IN
DEGREE PROJECT DESIGN AND PRODUCT REALISATION, SECOND CYCLE, 30 CREDITS
STOCKHOLM SWEDEN 2018,
Operator Feedback
MICHAEL GUSTAFSSON TAPPER
Operator Feedback
Michael Gustafsson Tapper
Master of Science Thesis TRITA-ITM-EX 2018:450 KTH Industrial Engineering and Management
Machine Design SE-100 44 STOCKHOLM
Examensarbete TRITA-ITM-EX 2018:450
Operator Feedback
Michael Gustafsson Tapper
Godkänt
2018-06-11
Examinator
Claes Tisell
Handledare
Conrad Luttropp
Uppdragsgivare
Atlas Copco Industrial Technique AB
Kontaktperson
Stefan Olofsson
Sammanfattning
Det här är en rapport som sammanfattar ett examensarbete av studenten Michael Gustafsson Tapper skriven under våren 2018. Examensarbetet är en del av mastern Integrerad produktutveckling inom spåret Teknisk design på KTH, Kungliga Tekniska Högskolan i Stockholm, Sverige. Dagens montörer i monteringsliner i fabriker får sin feedback från sina verktyg men ibland missas denna information av montörerna. Det här examensarbetet resulterade i en vidareutveckling av ett tidigare projekt in kursen MF2016 Industriell design högre kurs, del 2. Resultatet var en lösning av ett par skyddsglasögon vid namn Protective Signal Provider eller PSP inom produktfamiljen Operator Feedback, OPF. PSP använde ljus från LED- lampor och ljudvågor i form av vibrationer från benöverförande högtalare. En mer utförlig undersökning gjordes också för att para den PSP med Atlas Copcos monteringssystem controller PF6000. PSP använde sig av Bluetooth low energy för kommunikation och kan anslutas till verktyget och till monteringens gränssnitt. En utvärdering jämförde den tidigare externa produkten som kunde fästas på ett par skyddsglasögon med den nya produkten med integrerade komponenter. Utvärderingen resulterade i att den externa produkten skulle bli den bästa lösningen för Atlas Copco att fortsätta med eftersom regleringar, lagar och standarder gör PSP för komplex att producera.
Nyckelord: operatör, återkoppling, skyddsglasögon,
Master of Science Thesis TRITA-ITM-EX 2018:450 Operator Feedback
Michael Gustafsson Tapper
Approved
2018-06-11
Examiner
Claes Tisell
Supervisor
Conrad Luttropp
Commissioner
Atlas Copco Industrial Technique AB
Contact person
Stefan Olofsson
Abstract
This report is a document summarizing a master thesis by the student Michael Gustafsson Tapper written during the spring of 2018. The thesis is the part of the master Integrated Product Design in the field of Industrial Design engineering at KTH, The Royal Institute of Technology in Stockholm, Sweden. Today workers in assembly lines get feedback from their tools, but sometimes this transmitted information is missed by the worker. This thesis resulted in a development of a previous project in the course MF2016 Industrial Design Engineering Advanced Course, Part 2. The result was a solution with a pair of safety glasses by the name Protective Signal Provider or PSP within the product family Operator Feedback, OPF. PSP used light of LEDs and soundwaves in form of vibration from bone conductive speakers. A more extensive investigation was also done to pair PSP to Atlas Copco’s assembly system controller PF6000. The PSP used Bluetooth low energy for communication and can be connected to the tool in use and the interface of the assembly. An evaluation compared the previous external product that could be placed on a pair of safety glasses with the new internal product with integrated components.
The evaluation resulted in the external product being the best solution for Atlas Copco to proceed with since regulations, laws and standards make PSP too complex to produce.
Key words: operator, feedback, safety glasses
Preface
During this semester, I have had the opportunity of working at one of the best technical companies in Sweden Atlas Copco. I would like to thank all my colleges both at Atlas Copco and at school for the feedback and help along the way to the finish line.
At first I would like to thank examiner Claes Tisell and teacher Teo Enlund.
I would also like to thank Annelie Fredriksson, who were my student counsellor at KTH, you deserve great acknowledgment for helping me through my first two years. Without your support I would not have got this far.
At last, I would like to give a special thank you to my supervisors from Atlas Copco Stefan Olofsson, Robin Karlsson and the department Total Work Station as well as my supervisor form KTH Conrad Luttropp who have given me much help and support during these five month.
________________________________
Michael Gustafsson Tapper
Nomenclature
Arduino: Hardware and software manufacture BLE: Bluetooth low energy
CAD: Computer-aided design
DASP: The protocol that Atlas Copco’s tools use with the Power Focus 6000.
External device: Provide feedback from added device on the safety gear.
HK2: Högre Kurs 2 or MF2016 Industrial Design Engineering Advanced Course, Part 2.
Internal device: Safety gear with integrated feedback.
Luminous flux: Measurement of perceived power of light.
Open Protocol: The protocol that Power Focus 6000 use with the external assembly interface.
OPF: Operator Feedback PF: Power Focus 6000
PSP: Protective Signal Provider
Temple: The pins of the glasses that connects to the ears.
ToolsNet: Software for collection of data and process improvement.
ToolsTalk: Software for assembly line management.
Trigger material: Models or modules used to provoke feelings and comments. Used to evaluate concepts.
Table of Content
1. Introduction ... 1
1.1 Background and problem description ... 1
1.2 Purpose ... 2
1.3 Delimitations ... 3
1.4 Method ... 3
2. Frame of Reference ... 5
2.1 State of the art ... 5
2.2 Ergonomic analysis ... 8
2.3 Communication ... 10
2.4 Error proofing ... 11
2.5 Field trip to Scania ... 12
3. Specification of Requirements ... 15
4. Concepts ... 17
5. Glasses ... 25
6. Trigger material ... 27
7. Shape ... 37
7.1 Design ... 37
7.2 Evaluation of the glasses... 40
8. Results ... 43
9. Discussion ... 47
10. Conclusion ... 49
11. Future work ... 51
12. References ... 53
13. Figure References ... 57
List of Appendices
Appendix A. HK2 Executive brief Appendix B. Interview templates Appendix C. Arduino codes Appendix D. Audio tests Appendix E. Vibration tests Appendix F. Ideation phases Appendix G. Survey
Appendix H. Cost analysis
1. Introduction
1.1 Background and problem description
The purpose of the project was to design a pair of safety glasses or some other device with an integrated feedback system. The initial constrain was that the feedback system should include light, sound and vibration. The main goal of the project was to present a functional model and a physical product together with rendered pictures and a cad model.
