PROJECT -1
Collaborative Robot (CoBot) for Garment Manufacturing
The innovation process usually involves identifying problems. Problems are there to be solved and if the problem and solution are larger than your own workspace, then there is room for innovative solutions. If the innovative solution works, it could be a commercial plan, but the whole process to get there is a long way of investing; Investment in time, resources – both financial and material, and patience and persistence. The road to an innovative solution is often long and unknown, but the results will always be satisfying and profitable in every way possible.
Table of Contents
Project 1 – Collaborative robot for garment manufacturing
In this article, we take a closer look at the innovation project of the collaborative robot for garment production. What is the problem we want to solve? How are we going to develop the solution and what are our expectations of the entire process? We will answer those questions to give the reader a more informed view of what we are trying to develop and whether that is realistically feasible. We also discuss the timeline along which this should be done. The aim is to provide a total product that contributes to a better future for the apparel manufacturing industry and its employees. This is part of a series of articles for the participation program and hopefully gives the reader more background information, this is the second article in a series of 3 articles.
Author: YS Koen, Klaten, 14 April 2021
1978 – Keep this year in mind as you continue to read this article. It is now 2021, 43 years have passed. How would you like it if the world had stood still all these years? For 43 years. Follow me in these thoughts, imagine you are in the office, your manager pressures you and says you are so slow on your computer that he/she would do it even faster on a typewriter. Then you can complain that you do have a computer at your disposal, but that the technology under the hood dates back to 43 years ago. The manager does not want to hear it, stop complaining, fulminate this one, just do your job. You know what, says the manager, if you have not finished the work at the end of the day, you just keep working until it is done, and oh yes, you do not get paid for overtime because you are slow yourself.
You will say this is impossible, no one would accept that. No, this should not happen either. But take a look at the pants, the skirt or the dress you are wearing, at the shirt, the blouse or the top you are wearing. 95% guaranteed made under such conditions. Have you ever seen a YouTube video of a sewing factory where the production workers try to hit production numbers like zombies? It is not a pretty picture.
Another example, designer jeans can easily have a retail price of $ 250.00 or more. You would say real quality, you have to pay for it. Only a spinning mill in Japan can make that denim. That spinning mill gets its cotton from Zimbabwe because the Zimbabwean cotton is again of the best quality to make that denim. The distance from plantation to spinning mill is only +/- 12,000 KM. Then the cotton must be spun, woven, and dyed. Then two things can happen: the jeans are made in Japan and shipped worldwide, or the denim is shipped, and the jeans are made on site, close to the consumer. For the sake of convenience, let us assume that the store where those jeans are sold is in New York (+/- 10,000KM) or Amsterdam (+/- 9,000KM). Then you wear the designer jeans for the first time, which has already covered a distance of +/- 21,000 KM (+/- 13,000 Miles). Is that normal? I do not think so, those $ 250 jeans do more damage to the environment than my grandmother did in her entire life, by the way, she just turned 100 in August 2020.
You could say that both examples are exaggerated, but then I have to disappoint you because this is the reality we live in now. This article is not about who is guilty of the oppression of textile workers, but the fact remains. This is also not a judgment as to whether you should let the quality of clothing be determined by the conditions in which they are made. The consumer deserves what he pays for, so if you pay for quality, you deserve quality, but the environment should not suffer from our desires, because sooner or later everyone will fall victim to it. Okay, for that last problem everyone in the textile industry is working hard to come up with solutions. People are starting to realize that it is absurd for an average piece of clothing to travel at least 12,000 kilometers before it ends up in your or my wardrobe. So, let us get a little more aware of reality. 7.85 billion world population let us use this planet a little more efficiently. I am not an environmental freak, but even I see its absurdity and realize it should not still happen by 2021.
Nice of course, but how do we get to these topics when the story is about an innovation project for sewing robots? 1978 was the last year that a groundbreaking innovation was introduced in the garment manufacturing sector, I am not talking about the total textile industry, but the assembly of garments. While the textile industry has always been at the forefront of innovation since its long history, it has been deadly silent for the past 43 years. Now people will think it will not be as bad, well, just as bad as if your beloved jeans traveled 20,000 km before you could put those pants on for the first time.
Of course, there have been innovations, further development of existing techniques. But absolutely no groundbreaking innovations. For example, seamstresses no longer have to turn on pockets or hems themselves, there are now machines that can do it all for you. But is it really such a plus that the work is carried out in detail? Maybe it is a piece of technology that you can be proud of, but is it the solution everyone is waiting for? If the numbers look good in spreadsheets, does that mean it will work in practice? How does the employee experience this, is their work so much better now? Let us look at the dry reality. The purchase of such a machine that can-do detailed work, between you and me, for different model pockets, another machine must be installed. Cost price, I am not even talking about the topline machine, just a midline industrial pocket sewing machine, +/- € 12,000.00. (Photo 1.) What can that machine do? Let us not get crazy, but roughly between 60 and 100 pockets an hour. This requires one operator to control this machine, because yes, the process is carried out autonomously, but the machine will have to be “loaded”. Still, let us not get too crazy, just assume that an average production worker has to make 20 to 30 pants in an hour. That means that I have to have one industrial pocket sewing machine for a maximum of four production employees and that also costs me a production worker. How many trousers can a production worker do in one hour if he has to turn on the pockets himself, 15 to 20 at least. So, a groundbreaking innovation? You name it because I am crying out in the desert. Wait a minute writer, I hear you thinking, I watch YouTube and I see robotic arms with sewing machines (photo 2.). That is right.
