each tasks varied due to how often it was performed, but a minimum of five measurements was used. The average times was calculated by adding all the times together and then divide the total with the number of measurements.
During the observation it was made clear that many of the work tasks were non-ergonomic, especially the refill of glue. The project decided to make two REBA ana-lyzes on different postures when the operators refilled glue. To investigate the working environment further a meeting was held with two of the safety engineers where the project had the opportunity to ask them about the safety-and ergonomic work at Essity. This was done to make sure that the projects views on the oper-ator’s works tasks agreed with incident reports and the operators’ opinion.
In order to find out how much time the operators spent performing a certain task during one shift the project needed to know how many times each task was per-formed. To get as accurate result as possible the project decided to calculate the time for the material handling instead of manually calculate all the material delivered to Fenix 2. The project contacted the material planners and they provided Excel lists of all the material that had been delivered to and from Fenix 2. The Excel docu-ment from the material planner included data from seven days of production, and the efficiency of Fenix 2 during this time could be seen as “normal production”. The data from the Excel document was divided and grouped, so that, for example all material to the second floor was summed up. After this all material belonging to the same area were added together to get the total material delivered to Fenix 2 during a week. The weekly production was then divided to material delivered/8h. By doing this, general frequency of different material handling per shift could be established. During this time Fenix 2 produced different orders, had its normal RIS-stop and rescheduling. The result of the total time was then summarized in tables.
Even though the project is not focusing in production stop or the causes of mechanical problems, it is still
im-portant to know how much time this takes during a shift. To get the average time of the production stops, both planned and unplanned the project received an excel document from one of the Quality Engineers at Falkenberg. The document included all stops on Fenix 2 from the 1st to the 31st of March. The document in-cluded number of planned stops, number of unplanned stops, average duration of the stops and total stop time per day. The information received were then summed up and presented in the context.
The times collected by the project were analyzed to-gether with Marcus Lehto, to see if the results seemed reasonable and if there was something missing. It was decided that the project should continue with the map-ping, as there was still a lot of time missing compared with the working hours available. The project would look further into the operators walking and how they interact with the alarm/HMI.
A good way of analyzing a flow is to create a so-called spaghetti diagram, which fits well if the flow is un-clear or there are many different ways to perform a task (Sörqvist, 2013). A spaghetti diagram is according to Sörqvist (2013) developed from a graphical image of the environment where the flow is to be analyzed. Next, you draw lines that represent the physical material, product and/or individual flow. Spaghetti diagrams can be supplemented with facts about times and distances of the area (Sörqvist, 2013). Also Pettersson et al. (2010) states that spaghetti diagrams is a great tool to use when mapping activities.
The project decided to do a few spaghetti diagrams to visualize the operators’ movement at Fenix 2. The proj-ect started with making a graphical image of the layout at Fenix 2 using Microsoft PowerPoint, the advantage of using PowerPoint is that a layout can easily be drawn us-ing simple symbols. The layout was not made to scale as the project mainly wanted to show movement and not spend too much time on measuring up everything in detail. Different operators where then observed during a specific time, and the project drew lines as the oper-ators moved around at the Baby machine. The opera-tors were aware that they were being mapped, but they didn’t know what the project was observing. An analyze of the spaghetti diagrams was made together with Mar-cus Lehto and it was decided that this area needs to be investigated further.
The project needed to quantify the amount of walk-ing and the operators were therefore provided with step counters. The step counters were given to the opera-tors together with an explanation of the propose with the test. The step counters was left with the operators for a week and they were responsible for using them and in the end of the shift writing down the number of steps, distance and hours worked. The step counters were preset on a step length of 75 cm, which is not what reality looks like since all operators have different step length. This was done, as it would have been to time consuming for the operators to reinstall the step count-ers in the start of each shift. Therefore the analysis was made on the number of steps and since the shift varies in time the project converted the steps into steps per eight hours. After analyzing the results the project felt that it was important to include the operators and hear their opinions about the walking and their overall working environment.
Surveys are a subjective method and can be seen as a type of structured interview, without the interviewer present (Osvalder, et al., 2011). According to Osvalder et al. (2011), when creating a survey, it is important to formulate simple and clear questions as the questions can be interpreted unambiguously. Response options or scales can be used in combination with open questions where respondents can provide more detailed answers. The survey was written in a Word document, printed and handed over to the operators with a description. This was done to ensure that the project had perceived the operator’s situation and opinions in the correct way. As they work in shift the surveys were left in at Fenix 2, giving the operators the possibility to answer when time was available. The project then collected the an-swered surveys after two weeks and complied the results in Google Forms, a tool that automatically analyze the answers and provide charts and statistics. In order to use Google Forms the project needed to transfer the an-swers from the printed document into the digital form manually.
In the survey the operators received questions about the alarm at Fenix 2. Based on the results the project wanted to investigate this further and asked one of the electri-cians for help. He came up with a solution that could calculate how many times the alarm lamp was turned on during each shift. This was done to see if the alarm could be a reason for the amount of walking amongst the operators.
As a final step in the mapping the project also wanted to find out about ongoing projects in Falkenberg. This was done to ensure that the project does not do the same thing as any other project, and that Essity would get out the most of this thesis. Marcus Lehto gave the project names of other people that are involved or responsi-ble for ongoing projects. The project then contacted the people involved for a shorter meeting, in this way the project could gain more information for the project and the current state.
4.3.3 Decision
Together with the supervisor at Essity the context was analyzed with focus on both time and ergonomic. A de-cision was made to develop several areas founded in the mapping, with focus on creating improvement concepts for the areas. The tasks that took the most time would be prioritized, as this would benefit the company most. Second priority was to focus on reducing non-ergo-nomic work tasks and improve the environment for the operators.
It was also decided that the project should not develop the concept in detail as Essity will take the interesting areas and concept in this thesis and turn into future proj-ects. The project will instead focus on finding alternative solutions and calculate how much time each solution saves, or how it will improve the working environment for the operators.
4.4 Literature
To gain a better understanding of the areas covered by the project, a literary study was carried out. Literature was read at the beginning of the project and then con-tinuously supplemented as new areas became relevant. Reinecker and Stray Jörgensen (2014) believe that lit-erature search can be done using three different basic methods. These methods are called chain search, system-atic search and random search. Chain search means that appropriate literature is found and that one text leads to another text. The strength of this method is that it leads from one good reference to the next, the disadvantage is that the reader gets stuck in one point of view. The systematic search is used when searching for informa-tion about a specific subject. The same search words is used on several different databases to find the request-ed information. Random search, is just what it sounds like, information founded is based on a random search where the reader gets inspired along the way. The ran-dom method fits best at the start of a project (Reinecker