Figure 1. Storyboard illustrating how the user misses the task of torque tightening
Today the operators in assembly lines gets simple feedback from their handheld tools together with more specified information on a TV monitor. This monitor is sometimes placed on a pillar, a wall or in some other remote area. As the storyboard above in figure 1 shows, this information is sometimes missed because the TV monitor is not in the line of sight, the working environment is too noisy or the user is not alert enough. Assembly lines are mainly designed in two ways;
either the assembly workers are in just one long sliding assembly line, where the workers get their parts delivered to them. This is called a continuously moving line. The other process is when the workers are stationed in one group where they work from different angles on the same object and that is called stop and go line. There are of course also different combinations of the two lines.
The time that the workers have to complete their task is called tact time and that differs a lot depending on which company, what item being assembled and on amount of colleagues involved. Often the workers have earplugs or other hearing protection that isolates them even more from the necessary information.
Figure 2. The final concept from the HK2. An external device on the safety glasses combined with a charging station
The main idea was to include feedback via a wearable device of some kind. This could be safety glasses or for example a bracelet. The background of the project is an already accomplished project for Atlas Copco in the course MF2016 Industrial Design Engineering Advanced Course, Part 2 by the name Operator Feedback that needed further development. (Benjamin, et al., 2017) The result from the initial project in the advanced course, hereby shortened HK2 for Högre Kurs 2 in Swedish, was an external device that was placed on the side of the temples together with a charging concept. The external device with charger from HK2 is showed in the pictures in figure 2. The given feedback from the initial result was signals by light, sound and vibration. These three signals could be controlled by a smartphone app.
Since this thesis was founded on the results of the HK2 project this introduction is complimented with a minor summary from the executive brief from the report. (Benjamin, et al., 2017) The executive brief is attached to this report in appendix A.
Atlas Copco showed interest of the possibilities designing a new pair of safety glasses or another device with the components integrated inside. The main task of this thesis was then be to design the new device, and at the same time, determent if an integrated product is better than the already existing concept. It was suggested too use all the three ways of notification in the development of the new product.
1.2 Purpose
As mentioned in the project description above, the purpose was to design a pair of safety glasses or some other device that had an integrated feedback system. The signals sent from this feedback system should be by light, sound and vibration. One purpose was also to investigate the possibilities of how the device should communicate with Atlas Copco’s assembly system controller, called PF6000. To evaluate if the product should have the components integrated inside a new device or if the external snap-on solution was better, this thesis had to concern these three research questions:
• What could be gained by integrating the components?
• What errors are done by the operators today that can be obstructed by a new solution?
• How should the product communicate with Atlas Copco’s assembly system controller PF6000?
When answering these three research questions along with the increased knowledge, this would hopefully gain Atlas Copco’s product development.
1.3 Delimitations
The three pillars, which support this thesis, are components, ergonomics and signal system.
These areas limited the thesis and helped focusing on narrow it down to one product. These three pillars, the frame of reference, is explained further more in chapter two. Due to the lack of time a functional prototype and a separate physical prototype was done. This to distinguish the outer shape from the functions and focus on a prototype that works in practice. Also due to lack of time and too narrow down the scope, no analysis of solid mechanics and no life cycle analysis was done. Cad drawings and exploded view was also not part of the scope.
1.4 Method
The methods used in this thesis where chosen to apply an agile development (Stickdorn &
Schneider, 2011). This agile development with many phases and iterations should lead to a well- defined and elaborative product.
The structure of the agile development consisted in three ideation phases. The first phase resulted in eleven concepts in total. In the second phase the eleven concepts was reduced down to three and later in the third phase, those three concepts was developed into one final product.
In total, ten persons was interviewed during the initial phase with deep interviews; two industrial designers, three user experience designers and two ergonomic experts. This to gather information about the company and the customer values. Three assembly workers were interviewed before a field trip to the Swedish company of Scania. Furthermore, two persons was also interviewed regarding ToolsNet and ToolsTalk for information gathering.
To evaluate the concepts the usage of QFD matrix and Pugh’s matrix was carried out. A morphological matrix was also done to define the form and shape variation according to customer needs. Then 27 employees at Atlas Copco, of whom 24 were men and three were women, were interviewed with trigger material designed by the feedback from the field trip at Scania. These interviews where short and could be seen more like a survey.
2. Frame of Reference
2.1 State of the art
In the market today, there are many different safety glasses. They come in different sizes, shapes and material depending on the usage. A technology that some of the glasses used is headphones with conductive speakers. This was an interesting solution because the customer can still hear its environment, talk to people while listening to music without disturbing others around them. The magazine everyday hearing says the user can still use the headphones even “If you have a conductive hearing loss”. (Banks, 2018). This was also interesting since many people in Sweden have hearing problem, “Above one million people in Sweden have hearing damage, 57% men and 43% women”. (Bohgard, et al., 2015) If this is equally common in the assembly business, then approximately one of ten workers have trouble hearing, which is quite a lot.
Figure 3. Different smart glasses on the market
Different types of smart glasses in various shapes and sizes on the market is shown in Figure 3.
They are all more or less expensive and this is an obstacle. If the product of this thesis is to succeed, then it needs to have a low price to compete with glasses in the price range of 20 to 200 SEK (Jula, 2018). On the other hand, if the price is in the range as the regular safety glasses made out of just plastic, then the risk is that they will be seen as product of less quality. This means that the price needs to be higher than regular safety glasses so the user does not think that the glasses is something not worth saving and repair if broken. Atlas Copco’s tools are often in use for decades with service and have replaceable parts. From a circular economy point of view, it is good idea to have replaceable parts and providing services. (Benton, et al., 2014) The focus
should rather been on designing something more exclusive with changeable parts than design a device that is made for just one time usage.