But if I connect my grandmother to a robotic arm and give her a sponge, wiper and chamois, it does not mean I have a robotic window cleaner, I have a robotic arm, and I traumatize my grandmother. But it illustrates how there is so little creativity when it comes to collaborative robots. The car industry has begun to let robots collaborate with humans. A model of a robotic arm was used there, we are talking about 1962, that same model is still in use today. But does that mean that all industrial robotic arms must be the same? People say you should not change things if it works. My first Nokia phone also worked and check out what Steve Jobs has accomplished over the years and rethink if you should not change things when they work.
This is, of course, a totally absurd introduction to a serious story. Now let us get to the more serious side of it. Because that there is a story in it, that should be clear. The idea began to take shape in 2017.
Perceiving a problem
During a consultancy assignment, we (P.T. Emas Cemerlang Bersama) were given the opportunity to take over a clothing production company. The owner’s involvement in his own business had fallen to an alarmingly low level, so he was looking for a solution. The production facility’s capacity was that of a small, medium-sized player in a highly competitive market, both locally and internationally. The production facility is too large to only produce clothing for its own account, but too small to function as an independent producer for (international) customers. Nevertheless, we thought it was an interesting company where we could build a bright future with a clear vision and a stable foundation.
We started by assembling a management team consisting of people with extensive experience in this sector. The first goal was to make this production company profitable again. Because if you fail to make a small manufacturing company profitable, how can you expect a large investment to pay for itself? Together with the management team and the staff present (300+ employees) we managed to implement changes and return to operational profitability within the first year. We then submitted an investment proposal to the network of Golden Intellect Corporation Limited (HK) and American Capital Investment Limited (HK), which was well received. Within two weeks, contracts were signed for a direct investment of US $ 5 million and an additional credit facility of US $ 3.5 million. After we were assured of the required investment, we were able to get started with preparatory work. However, the deeper we went into the matter, more and more questions arose that we could not answer.
Major changes are imminent in the textile industry that will radically change the current production model. One of the biggest changes is that centralized mass production facilities will soon become a thing of the past. They want to bring the production of clothing closer to the consumer. As we described earlier in this article, clothing that has to travel tens of thousands of kilometers before it reaches the consumer is outdated. For a long time, low-wage labor has benefited, especially in Asia. But in Asia salaries have been under pressure for some time. Subsidizing the sector by governments of production countries to remain competitive is a dead end. The only reason why production has not moved yet has everything to do with the presence of sufficient employees. While there is still a growing labor market in Asia, the West is struggling with a saturated labor market, jobs in the clothing manufacturing sector are not popular there. The transition is about to happen, but there is a missing link. That is where our problem lies. During the research for the major investment, we wanted to gain insight into the new technological developments that were going on in the field of clothing production. If you are going to invest, you are doing this for the future, then every expenditure must also be focused on the future. But when we thoroughly analyzed the condition of the machines, we came to the conclusion that little, or no attention is paid to the changing era of Industry 4.0.
The biggest cost item in such investments is machinery, the tools how that clothing is made. The most valuable thing in a garment production company is the people, but if we do not give people the tools to efficiently perform their tasks, how can we expect to be competitive. We have spent months researching what is being done to develop new technology within this sector. We can build the entire company and cover it with the latest technologies in all kinds of areas, from Smart & Green building to super smart distribution systems and from top-of-the-line automation to digital pattern processing and so on. But when it comes to the main tasks of the company, assembling clothing, there is nothing to be found that meets our expectations.
The question we had asked ourselves was: how can we make our production employees perform the work more efficiently, under reduced time pressure and with fewer errors and interruptions? You can have the best machines available, but that does not automatically mean that you have a more efficient execution. Let us put it right, the machine builders are not to blame, they continue to implement innovations, we are regularly amazed at what they have come up with. The production sector itself is guilty of the situation that has arisen. The entire textile industry has benefited from low wages over the last 40-50 years, which has prevented real innovation. But where the textile industry has always been one of the frontrunners with innovation, it has been painfully silent for the last 43 years for clothing production.
We have the answers ready for the individual parts of the total garment assembly process. You cannot think of a problem where there is not a machine available to solve it. However, there is no answer to this: how can we make everything work together so efficiently that the contractors (the production employees) also benefit from it. Because thinking that if there is a bottleneck somewhere in a part of the entire process, just to isolate that part and develop a machine that solves the problem, is not tackling the problem. What has happened is that we have started segmenting production instead of integrating it. More departments have been created where we can all house the micro parts. Which actually resulted in 2 macro departments.
The first department is basic production. Most of the work is done here, as the base product is processed into one garment here. But because ‘bottlenecks’ have been removed from this process, the work has become simplistic, which has created a great deal of mindlessness in the performance of the work. The only thing that matters in this part is the numbers that can be produced, preferably with as few employees as possible. In numbers, this means: 70% of the work is done by 20% of the employees. Within the second department, for the sake of convenience we call it the finishing department, there it is very busy with the number of employees and actions they have to take to finish a garment. There is a machine for every part. But every machine must also be manned. One machine in this process can handle the amount of work that produces one to two machines from the first department. And we encounter a problem there. Whereas machines are designed to remove bottlenecks in the assembly process, those same machines now create an even greater bottleneck. These machines are so expensive that there is no equivalent investment in the amount of machines to keep up with the production numbers of the first part.
So, where we started asking ourselves questions and trying to find answers, more questions came up. There was only one thing left and that was to have all investment agreements dissolved and find appropriate answers. First, let us give a brief history of the textile industry and innovation.
History of sewing and fashion production industry
17.500 BC – First Sewing Needles with Eyes;
Archaeologists and anthropologists have discovered sewing needles with eyes dating back to 17,500 BC, which were likely made of bone, and used to sew skins and furs.
202 BC-220 AD – Han Dynasty Uses Sewing Needles and Thimbles;
Although some ancient sewing needles date back nearly 25,000 years ago, during the last ice age, Chinese archaeologists found something interesting that dates back to 202 BC. Sewing needles, along with the oldest thimbles in recorded history, were found in the tomb of a government official from the Han Dynasty. Even in ancient history, sewing was an important part of life– and more advanced than we might think.
1200 – Buttons Become Popular in Europe;
Due to the Crusades, Europeans encountered many other cultures. As a result, Europeans began using buttons and button holes to fasten their clothing. Soon after, buttons became a driving force in the clothing industry in Europe.
1730 – Cotton Spun by Machinery;
In 1730, cotton thread was spun by machinery in England. This introduced cotton thread to a much broader audience, and it spread “like wildfire” across the British Colonies and the world at large.
1730 – Needle Factory Founded in Aachen, Germany;
Stephan Beissel founded an early needle factory in Aachen Germany, a town then famous for its needle-makers guild.
1755 – Charles Weisenthal takes out patent for mechanical sewing needle;
Charles Weisenthal, a German immigrant living in London, took out a patent for a needle meant for mechanical sewing in 1755. No record of any machine to accompany the needle has ever been found, but this is recognized as one of the first events that would culminate in the sewing machine.
1790 – Thomas Saint patents early sewing machine;
Though his machine was never built, a London cabinetmaker successfully patented a crude sewing machine in 1790. Thomas Saint also built plans for his machine, which were not discovered until the 1800s. It would not work without modification, but it was an important step on the road to the sewing machine.
1812 – Brass Buttons Mass Produced in America;
In 1812, The Waterbury Button Company began mass producing brass buttons for American military uniforms. 1812 marked the start of a renewed war between England and the United States, and brass buttons became popular for American military uniforms, as well as popular for domestic and import use among civilians. Brass buttons were a strong commodity throughout the 1800s and are still used in sewing to this day.
1812 – James and Patrick Clark Mass Produce Cotton Thread for Sewing;
In 1812, England was being blockaded by Napoleon’s warships, which led to a silk shortage. As a nation used to luxury goods, they soon found themselves in quite a bind. Their luxury needs weren’t being met because of the embargo, and they lacked sewing thread almost entirely.
Enter James and Patrick Clark of Clark and Co, based out of Paisley, Scotland. The Clark family came up with a way to twist cotton threads together, producing an excellent thread for sewing. Their thread was the first such material mass produced for sewing, in fact. England, and the rest of the world, appreciated their efforts.
1830 – Barthelemy Thimonnier invents first practical sewing machine;
In 1830, Barthelemy Thimonnier was awarded a patent by the French government for his sewing machine. Though he initially imagined an embroidery machine, he found its true purpose as a sewing machine. It was practical and efficient, used a barbed needle, and was built almost entirely out of wood. At some point, he had a factory running with 80 machines. He also sold his sewing machines for commercial use and ran one of the first-ever garment factories, throughout his tumultuous life.
1831 – French Tailors Riot Over Barthelemy Thimonnier’s Sewing Machines;
Though his patented sewing machine changed the world and he ran what was likely the world’s first sewing factory, Barthelemy Thimonnier had his fair share of detractors. Many French tailors were afraid Thimonnier’s invention and his factories spelled the doom of their livelihood.
On January 20, 1831, some 200 hungry and fearful French tailors set fire to Thimonnier’s factory and his mostly-wooden sewing machines. Reportedly, Thimonnier never fully recovered from the after-effects of the riot.
1834 – Walter Hunt Invents First Lockstitch Sewing Machine;
In 1834 (or 1833, according to some accounts), prolific American inventor Walter Hunt created the first lockstitch sewing machine. Hunt’s sewing machine is said to be the first that didn’t mimic the movements of the human hand– instead, it was a curved, eye-pointed needle machine that passed the thread through fabric in an arc motion. On the other side of the fabric, a loop was made while a second thread, carried by a shuttle, ran on a track and passed back through the loop. The lockstitch was born.
Though his design was brilliant, Hunt did not patent his machine. He worried it would put seamstresses out of business and cause unemployment. That was a real concern in the sewing industry (see 1831 above), so his fears were not unfounded.
Hunt also invented the safety pin, a predecessor to the Winchester repeating rifle, road sweeping machinery, nail making machinery, and a safer household oil lamp, among other inventions.
1844 – Elias Howe invents modern sewing machine;
Though many other brilliant inventors produced mechanical sewing equipment, most Americans claim that Elias Howe, from Massachusetts, invented the first modern sewing machine in 1844. Unfortunately, the machine didn’t catch on immediately, even though it amazed people in sewing competitions, where it out-sewed some of American’s finest tailors.
Howe left America for England to sell his design, which did not end well. When he returned to America, he found several entrepreneurs, including Isaac Singer (yes, that Singer), making money from his patent. Though Howe eventually made his fortune, it was a hard-won battle.
“No useful machine ever was invented by one man; and all first attempts to do work by machinery, previously done by hand, have been failures. It is only after several able inventors have failed in attempt, that someone with the mental power to combine the efforts of others with his own, at last produces a machine that is practicable. Sewing machines are no exception to this.” – James Edward Allen Gibbs, inventor and sewing machine innovator.
1851 – First Singer sewing machine patented;
On August 12th, 1851, Isaac Merritt Singer patented what’s known as the first modern and practical sewing machine. Though patent and legal disputes abounded around this time, Singer was eventually able to formalize an affordable payment plan for his machines, bringing them into many American households.
1874 – Husqvarna Begins Sewing Machine Production;
Though they’re known for their excellent Viking sewing machines, Husqvarna was not always in the sewing business. Prior to 1872, Husqvarna produced high-quality rifles and other military armaments for the Swedish Crown. In 1872, orders for military hardware stopped coming in, so Husqvarna had to adapt.
In 1874, they turned their formidable steel forging and metal bending skills toward making many household items, including pots, pans, and bicycles. Soon enough, however, their sewing machines took the spotlight.