Figure 4. Smart glasses for everyday use by Vue
Figure 4 above illustrates the interior design of smart glasses from the company Vue with bone conductive speakers, microphone, Bluetooth, touch screen on the side and inductive charging (Vue, 2016). The technology of these glasses are in particular interesting because they use the same type of signals as specified in the project description. The glasses from Vue included all components for the three signals along with a touchpad.
Figure 5. The interior of google glasses
As the figure 5 shows google glasses can fit a lot of components inside a volume of small size.
Google glasses as well as Vue use components for application that is redundant for the user in assembly lines. However, they show the possibility of using that technology.
Figure 6. Huawei Watch 2
Bracelets on the market with the desirable features are smart watches or fitness trackers. The small size does not seem to be an interference for the required components. As figure 6 of the smart watch Huawei watch 2 illustrates, different smart watches also keep components compressed and they can provide similar signals together with Bluetooth (Huawei, u.d.). Other smart watches on the market are Gear S3 from Samsung (Samsung, 2018) and moto 3602 by Motorola (Lenovo, 2018).
Figure 7. Inductive charging of smart glasses
Charging is divided into two areas. Either it is the common way with pins or it is by inductive charging. The charging of the smart glasses in the figure 7 above is provided by inductive charging. The inductive charging uses electromagnetic induction between two objects to transfer power. This is done with an electromagnetic field. Humavox lists the pros and the cons of wireless charging. (Humavox, 2015) The inductive charging is very easy to use since no connection is needed to transfer the power, the device in need of charging just has to lie on top of the power source. The negative part about using induction is that it is more expensive, has a less effective charging and creates heat. The environment in a factory is also harsh which is not suitable for a charging station that is too advanced and hi-tech. The focus was to have something
that always works since every second lost have negative effects on the economy of the company.
The natural choice was then to go with regular charging with pins.
Two ways to transmit the signals are either to use some sort of Bluetooth or Wi-Fi. The webpage Diffen compares Bluetooth to Wi-Fi. (Diffen, 2018) Linklabs on the other hand lists and compare regular Bluetooth to a more energy efficient BLE. When comparing the different solutions BLE seems to be the best solution of all three because of its low cost and energy efficiency. (Linklabs, 2015)
TED Talks brings up an interesting solution to send vibration signals to the users back. By doing this the user can get signals to different parts of the back depending on the picked up sound. This means that the user can sense the words even if he or she have hearing problems (TEDtalks, 2015). This way of using haptic technology is just an example on the possibilities. IPhones and other smart phones are using haptics so the user can feel the difference between vibration from a phone call and a text message. The haptic technology is becoming more of something that is common today in the human too machine interaction. Haptic technology was therefor something that was possible to use in Operation Feedback.
The feedback from already existing tools from Atlas Copco is limited to light, sound and vibration. Tools on the market that would be suitable to combine with Operator Feedback are STR, STB or STwrench. These tools have different types of signals. STR and STwrench uses all three types of signals and are suitable for using as a reference item in different evaluation methods such as Pug’s matrix or QFD matrix.
Interviews with three assembly workers was carried out as preparation for the field trip to the Swedish Truck company Scania in Södertälje, south of Stockholm. The questions being asked, in appendix B, was not lined up according to the funnel method. Instead, they were more open and loose with the purpose to let the interviewee speak freely about their experience working in assembly line. All the three interviewees had worked one season during the summer of 2016 at Scania but in different departments of the assembly line.
One worked with the logistics at the engine department, one with injection- pump department, and the last one with epicyclic gearings at the department of gearbox assembly. They mentioned that the tact time differed from one and a half minute to one minute 58 seconds and their workday was different from each other’s. The usage of safety gear was different; some used gloves and vest while one did not use any of them, not even hearing protection. However, it was mandatory to use steel toe-cap shoes and working pants. Two of them did not use any tools, but one of them did.
2.2 Ergonomic analysis
During the first ideation, four employees at Atlas Copco was interviewed. Two of the interviewees was working with user experience design and the other two with ergonomics. A lot of crucial information about the user could be gained from this.
Figure 8. Word cloud designed from the interviews in English
These four interviews was held according to the funnel method, the template is attached in appendix B. These interviews was recorded and then transcribed. From these interviews the transcription led to important key words and quotes could be picked. The word cloud above in figure 8 is a collection of words of more importance from the interviews in English. The most important words from the English interviews were light, use, work, red, user and see. The conclusion that could be drawn from this is the importance of light and how it affects the user. The color red is also mentioned quite often since it signalize wrong or error in the tools of today, thus it is convenient to continue using it.
Figure 9. Word cloud designed from the interviews in Swedish
From the Swedish interviews, the word cloud in figure 9 above could be designed. Since the meaning could be lost in translation the writer decided not to translate the transcription before the designing of the word clouds. From the Swedish interviews the most important key words were sound, tools, vibration, light, Hz and work. Several of the interviews led to the discussion about sound and how bad it is for the user to experience a high sound level. The high sound level can also lead to cognitive stress when the user gets signals from several directions. One other possible scenario is that the operator ignores the signals if they are experiencing it too often.
This is why a signal only will be sent if the operator is doing something wrong or if there is a
assemblies which means that the operators can get items in different sizes and shapes on the same line. It is a difference from before when the operator gets the same item on the assembly line all the time. One other conclusion that could be drawn from the interviews is that vibration is an option but then the intensity should not be too high.
2.3 Communication
As earlier mentioned two of the main focus areas are the transmission of information and the communication with a controller. One of the latest controllers by Atlas Copco is the Power Focus 6000, also called PF6000 or just PF. It is a device that works as a interpreter between the interface of the assembly and the interface of the tools. PF gather information about the assembly and stores it for technical documentation. PF can be connected by wire, Bluetooth or Wi-Fi to the tool and uses the DASP protocoll, a system made by Atlas Copco. External devices or systems by third party vendors are connected to the PF by Bluetooth or Wi-Fi, then Open Protocol is used as a communication protocol.