1880 – First electric sewing machine;
1880 marked the first time in recorded history an electric motor was affixed to a sewing machine. These motors were added to retrofitted sewing machines, invented and implemented by Philip Diehl, a contractor who worked for Singer.
1889 – First practical electric sewing machines sold for mass market;
In 1889, Singer introduced a sewing machine that could be purchased with an already built-in electric motor. This was the first electric sewing machine intended for home use.
1900 – Heinrich Stoll creates the flat bed purl knitting machine;
In 1900, hosiery was in great demand. Germany’s Heinrich Stoll met that demand with his flat purl-knitting machine. Stoll’s machine was both efficient and powerful, resulting in more than enough knitwear to go around.
1910 – Circular bed purl knitting machine invented;
By 1910, knitting machines had come a long way. Many were automatic and motorized. They still couldn’t produce the foot and length of a stocking in one single production step, but the invention of the circular bed purl knitting machine allowed the length (or tube) of a stocking to be completed in one step. The foot was then knit onto the length, which was standard practice until the 1970s.
1920’s – “Portable” electric sewing machine popularized;
The first practical electric sewing machine was invented by Singer in 1889, but electric sewing machines didn’t become portable until the 1920s. Though they were technically portable, these machines were both heavy and expensive. Sewing machines became much more lightweight in the 1930s.
1932 – PFAFF 130, the Universal Tailoring Machine, Released;
The PFAFF 130 was a high-performance zig-zag sewing machine, and it is still used widely by serious sewers who appreciate classically-built machines. It saw many industrial uses in its early days and was used by many professional tailors and seamstresses.
The 130 came about in 1932 but didn’t reach American shores until much later. The Necchi is widely considered the first zig-zag machine that came to America for home use (see below), but the PFAFF 130 was equally important. Likely because of its German origins, it didn’t make it to American shores until after World War II.
Though it is very heavy by today’s standards, the 130 was a high-speed and high-performance portable machine, used widely by the Merchant Marines after WWII because it was very effective at mending sales. Because of its versatility and popularity, it was known as the universal tailoring machine.
Overall, the PFAFF 130 is one of the most important sewing machines in history. It set a high bar for Pfaff’s future products– a bar they reach or exceed to this very day.
1947 – The Necchi introduces zig-zag sewing for home use;
Prior to the Necchi, zig-zag sewing machines only saw industrial use. Zig-zag sewing is used when straight stitches just won’t do. It’s used to reinforce buttonholes, stitch stretchable fabrics, and for transitional project work such as temporarily joining two pieces of fabric, edge-to-edge. Necchi brought zig-zag sewing into the home.
1949 – Heinrich Mauersberger invents the sewing-knitting technique and his “Malimo” machine;
The Malimo machine, which creates textile substrates through a stitch-bonding process at the speed of a sewing machine, was invented by Heinrich Mauersberger in 1949. Mauersberger got the idea from his wife, who repaired a piece of underwear by stitching both across, and up and down the damaged fabric.
The Malimo, in its various stages, bonded support fabrics to other fabrics, for the composite industry. Karl Mayer also produced notable Malimo machines. The Malimo was constantly upgraded and repurposed, and some versions of the Malimo machine are still used today.
1950’s – Portability and Variety;
Although portability and other sewing machine innovations began in the late 1940s, with the Necchi, sewing machines really started becoming lighter and more versatile in the 1950s. The Elna, a Swiss zig-zag machine from 1950, was made from a lightweight alloy and only weighed 19 pounds– much lighter than its 40+ pound predecessors. It also had a free-arm, ran quiet, and was featured in a new color: green.
Elna made further waves in 1952. The Elna Supermatic was the first machine to use cams or discs that were interchangeable, allowing for a wide variety of stitches.
1978 – Singer Invents first computer-controlled sewing machine;
Singer introduced the Touchtronic 2001, the world’s first computer-controlled machine, in 1978.
Source of text: Jones Sew & Vac.:
It may not be the most accurate and comprehensive historical representation, but we have tried to show that the textile industry has been familiar with change and innovations throughout history. The textile industry, for example, was the stage for Joseph Marie Jacquard, Charles Babbage and Augusta Ada King-Byron (Ada Lovelace) where they pioneered the development of the first programmable computer systems.
Unfortunately, what is being developed now is no longer as innovative as what the textile industry was known for. The industry still runs on inventions dating back to the previous two centuries. Unfortunately, time does not stand still, and so innovation will have to be done to help the sector enter the next century. People continue to need clothing; clothing will always continue to develop. But the big question is how the production sector will respond to this. What disappoints us is that the production sector is not involved in this discussion. This is entirely due to the passive attitude of the sector. However, do not be surprised if decisions are made for you, rather than with you, when the future is determined.
The next chapter will be devoted to the problem we have analyzed. What do we want to do about it, how do we think we can solve that problem and create a commercial plan for it?
Analyzing the problem and the solution
We have already mentioned it in the previous chapters, what is going on and what can be done? As discussed above, everyone will agree that what is currently considered normal within the industry should no longer be considered the norm. The way in which clothing is now produced is outdated. Work procedures are further eroded, causing job satisfaction to drop to very low levels. We must realize that this is not the solution. Working longer hours to achieve certain production volumes is only counterproductive. So, what do we think needs to be changed?
A general production employee stays with the same employer for an average of 2 years, after which he leaves the sector or looks for another employer. Why is that bond with an employer so low? Simple yet easy to point out, because of the mind-numbing effects associated with activities Certainly in the shadow of Industry 4.0 something can be done, but then a total solution must be found. A number of things are being done to re-develop production processes. But our research has shown that the sector itself is not or hardly involved in this. The ideas usually arise within the labs of technical universities or are coming from machine manufacturers who are already suppliers within the sector.