Figure 10. A figure illustrating the network of communication between the tools, the Power Focus 6000 and the external interface in the factory
There are several ways the device could link to the chain of connection. In figure 10, a pair of glasses illustrates the device. One way is to add another link with DASP and extend the chain.
One other way to link is to communicate with the PF and send the same signals that tools get directly to the feedback device. The PF will then work as a hub. The third way to link the device is to skip the whole chain and connect directly to the external system. One possible outcome in the future is that the tools of Atlas Copco communicates directly to the external system without the use of the power focus.
Figure 11. Picture of the assembly in the expo facilities at Atlas Copco
Along with the guided tour, the writer got a small field test and information about the complex systems of communication and technical documentation. One interview in ToolsTalk and ToolsNet was also done. In figure 11, an assembly line from the tour is shown with the PF, tools and other items.
2.4 Error proofing
You could break down the tools of Atlas Copco into five levels of human too machine interaction regarding torque tightening. This human to machine interaction is very important and it was needed to be decided where the device should be implemented.
Figure 12. The five steps of tool tightening of Atlas Copco
Figure 12 illustrates the five steps of torque tightening. The first step is if the screw has been tightened with the correct torque.
The second step is if the correct batch is ok. This means that the tool counts the screws.
The third step is if the joint is ok. By just detecting if the torque is ok, then you do not consider what is called ‘screaming joints’. Screaming joints is when a screw gets stuck or the friction in the surface is too high. The way to solve this is to count the degrees when the screw is at the bottom of the hole.
The forth step is if it is safety critical. Items designed in this level are products that needs extra safety, like for example airbags. The tools that are on this level are STR and STB and they use a transducer with a plus minus five percent margin of error.
The fifth and last step is the zero fault fastening. This is when there should not be any errors at all.
2.5 Field trip to Scania
During the second ideation phase a field trip to Scania was done. The main idea was to gather information about how the environment was in the assembly line, what type of tools and safety gear the workers used as well as how Scania had organized their assembly line. This could then be put in comparison to the field trip to Volvo construction site in Eskilstuna that was done in the HK2 project.
However, some general information needed to be gathered before the field trip. Three persons whom worked in three different assembly lines at Scania was interviewed with the interview template in appendix B. The person who used tools used screwdrivers and nutrunners and got the signals from the tools. The signals was green light from a lamp. The other two persons did not use any tools since they assembled parts, screws, pins or nuts as preparation for the next person on the line. The only signal system they used was a radio to communicate among each other.
One worked in clean room environment and had to use a full body suit since that person did assemblies with injector and pump assembly. Regarding the environment, all three mentioned that their environment was clean and light but the sound level could sometimes be intense. Some tools in the area was pneumatic or hydraulic which creates an occasional high-level sound, therefor many workers used earplugs although it was not mandatory. All three mentioned that the temperature was high, especially since it was summer and they had to use their working clothes. This combined with the heat generated by the movement of tools and people resulted in a very warm environment.
They all mentioned that the consequence of missing signals was that their own stationed stopped but not the whole line. All they had to do was to call for their team leader and they got the help they needed. This means that Scania has a well-developed system for errors. Nevertheless, the whole line stopped at one point because of errors from an ordered part form other suppliers.
After collecting this broad information, enough knowledge was gained to do a visit.
Figure 13. Picture of a parking lot and office from the field trip at Scania
The field trip firstly gave an interesting view of how organized and well prepared the assembly line was for stop and delays. The writer got a guided tour together with Joakim, another master thesis student. At Södertälje, Scania makes axels, gearboxes and motors while other parts are manufactured elsewhere. The frames originally comes from Luleå and are then transformed to Södertälje. When in line it takes four to five hours to create a truck from a frame and they have the capacity to make 60 trucks a day. The trucks are made after demand and takes 8 weeks from order to delivery. If the line comes to a halt it cost 15-20 million SEK for each minute passing.
This mean that it is a high pressure on the operators keeping the tight time schedule.
No pictures was allowed inside the factory at Chassi Porten, therefor the figure 13 above shows the exterior of the buildings. The interior environment was light, clean and everything was well organized. The guide mentioned that they recently changed the interior light to a more white light to imitate bright daylight. The temperature was a bit chilly just in a skirt, but full dressed in working clothes and moving around a lot, then it is probably a good working temperature. The line was almost constantly moving with few stops or errors during the time of the field trip. The tempo or tact time did not seemed to be that intense either.
The pneumatic tools they used was mainly from Atlas Copco, approximately 90 percent of all pneumatic tools. Almost all the electric battery tools was from Desoutter, which is company of the Atlas Copco group as well. The battery driven tools was used as a complementation to the cable tools if they stopped working. Most of the electric tools was from Atlas Copco as was mainly driven via cable connected to an arrangement in the roof. Most of the tools hang from the roof above the head with a leach to bring it down to the operator. Tools that was noted down was Tensor ST, socket selector, twin spin, SAB-bnl along with others. Scania mainly uses PF4000, and in some stations PF2000, in connection with their tools. The safety gear that the operators use are shoes with steel toe-cap, safety glasses and earplugs. The steel toe-cap shoes and the safety glasses was mandatory while the earplugs was used by some, but not all operators.
Figure 14. The two types of the most common safety glasses at Scania
The department at Chassi Porten had two regular types of safety glasses in use that where the most common, but they also had several other types rotated in the assembly. The two most common, see figure 14, was made with a combination of plastic and rubber types and had the same basic shape. The pair in the front had a separable lens that could be removed if broken, scratched or lost. It also had light green rubber on the temples creating friction with the white hard plastic. The pair in the back had a fixed lens screwed in place with the temples. The ones in the back match up with the most common safety glasses on the market while the ones in the front are less common.
Since it is mandatory to use safety glasses at Scania from several years ago they also provide safety glasses with correction for people with impaired vision. The operators had the choice of 5 different safety glasses with impaired vision correction.