We understand that these machine manufacturers are approaching the problem from their own angle, but why are the “new” developers doing the same? What is the case, everyone working on it is now approaching the problem from the corner of the machine? Theoretically, this approach is justified, but in practice it works out differently, if you have never or very little time spent on the shop floor of a production company, you will never know the real problem.
The developments that are now taking place all have the same approach, they transport the sewing machine along the various pattern parts, as it were. Or the pattern parts along different machines. A clever idea, but the existing sector will not implement it quickly. If we do not involve the machine manufacturers in this part of the story, we are currently working on 3 solutions:
- Connecting the sewing machine to a robotic arm – this is not a solution for fitting clothes at all. What it can serve for are special activities such as sewing seat covers and mattresses or derivatives thereof. But that’s not the bulk of the job, so it will only serve a niche of the sewing industry.
- Sewbo – they come up with a creative solution by adding a liquid to the textile fabrics that stiffens the fabrics, then a robot arm can more easily handle the different pattern parts and feed them to the sewing machine by means of suction cups. Nice idea, but our reservations about this solution are twofold; it is only suitable for very simple work and what happens to the liquid when the assembly process is completed? Sorry for all the trouble, but we cannot consider this a serious solution.
- Softwear Automation – This solution has the best chance, in short, the people at Softwear Automation come up with a horizontal implementation of a three-dimensional problem. In our opinion, this immediately reveals the shortcoming. Over the years, the employees of garment production companies have been given an increasingly smaller workspace, due to economic motives, the smaller the work environment of one employee, the more employees can work on the same floor space. So why expect people to work in a 3-dimensional environment, but with automation this is not necessary. They show off data that they produce more than an average production worker, but they use the same floor space as 10 workers and use not 1 but 4 machines, so capacity should be compared to 4 to 10 production workers. It is very clever what they have come up with, but in our humble opinion it is only a solution if you fully attune the production to the sewing robot, instead of having it collaborate in an existing production line.
The second point of criticism is that we oblige people to perform very complex actions, but because we currently do not know how to automate these actions, we let the robot do everything in horizontal execution. One of Softwear Automation’s comments was that textile fabrics are not rigid and differ in composition and structure and that it is therefore very difficult to perform actions in a 3-dimensional version. But that testifies to absolute poverty. We oblige production workers to find out for themselves, while we, as the inventor of automation, show a lack of creativity and want to shape the situation according to our own flaws.
The following questions will have to be answered, what obstacles are we trying to tackle with the development of the collaborative robot? Does the robot by definition have to achieve higher production numbers than humans? We will answer this question negatively because speed does not always produce the desired result. We work with a large variable, the textile fabrics. The fact that each pattern piece has different stitch sizes is not the problem, but textiles have the maximum stretch and tensile strength with which sewing operations can be performed. If you pull certain fabrics too far while sewing, the stitching will not give the result you are looking for.
The last part of the previous question is easy to answer. The production result should never be less than the human variant. Then the next question automatically comes; measured to what? If the cobot achieves 10% lower production numbers per hour, have we not succeeded? We think so, because a cobot does not have to go to the toilet, no lunch break, or for whatever reason of possible lower production numbers. A cobot does its job and, if properly managed, achieves an equal number of production numbers, whether it is the first hour or last hour of the day.
The quality of the implementation cannot be less, from a technical point of view, a cobot must guarantee a better uniformity in quality. People can make mistakes under pressure, think of the panting manager, reason enough for people to drop a stitch here and there, something a cobot will not do quickly. Our starting point will therefore not directly be higher production numbers, but a better coordinated production in which we filter out coincidences. Can this be done on competitive terms? What do we have to compete with, is the next question? Costs: purchase costs and operating costs; the purchase cost is an isolated factor. Returning to what our competitors are doing, the purchase price of a Softwear production platform is one of the reasons why it will not be directly accepted by today’s apparel industry players. But what is a good price? You may wonder whether the cost price/asking price should be discussible at this stage. You could say: let the innovation speak for itself. I agree, but also not, because getting the innovation accepted by the industry really depends to a large extent on the purchase price. We keep a close eye, but the purchase price should be somewhere between € 10,000.00 and € 30,000.00. Operating costs is another factor that must be analyzed, what is the energy consumption, maintenance costs, time for maintenance that the cobot cannot be operational. Additional costs such as software and operating system updates.
Remember, the industry has benefited from low labor costs for the past 40 years. If a Cobot will cost € 30,000.00, compare that with, for example, 6 employees for a starting salary of +/- € 133 per month, for these amounts we can keep 3 employees working for 6 years. That is a rough calculation. But then you will have to come up with a lot of persuasion to have the work of these 3 employees done by cobots. Because can a cobot compete with 3 employees? In our experience, only if you let that cobot work longer than 8 hours a day, 365 days a year. Not all facilities are designed for this. But that is the immediate advantage of the cobot, it will not complain if it has to complete 2 or 3 shifts in a day. Then there is another issue that is never discussed, job loss. Something that gives us sleepless nights because that must be taken into account. The garment industry in Indonesia employs 5.2 million employees, which is about 4% of the total labor market. These are mainly low-skilled workers with few opportunities in other sectors. What will the innovation mean for them if we succeed in drastically changing production activities? We have taken steps in this sector to ensure a better future for employees. But our development can actually change the activities drastically and effectively reduce jobs. That is why we want this to be an employee participation project. A Cobot is a collaborative Robot and not an autonomously operating robot. The question is whether an autonomously operating robot will ever enter the textile industry. But you do not help current employees with that.
The solution that we can provide for this is to make training and (re)education part of the daily activities. The basic educational program should be taught during working hours. Because we want to make this part of the daily activities, you guarantee the involvement of the staff. But their input will also help to develop a better system. The cobot will be a continuous process of development for both the hardware and software side of the innovation. On the hardware side, it will be a collection of existing technologies that will be tailor-made. We do not want to install an existing (duo) robotic arm and then see how we can make it suitable. The mobility of the existing arms is less suitable for our purposes, as we have limited freedom of movement. It will be a combination of existing techniques. But that does not mean we want to limit ourselves, because should it consist of 1, 2, 3 or more arms (tentacles)? Will it be one system or multiple systems in line that communicate with each other? We demand a high degree of flexibility from ourselves to make the development applicable within different systems of production lines. While a total solution will be our preference, this does not necessarily mean that it will also be the ideal solution for every future user. That is why the total solution will have to consist of different modules that can be adapted to the model that the user requires.
The total solution will be a model with 2 production arms and 1 or 2 arms for the supply and removal of production parts. The production arms (right and left) will consist of an upper and lower arm, tailored to the human anatomy. The connection between those 2 parts will have a freedom of movement comparable to the elbow. The upper arm is given a rigid construction that has the function of bridging the distance between the cobot and the position of the operation. The forearm will have a flexible construction, comparable to a (vertebral) spine. As a result, the wrist joint may have limited freedom of movement in the palmar / volar and dorsiflexion. This improves the strength of the construction and we hope to achieve a higher accuracy. In addition, we have more freedom to adjust the pronation/ supination and adduction/abduction movements to our advantage. We can make use of these freedoms in design and construction because we are not limited by “too great” weights of the material with which the cobot has to work, textile materials are of course not heavy in weight. At most we are talking about kilos and not about tens of kilos that the cobot should handle. The accuracy of point of operation is much more important because we are tied to a static point by the sewing machine. The gripper / hand mechanism will, for the most part, mimic a human hand. Flexibility and grip in particular will have to be tailored as much as possible to the capabilities of a human hand. If this does not happen, our mission will not succeed in controlling the manageability of the various textile materials. Something that is very important in our plans. On the other hand, all the time and energy we spend on this will be a very good investment. Because if the mission is successful, the uses are endless, and the hand mechanism can become a commercial success.
The steps to be taken
First Step - Data Collection
The data we want to collect, and process has a dual purpose. First, it will become clear from the data what the radius of movement and freedom of movement should be. Let us be clear in one thing: when it comes to putting together clothes, nothing beats human execution. The way we can control the pattern parts, grasp, and move the different textile fabrics, solve problems that arise due to the flexibility of the fabrics because they often have to be twisted in every nook and cranny to get the right stitching, all without thinking. We can have a legion of staff; we have access to thousands of experts by experience. But if you ask them questions in this area, you will get hundreds of different answers, and in the end, they should owe you the real answer. Because the actions are performed without consciously thinking about it. In the first instance, we need the production data of the actions to answer important questions for the design of the cobot.
In the second instance, we can use that data again to make instructional software for the cobot. Because every design / pattern of clothing needs its own instructions and often hundreds of different actions. So, we want to try to make standard templates from the data we collect and then the specific properties per pattern only need to be supplemented. That is why data collection will always take place, because much can be learned from human executions. If we have good control of this process, an imitation module can also be placed where the cobot mimics all movements of the human operator. This can also be a more efficient way of working. In any case, for learning to understand all movements, by both the developers and the cobot itself.
To collect data, we will equip the human operator with sensors and store the data for analysis and processing. Our plans differ from what is being developed so far. We start from the existing production methods, i.e., three-dimensional production instead of horizontal production. The data collection is about digitally mapping every movement and action that a production employee does during the assembly process. This can be done by using a “motion capture suit”. It will not be a full body suit, but the upper body and arms and hands. So that we can digitally analyze every shoulder, arm, hand and finger movement. There will be censors in the outerwear who meticulously chart every movement. Not just once or with one person, but hours, days, months and maybe years in a row. For every product that is made, but also for all the different parts of a total production process.
These types of motion capture suits are widely used in the gaming and movie industry to mimic movements. We will initially use them to analyze movements. A software program / algorithm must be developed that can translate these movements for production purposes. We are going to use the staff present for the data collection, who will also become data processing assistants. They will continue to do their normal work, only every movement will be recorded in detail. This data collection will be done together with a University in Yogyakarta that has a computer science faculty. We will also employ data analysts on a permanent basis.
Second step - Design & Development
An industrial designer and his team will conduct research and development into robotic arms for light production purposes. These robotic arms, including the hands, must simulate the movements (mimicking) that are collected in current production processes. What that design will look like also depends on the data we collect. The data will have to show whether we should focus solely on the mobility of the arms and hands, or whether we should include the shoulders. So, it will be important to investigate not only the mimicking of the movements of the arms and hands, but also the spatial movement within the environment of a workstation, hence three-dimensional. Because with that we can make a total design. Within this step, the pressure sensitivity of the hand and fingers should also be an important point of attention. The manufacturing process is with textile fabrics, which are sensitive to strength during handling. New developments in the medical industry will also be reviewed.
In addition, we must be creative in development. Most robotic arms and hands are mainly motor-controlled. However, pressure adjustments of motor-controlled robotic arms, especially for the grip of the hand, are unlikely to be accurate enough for the correct pressure during production run. When it comes to pressure, don’t think in terms of strength, but rather in sensitivity. The attention will therefore also have to be on deviations in this. Then we may consider pneumatic or hydraulic applications. A lot of work can be done with sensors. A reference can be made to the modern braking systems of cars, not cables but sensors. More will become clear about this during the research period. The same goes for the different materials to be used. Are plastics sturdy enough, or should we start using light metals?
All departments will work closely together, because we want to work towards a prototype as quickly as possible. The first prototype will be a single operating unit, but we want future models to work together in a system context. Nevertheless, we will first go for a single operating unit because it will allow us to understand more about the technology needed for later models. In addition, it would be an ideal module if our employees could also work on the production units, if only to solve any obstacles and problems. Interaction between man and machine (Collaborative Robot or COBOT).