One thing that was noticed was that the user did not check the light in their tool. One example was that one operator used a Tensor ST with an extension. The tensor had a green light near the handle while the operator watched the nozzle.
3. Specification of Requirements
From the interviews and the research in the state of the art two lists with requirements and objectives where made. The first list where constructed for a device that where a pair of glasses.
The second list was with the requirement for some other device than glasses.
Table 1. Requirements and objectives for glasses and other devices
Requirements for glasses Requirements for other devices Give feedback when doing an error Give feedback when doing an error
Give feedback when something new is coming Give feedback when something new is coming Not cause scratches on object in assembly Not cause scratches on object in assembly User should not get hurt by it User should not get hurt by it
Sound volume maximum: 85 dB during 8h Sound volume maximum: 85 dB during 8h Illuminace range: 20-500 lx Illuminace range: 20-500 lx
Light minimum angle: 45º Light minimum angle: 45º
Vibration intensity maximum: 2.5 m/s2 during 8h. Vibration intensity maximum: 2.5 m/s2 during 8h Wireless power supply Wireless power supply
Wireless communication with tool Wireless communication with tool Shape not depending on users appearance Shape not depending on users appearance User able to detect light signal User able to detect light signal
User able to detect sound signal User able to detect sound signal User able to detect vibration signal User able to detect vibration signal Protect eyes from shatter, dust, explosion and heat
Prohibit fog
Prohibit scratches on glasses Correct defect of vision
Objectives for glasses Objectives for other device
Protect from chemicals and optical radiation Include signals: sound, vibration and light Include signals: sound, vibration and light Should be easy to use
Should be easy to use
From the previous HK project, the constrains from Atlas Copco’s point of view was that the user should be able to detect sound, light and vibration signal. (Benjamin, et al., 2017) Therefor the same constrains was set in table 1 in this project as well.
From the interviews people said: ”just symbolize that it is not ok. Then it is just red.” By just using red light there should not be any interference if the user has red and green color blindness. Red and green colorblindness is the most common visual impairment when it comes to color worldwide;
therefor it is not possible to design a new product using those colors in combination. (NEI, 2015) This is also a good idea since employees at Atlas Copco mentioned: “we try to minimize green.”
The limit for the sound level was set to 85 dB(A) since that is the absolute maximum sound intensity a company can have during a 8 hour working day according to arbetsmiljöverket in
sound power of 60-70 dB and the “Pain threshold is around 125 dB(A)” (Wallin, 2010) The capital ‘a’
in dB(A) stands for amplified and means that the sound is amplified to be adjusted for human hearing. The frequency of the sound is what determent if humans, dogs, bats or other animals can hear it or not. “For mankind audible sound is that which falls in the 20 – 20000 Hz range. For frequencies lower that 20 Hz, we speak of infrasound, and over 20000 Hz of ultrasound.” (Wallin, 2010) According to Mats Bohgart areas like electronic workshops should have a illuminace of 1500 lx.
An area for precision mechanics however should be 1000 lx but unmanned corridors is as low as 20 lx. (Bohgard, et al., 2015) These numbers are recommended and applies for normal conditions.
Atlas Copco uses the webpage ENAV, among others, to collect necessary data and information from Swedish Standards Institute about standards in workplaces. According to the article 12464-1 by ENAV if the surrounding area is above 750 lx then the illuminace within the direct surrounding area should be 500 lx. (SIS, 2011) Given that the illuminace is between 20 to 1500 lx in a general assembly line, then the device should have a maximum illuminace of 500 lx and be adjustable to a lower number. 20 lx was set as a minimum to match the low light of unmanned corridors.
The light angle was set to at least 45 degrees since “The angle between the direction of sight and the source of light should not exceed 45 º to avoid direct flashing.” The definition of direct flashes is therefor when it is below this limit. Reflexes of light in material can still occur and that was something that needed to be considered. (Bohgard, et al., 2015)
Romain Haettel, who is anergonomic specialist and with a PhD in sound and vibration stationed as an employee at Atlas Copco, mentioned that you are not allowed to work with a vibration greater than 2.5 m/s2 for a 8 hour workday. With this knowledge, the vibration intensity was set to be no more than that.
According to Wallin in the low frequency 2-100 Hz region, the body can for some purpose, and for low amplitudes, be regarded as a particle system. The individual organs and body parts have eigenfrequencies with the eye and intraocular having eigenfrequencies structures of 30-80 Hz and the head of about 25 Hz. (Wallin, 2010, p. 57)
4. Concepts
As mentioned above in the method chapter the structure of the concept evaluation was in three phases. The first ideation phase resulted in eleven concepts, the second in three concepts with glasses and the third phase in the final concept with a pair of safety glasses with the use of fiber optic cable combined with bone conductive speakers.
Figure 15. A concept tree with the eleven concepts in first phase and the three concepts in second phase. The ones in squares was the winning concepts from the evaluations
The first ideation took place during three weeks and resulted in eleven different concepts where sketched down above in figure 15. The concepts of the first ideation were made up with the help of the interviews in ergonomics in appendix B and with the research in the state of the art. All eleven concepts was inspired by already existing safety gear, or other items that could be used by an active person.
Figure 16. Sketches of concepts for head
During the first ideation four different sketches with concepts for head where designed. Those were diadem, glasses, hearing protection and helmet. The four concepts can be seen above in figure 16. Glasses, hearing protection and helmet are all products that more or less already exsists in factories. The diadem on the other hand is not commonly consider to be part of the wardrobe of a worker in assembly line.
Figure 17. Sketches of concepts for hands
Three other concepts for hands were sketched down in figure 17. The first hand concept was bracelet, the other two were ring and gloves. Gloves are used in some factories where finger prints or greasy hands stains are not accaptable like for examples assemblies of gearboxes or sensitive electronics. The bracelet was inspired by fitnesstrackers and smartwatches from the state of the art research while smart rings can be used to lock and unlock your smartphone. Similar solutions could be used for locking your assembly tool.