Third step - Operating System and Software development
The data collection, together with the prototype production unit, must lead to an operating system and software for controlling the production units. For this, the data analysts will work together with software developers and programmers. This software system will also get a “Machine learning” application. So that the computers and robots can learn from the data that we collect analogically from our employees. This includes pattern recognition and problem solving.
Both in steps 1, 2 and 3, much of the attention will be paid to classical Mechanics, for research purposes and in the development of the Collaborative Robots. Classical Mechanics, Kinematics and Kinetics and all subdivisions below. Here is a small explanation where research and development will mainly be done: Classical mechanics, also called Newtonian mechanics, is the form of mechanics defined by Isaac Newton (Philosophiae Naturalis Principia Mathematica, 1687). Classical mechanics is part of physics. Newton postulated his three laws of mechanics, making it possible to use mathematics in physics. Later, the work of Newton was built on by, among others, Joseph-Louis Lagrange and William Rowan Hamilton.
Classical mechanics is applicable in ‘everyday’ situations. Until Albert Einstein came up with the theory of relativity, physicists assumed that classical mechanics accurately described the movement of objects. From the beginning of the 20th century, classical mechanics no longer proved to be sufficient to explain all observations. Fundamental expansion proved necessary with the theory of relativity and quantum mechanics. Classical mechanics only applies when there are velocities that are small in relation to the speed of light, when the force of gravity is not abnormally strong and when the behavior of matter on an atomic scale is negligible. In everyday life, classical mechanics are therefore still sufficient.
The field of Classical Mechanics deals with the study of bodies (objects) in motion, in particular the physical laws that control bodies (objects) under the influence of forces. Many of the mechanical aspects of robotics design are strongly linked to the principles of this field. This section will describe some of the core concepts of Classical Mechanics that are important to us.
SPEED
A measure of how fast an object is moving. Describes a change in position over time (or more simply, how far an object will travel over a period of time.) This measure is given in units of distance per time (i.e. (kilo) meters per hour or centimeters per second).
ROTATION SPEED
Speed can also be expressed in rotation. This refers to how fast something is moving in a circle. It is measured in units of angular distance per time (i.e., degrees per second) or rotational cycles per time (i.e., revolutions per minute). When someone talks about “RPM”, they are referring to rotational speed. When talking about the RPM of a car engine, someone is describing how many revolutions the engine makes per minute.
ACCELERATION
A change in speed over a period of time is described as acceleration, the higher the gear, the faster the speed change. If a car goes from 0 kilometers per hour to 60 kilometers per hour in 2 seconds, this is a higher acceleration than if the car goes from 0 km / h to 40 km / h in 2 seconds. Acceleration is a rate of change of speed. No change means no acceleration – if something moves at a constant speed, it does not accelerate.
FORCE
Accelerations are caused by forces; they are influences that cause a change of movement, direction or shape. When one presses on an object, they exert a force on it. When a robot is accelerating, it does so because of the force that the wheels exert on the floor. Power is measured in units such as Pounds or Newtons.
For example, the weight of an object is the force on the object due to gravity (acceleration of the object towards the center of the Earth).
TORQUE
Force directed in a circle (rotating an object) is known as torque. Torque is a turning force. When the torque rotates an object, the object creates a linear force at the edge. In the case of a wheel spinning on the ground, the torque applied to the wheel axle creates a linear force at the edge of the tire where it hits the ground. In this way one defines the torque, a linear force at the edge of a circle. Torque is described by the magnitude of the force multiplied by the distance from the center of rotation (force x distance = torque). Torque is measured in units of force * distance, such as Inch-Pounds Newton-Meters.
KINEMATICS
is a branch of classical mechanics that describes the movement of points, bodies (objects) and systems of bodies (groups of objects) without taking into account the forces that caused the movement. Kinematics, as a field of study, is often referred to as the “geometry of motion” and is also seen as a branch of mathematics. A kinematics problem starts with describing the geometry of the system and explaining the initial conditions of known values of position, velocity and / or acceleration of points within the system. Subsequently, the position, speed and acceleration of unknown parts of the system can be determined with the aid of arguments from the geometry. The study of how forces act on bodies falls within kinetics, not kinematics.
KINETICS
In physics and engineering, kinetics is the branch of classical mechanics that deals with the relationship between motion and its causes, especially forces and couples. Since the mid-20th century, the term “dynamics” (or “analytical dynamics”) has largely replaced “kinetics” in physics textbooks, although the term is still used in technics.
Fourth step - Intellectual Property
Legally recording and protecting the systems and software that we develop. Both contractually with all parties involved, as well as patenting the systems, software and production units. We must ensure that we can claim intellectual property in one of these areas, so that we can exploit it commercially and that another party does not exploit our development without paying for it.
We will put together a legal team for this, which will focus on all issues related to this. This also applies to the employment contracts of our employees and any future partners. But also, for the legal framework if we use the technology of others in our cobots. These are complex matters, but very important for our future and that of our shareholders and financiers. All this will relate to local law as well as international law.
Fifth step: Implement Cobot in production facility
Coordination with existing production lines. The automated production units must be integrated and installed in the existing production lines. Will they work in combination with the production employees, or will the production units work completely autonomously, possibly with the guidance of employees? The employee as a teacher and navigator for the automated production units. In addition, the entire production line will have to be set up even more systematically for the use of these new automated production units. This brings about an even higher degree of efficiency in the production process. The current production procedure is too dependent on the problem-solving capacity of humans. Not taking into account a decrease in production speed.