Figure 18. Sketches of concepts for other parts of the body
The four last concepts in figure 18 where for other parts of the body like ankle monitor, barrette, shoes and vest. Shoes and vests are already something that the workers use and the ankle monitor comes from the punishment method that is sometimes used. The barrette, in conformity with the diadem is something that is not regularly used in an assembly line but can still be considered a possible solution. Regarding the ring and the barrette, the challenge was to include all the components needed for the function to work properly.
Figure 19. The tool STwrench
The first evaluation was done with the help of a Pugh’s matrix with the tool STwrench as a reference product. Several products from Atlas Copco have the option to use signals of light, sound and vibration. STwrench in figure 19 above is one of the tools using these signals and therefor suitable for the evaluation. (inc., 2018)
Table 2. The first evaluation in form of a Pugh’s matrix ranking the concepts against Atlas Copco’s tool STwrench and the customer requirements
Concepts
Custormer requirements STwrench Diadem Glasses Hearing protection Helmet Bracelet Ring Gloves Ankle monitor Barrette Shoes Vest Give feedback when doing an error = = = = = = = = = = = Give feedback when something new is coming + + + + + + + + + + + Not cause scratches on object in assembly = = = = - - - = = = = User should not get hurt by it = = = = = = = = = = = Sound volume maximum: 85 dB during 8h - + + + - - - - - - -
Illuminace range: 20-500 lx - = - = = - = - - = =
Light minimum angle: 45° - + - + - - - - - + -
Vibration intensity maximum: 2,5 m/s2 during 8h = = = = = = = = = = =
Wireless power supply = = = = = = = = = = =
Wireless communication with tool + + + + + + + + + + + Shape not depending on users appearance - = = = = = = = - = = User able to detect light signal - = - + - - = - - - - User able to detect sound signal = + + + = - - - - - = User able to detect vibration signal = + + + + - + + - = =
Plus 2 5 5 7 3 2 3 3 2 3 2
Minus -5 0 -3 0 -4 -7 -3 -5 -7 -3 -3
Equal 7 8 6 7 7 5 7 6 5 8 9
Sum -3 5 2 7 -1 -5 0 -2 -5 0 -1
Ranking 7 2 3 1 5 8 4 6 8 4 5
The results from the evaluation is illustrated above in table 2. From the evaluation the helmet was a clear winner with the glasses and earmuffs on second respectively third place. The helmet is not used that often in assembly lines but this could be a solution for companies in the business of mining tools. Helmets are also often in combination with glasses, which make the integrated solution in helmet redundant. Glasses was the most technically possible device, since it could include light, sound as well as vibration in form of conduction. Shoes might be uncomfortable when having vibration and it is not possible to use conduction. Light is not in the line of sight and sound would be transmitted by air. This all means that it is not technically possible. Hearing protection is not uncommon but often uses earplugs instead. A light signal is not possible. Its
purpose is to cancel sound in a noisy environment, which means adding sound is maybe not a solution.
From the interviews in the ergonomic analysis one person said: “It is very important that one does not scratch a car or something. No rings or watches.” Since bracelet as well as rings are not allowed in many factories, in particular car factories because it can scratch the surface of the items built in the assembly line. The same rules applies with gloves as with bracelet and rings. The gloves can stick with the items in movement which can lead to accidents, therefor it is not technically possible to apply these solutions for usage in assembly lines. It is instead possible in other businesses.
The ankle monitor is still a technically possible solution since it is not in the way of the usage and not in risk of scratching the surface of the items produced in the assembly. However the ankle monitor is typically used as punishment which could lead to the user feeling surveyed by his or hers own company. Naturally, this could lead to trust issues and interference with the union.
Diadem and barrette are solutions based on the user having enough hair to fit a diadem or barrette on his or hers head. Since they are not suitable for all users then they were given low ranking in the Pug’s matrix. A vest is very likely to be used in a factory but a low ranking were given because the only signal that is possible is a vibration signal. This might not be enough for the user to detect when he or she is moving around. At the same time, it maybe too heavy compared to a regular light vest.
From the first evaluation many concepts had been excluded and the main focus was now to look more into safety glasses. The study from the state of the art gave the result that there are many different safety glasses on the market and comes in many different forms and shapes.
When the decision of designing a pair of safety glasses was made, then the function needed to be defined so the right components and shape could be set. Therefor a morphological matrix was created. The lenses could have been two or just one, the temples could have been wide or thin and the choice still remained between the options light, sound or vibration.
From the interviews, regular speakers and in-ear headphones were excluded as a technical solution. However, sound in the form of vibration to the bone, called bone conduction, was still a possible solution. From the interviews in ergonomics, one person mentioned: “we usually say that sound signals can be hard to distinguish in that kind of environment. Besides that it can create a lot of stress.” As well as regular speakers were excluded, any form of vibration that were not in the form of bone conduction was also rejected. Vibration to the eye section of your head can feel uncomfortable but one other person from the interviews said: “Spontaneously I think the vibration is an interesting area.
When we talk about vibration in the purpose of information translation. And the reason is that you really get instant feedback. You don’t have to change focus.” Since vibration is still a powerful source when it comes to information transfer, especially in combination with haptic technology, it should not been excluded in this ideation phase.
When it comes to the light source they were several suggestions. One interviewee said: “another idea for light is also you can project light.” This solution together with a regular LED and a multicolor RGB LED where the three ways to illuminate the light. The multicolor LED is slightly more expensive but was still the best solution since it was then optional for the customer which color they preferred to use.
Figure 20. Evaluation of the morphological matrix
The size, shape and number of lenses for the glasses could differ a lot. If the demand for correction of defect vision is high then the multiple lenses were to prefer. This, because the defection of vision could differ from right to left eye. As a conclusion from the interviews, state of the art and the literature study the morphological matrix was modified to the matrix in figure 20.
Table 3. QFD matrix with customer requirements, ranking, product properties and a sum of the score
Rank Product Properties
Customer Requirements Atlas Copco Worker at continuously moving line Worker at stop and go line Maximum sound volume Illuminace Light minimum angle Maximum vibration Sell price Wireless communication with tool Battery driven Battery length Adjustable settings Several lenses Transparent material Soft material Hard material Durable material Flexible material Light material Give feedback when doing an error 5 5 5 1 9 1 9
Give feedback when something new is
coming 5 5 5 1 9 1 9
Protect eyes from shatter, dust, explosion
and heat 5 5 5 3 9 9
Prohibit fog 1 5 3 1
Prohibit scratches on glasses 1 4 5 9 9 9 3
Not cause scratches on object in assembly 5 5 5 9
Able to detect signals 5 5 5 1 9 9 9
Not too expensive to produce 5 1 1 9
Prohibit throw away too easy 5 1 1 3 3 9 3
Long battery life length 3 5 5 9 9
User should not get hurt by it 5 5 5 9 9 9 9 9 9 3
Should not be uncomfortable 5 3 5 9 9 9 9 3 3 9 3 9 3 9
No cables 5 5 5 9 9 3
It should have a wireless power supply 3 5 5 9 3
Protect from chemicals and optical
radiation. 1 5 5 3 3 3
Correct defect of vision. 1 3 3 9 9 9
Should not desturb work 1 5 5 9 9 9 1 9 3 9 3 3 1 9
Goal values with units 85 dB 20-500 lx 45° 2 2,5 m/s 200-2000 SEK Yes/No Yes/No 8h Yes/No 2 Yes/No 5,1 m/s 5,1 m/s 5,1 m/s 5,1 m/s 60 g
Objective targets Speaker LED LED Vibration motor Price Bluetooth Battery Battery Modular Two lenses Transparent lenses Rubber timpels ABS Plastic ABS Plastic ABS Plastic ABS Plastic Score Atlas Copco 114 234 64 226 45 69 90 75 162 63 46 136 54 84 96 84 Score continuously moving line 132 252 82 212 9 99 114 123 162 72 86 122 81 141 105 90 Score stop and go line 150 270 100 230 9 105 120 123 180 81 90 140 90 150 90 108
Since the decision was made to make a pair of safety glasses a QFD matrix in table 3 above was done. The customer values where based on the same values in table 1 and 2, but was also in some cases divided, complimented and arranged so the product properties could match the European standards regarding safety glasses. Therefor it may look like the customer values differ from table 1 and 2 compared to table 3, but they are rather more detailed.
According to the European standard of EN 168 the glasses should “withstand the impact of a 22 mm nominal diameter steel ball, of 43 g minimum mass, striking the ocular at a speed of approximately 5,1 m/s, when tested in accordance with 3.1 of EN 168:2001.” (SIS, 2001) This gave some actual numbers and facts to the QFD matrix. Therefor the material is in the units of m/s and not MPa used for yield strength. One of the pair of regular safety glasses was weighted in the lab at Atlas Copco to the mass of 28 g. The mass of operator feedback was then doubled and set to a maximum of 60 g to include the electronic components.
The interviews, the state of the art and research reflected the product properties in the matrix.
The ranking differed from the workers at continuously moving line and the workers at stop and go line. This was because the workers on the stop and go line moved a lot more than the workers that stood still next to a moving line. The workers also had another ranking than Atlas Copco.
The vibration and the light seemed to be the best solutions according to the matrix as well as the interviews. Regarding the choice of material, it did not matter as long as it was a light and durable material that follows the terms of European standard of EN 168. A lens with transparent glass that was divided for each eye was the preference for the users.
5. Glasses
The second ideation brought three glasses with different solutions. The first concept was with a fiber optic cable, the second concept was with a sandblasted surface and the third and last concept was with a reflective tape. The vibration motor was placed behind the ear in contact with the skull to generate sound via bone conduction. The design of the glasses was scaled and simple with one yellow vertical line on each side of the temples. Two lenses was chosen from the morphological matrix because it is then possible to have lenses with different reflective errors in the glasses.
Figure 21. The fiber optic concept
The first concept was inspired from the final concept in the HK2 project. The concept used a fiber optic cable in the lower part of the lens that was in connection with a LED. When the LED shine the light was transported through the material and shines in the other end of the optic cable where the fiber is cut crossways. In figure 21, the optic cable was placed in the lower part of the line of sight since it is where the worker is looking. The worker often looks down on their gear, tools or assembly item. During the field visit at Scania, every operation was below shoulder level since Scania is focusing on making the working conditions more ergonomic. One way to catch up transferred light through glass is by breaking the fibers. The second concept used this method by cover a part of the screen with a frosted glass surface. This means that it is very flexible since it could be placed anywhere on the lenses.
Figure 22. The frosted glass concept
The LED was placed directly inside the glasses at the edge of the lenses to transport the light through the lens. As figure 22 shows, the area where the glass is made rough is where the light spreads out. This concept works with larger areas but can of course be applied in different shapes and sizes. The color and shape follows the same design as the first concept with a dark gray color and a vertical yellow line.
The third concept used a LED as a projector against the lens. The lens was then complimented with a reflective tape to notify the user of the signal.
Figure 23. The tape concept
As figure 23 above showes, the tape is placed in the lower part of the field of vision of the left eye. The glasses follows the same design as the first and second concept with a dark gray color and a yellow line on both sides of the timples.
6. Trigger material
With the results from the first evaluation and the supplement of the QFD matrix, all the concepts in phase two was now limited to the usage of light and bone conductivity as a form of signal transfer to the user. However, both the technology for the light and the sound needed to be tested before the final product. Two functional models for the bone conductivity and three functional models for each light concept was used as trigger material. These three light concepts along with the two sound concepts was tested in the facilities of Atlas Copco after the information gathering from the field trip to Scania.
Figure 24. The small DC motor for bone conductive testing
From a video on youtube the idea of designing a custom made bone conductive headphones was born. (JamLabs, 2015) The regular headphones that you can buy in store receive pulses of current in the same beat as the music. The speakers of the headphones could then be removed and replaced with a small DC motor as in figure 24. When connected with the wires of the headphones, the DC motor is then responding and vibrating with the same beat as the headphones. This vibration coms from the AC, alternating current when it is rotating. This rotation was not visuall by naked eye but could be felt. The DC motor was soldered with the wires and then tested. The first tests was done next to the timpels, skull, cheek and ears without results. From the video on youtube it was recommended to bite the motor. With the use of earplugs in a quiet office the music could then be heard inside the head when biting the motor.
Figure 25. The small motor from a holiday card that plays music
Figure 25 shows the small motor from a music playing holiday card. The stripped holiday card includes a small motor playing the music when the card is opened. This type of motor uses piezoelectricity to produce music and can be used in the same way as the DC motor in the previous tests. However the metal in the motors core did not react to the soldering as desired, therefor no tests where possible with the piezo motor. Enough knowledge was gained to order parts online from the company Digikey. The parts was ordered to assemble a pair of glasses with bone conductivity combined with a light solution.
Table 4. Part list of components for own conductive speakers
Item Number Company Price (SEK) Amount
Bone conductor 1528-1948-ND Digikey 76,9 2
Audio amplifier 1528-1381-ND Digikey 85,5 2
3,5 mm Stereo jack SC1459-ND Digikey 10,9 2
Adafruit Feather 32u4 Bluefruit LE 41014094 Elektrokit 389,0 2
Battery LiPo 3.7V 400mAh 41016062 Elektrokit 99,0 2
Total 1322,6
As table 4 above shows the total price of 1322,6 SEK for the ordered components without the price of the shipping. By the time of the calculation one USD had the value of 8,59 SEK. These components was for making two models for testing the best placement of the speaker, behind the ear or in front of it next to the temples. When the bone conductor and amplifier was soldered together as adafruit recommend, the componets was combined with the Adafruit feather.
(Adafruit, 2015) The Adafruit feather was used in the same way as in the previus HK2 project.
Figure 26. The circuit diagram with a bone conductive speaker, audio amplifier, stereo jack, Adafruit Feather 32u4 Bluefruit LE, 3,7V 400mAh battery and a multicolor RGBW LED. The RGBW LED is illustrated
here with fritzing as a RGB LED. The missing white pin should be connected to pin 11
With the standard code ‘BluefruitLE_nrf51822’ in appendix C from adafruits webpage (Ada, 2017) the different pins could be controlled with a smartphone. The pins were connected according to the circuit diagram in the figure 26 above and could be turned on and off since the Adafruit feather had a BLE module. The RGBW LED, which is a standard component in the sortiment of Atlas Copco, could be controlled so the red, green and blue pin could each be enabled separately or in combination to create different colors. The amplifier was connected to one of the pins so it also could be on or off together with the light.
Figure 27. The temple model and the neck model. The two functional models for testing placement of the bone conductive speaker
When the componets worked well with the code and the Bluefruit app it was time to combine them all with a cover. Two types of cover was created in Solid Edge ST9 and then 3D-printed with a Ultimaker 3 Extended. The figure 27 above shows the two different covers, the temple model and the neck model. And after testing them it turned out that the best placement was behind the ear because of access to the skull. Full scale tests was done with audio in appendix D and vibration in appendix E resulting in neck model providing the highest audio recording of 50,2 dB when in a free placement, and the tempel model giving the highest vibration intesity of approximently 0,7 m/s2 in Z-direction when having the frequence of ~2,3 kHz. The temple model gave a maximum sound level of 49,5 dB(A) when placed against a table and the neck model gave a maximum vibration of approximently 0,5 m/s2 at a frequence of ~1,3 kHz in Z- direction. In the tests, the Z-direction represented the direction against the skull surface.
When the audio test was carried out, it was noticed that the cover for the motor was working as a sound box. This had to be prevented if the feature of the product is bone conduction.
Figure 28. Test equipment for recording and measuring sound at Atlas Copco
Together with Romain Haettel the sound level could be recorded in the quiet room in figure 28 and the vibration could be measured. The app Signal Generator from App Store was used to generate audio test tones in frequences between 80 hz to 20kHz for the test equipment. Three tests was done on four different surfaces for each model resulting in 24 tests in total, read appendix D for more detailed information. The surfaces tested against was a table, a head, hanging free and a coffie mug. The tests of vibration was done with an accelerometer in 3 directions; X, Y and Z, with Z direction being directly agains the motor.
When the audio and vibration test was done two samples was recorded for given feedback. One sample for the feedback when a new item is delivered on the assembly line and one for the feedback when the worker have given the wrong torque. The samples was repeated in English, Spanish and last Swedish.
Figure 28. The Arduino set up station for LED testing
The first tests with the light equipment was basic testing with different types of LEDs. In figure 28, the whole set up where made up by one Arduino Uno from the Swedish company Kjell &
Co, a nine volt battery and a pin board to try out which LED that was the best. The code for the Arduino can be seen in appendix C.
Figure 29. Different shapes of light spreading material and the RGBW LED
The first testing was done with a RGB LED that you can find online and in several different local stores, it turned out to be compatible with the Arduino. One other test was done with an RGBW LED from Atlas Copco. This LED in figure 29 are used in the tools QMC21 and QMC41. To use the same components as already existing products is both time and cost effective so the decision was made to use the standard component. It was also noticed that the RGBW LED did shine brighter and had more power when transporting the light through the white plastic light reflector.
The RGBW LED in the left picture of figure 29 are shown together with other standard parts in the tools of Atlas Copco. Some of the tools uses a special fiber optic plastic for transferring the light through the transparent material and then spreading it on the edges. The light is first absorbed by the white round plastic part that is snapped on the LED circuit. The transparent material is then snapped on to the white part. By using the snap on method, the tools can then use different shapes of transparent material. The different fiber optic parts are illustrated in the same figure. The right picture shows different LEDs found in Atlas Copco’s workshop. A larger plastic part was 3D-printed to fit the other LEDs for more testing. Unfortunately the material was too dark and densely packed to spread the light in the correct way. The other LEDs was also not bright enough and had another light spreading angle.
Testing was done to verify the three concepts. The first thing to be done was to cut out a hole in the glasses that matches the same shape as the transparent part from the tools today. By doing this, it was possible to fit the white plastic reflector just by a snap fit.
Figure 30. Cut hole for press fitting the RGBW together with the modified light reflector