We on the board are not computer and robot experts, but we consider ourselves to be creative thinkers, where we care about the health of the companies in general and the well-being of our employees in particular. According to the Danish philosopher Christian Madsbjerg, we do not all need to be computer and robot experts:
“His argument is a settlement of the belief in exact sciences that he believes has gone too far in which computer science spews enormous amounts of ‘hard’ data, without being able to give it meaning in the most intelligent way. Because for that, Madsbjerg thinks, you really need a foundation in the “soft” sciences, such as philosophy, anthropology, social psychology and political science.”
All the research has been completed so far, now we will have to proceed to the actual design of the cobot. We have a clear idea of what this should look like, but a prototype provides complete clarity as to whether we are on the right track in research and development.
Implementation of innovation
Before we can talk about or go into the next phases, we will need to have a prototype of the cobot at our disposal. All our energy, time and attention will be devoted to the development of this first prototype. We expect to have the first prototype available for testing and using it within a development period of 9 months. This prototype will be able to simulate movements of a human operator (mimicking) or be controlled by an operator. We want to be able to understand the technology of the cobot before expanding the cobot with a semi-autonomous control system. Mastering the technology will be a very big task. But if it works, we can tune the operating system accordingly.
The reason why we choose this route is that we do not want to do double work, we can only tailor the operating system if we know what the possibilities of the cobot are. Does the cobot have the movement options as we expect them? Again, theoretically everything should be realizable, but is it actually applicable? Therefore, we are cautious about the implementation of the following stages of development. Everything stands and falls with the possibilities of use of the cobot. What we want to achieve may be new, the techniques we use for this are not new. If we do not have to reinvent the wheel, we certainly will not. But in this composition, in which we want to let all parts co-operate together, was never done before and therefore we need to know how everything reacts to each other.
The reader may think that a cobot without an operating system is not a cobot, but we are talking about 2 different things. From the very first development, the cobot will have a circuit management system that allows the robot to move by means of sensors and actuators. Comparable to a fly-by-wire system in an airplane. The operating system we are talking about is the computer-controlled management system that will control the entire operational activities of the cobot without the need for direct manual signals to perform tasks. When the cobot has a general operating system, it becomes easier to integrate the cobot into the total production line. As soon as the cobot has a computer-controlled management system, it can communicate with the digital sewing machine and respond to the supply and demand of the production line’s distribution system. In addition, we can build in a progress and warning system that monitors the progress of the production and detects any errors.
The ultimate goal that we try to pursue is the full integration of the cobot in the total production line. It is a gigantic task that awaits us. Because it is one thing to master the technique of the cobot, a completely different task will be to give the cobot instructions with which it can perform all actions independently. The collection of data from production movements will also play a crucial role in this. An algorithm will be generated from the data with which we are able to create a series of instructions for the cobot. We will have to create a library of all movements in combination with pattern parts. At this stage, we cannot go into detail on how the series of instructions is constructed. Much of the innovation will be developed in open source, but we do not want to disclose all the details in advance. We are going to invest a lot of time, money, and energy in the development of the cobot, we want to be the first to benefit from this. This for us, but certainly also for our investors.
Once the prototype is operational, we will propose to the investors to separate this project from the rest of the business. A special purpose entity (SPV Company) will be established. This becomes the project company in which the innovation project will be further developed. This has the advantage that the new company can be structured in such a way that it is suitable to attract further financing. Various scenarios are possible in which we can secure the necessary funding. One option could be to expand the convertible bond program. Another scenario could be that we bring one or more strategic partners on board. For this scenario, we will have to set well-defined requirements that a strategic partner must meet. For example, a strategic partner should never demand that the cobot be made available only to them. Because then we may develop something that we ourselves are not allowed to use.
In addition, we are explicitly examining the various options within which we can guarantee the financing up to the conversion into shares. Again, the first phase of development up to a prototype is realistically possible within the budget of three-quarters of a million euros (€750,000). It should also be understood that this amount will not be nearly enough to develop a final product. We expect that an amount of up to 50 million euros will be needed to be able to deliver a complete commercial product ready for production. We have often wondered whether an investment of 50 million euros in an innovative project is feasible. It is, of course, a large sum of money, but it is not the end of the world. We will have to sell 20-25 thousand units to recoup such an investment. That is at an average selling price of € 30,000.00 each and a margin of 7.5%. In other words, a market position of 0.5% to a maximum of 1%. In this calculation, we only take into account average estimates of employees who perform work that the cobot can perform. Then we will have to see if the 0.5-1% is a high number or at least reasonable. For us that is just the beginning, because if we succeed in our goal, then a market penetration of 5% is the minimum we want to achieve. In other words, more than 110,000 production units.
We expect that there will be many more production houses, on a smaller scale, but closer to the consumer. Then we will be in a much better competitive position, since in the West the purchase value of the cobot will not exceed the annual salary of an employee. If the product is good, then there is a market that will far exceed 200,000 production units. This is only in the production of clothing, for general production. The more production units (cobots) we can place with customers for a longer period of time, the more data we can retrieve to improve the technology and thus make the production unit suitable for total production, general production and finishing. We are talking a million-dollar market in the apparel manufacturing sector, but an even bigger market for all light assembly manufacturing awaits us. We are talking about a future where everyone understands why we split off the innovation project and developed it independently. Then we have not even included the income for service and maintenance, and of course the subscriptions for our template library.
Risk of failure
To be on the safe side, let's look at the project from the other side: what if we fail in our goal of developing a semi-autonomous cobot for the production of clothing, will the entire investment be lost?
We can answer this question immediately with a resounding NO. This innovation project is not set up like this, we are not going for an all or nothing approach. We are well aware that we are pursuing a audacious goal and there is always a risk that what we are trying to achieve is simply not achievable. Therefore, we will have to be conservative when raising funds from third parties. As long as no prototype is available, we will only raise funds that are needed. It would not be appropriate to park large sums of money in a bank account without a direct purpose.
What if we cannot realize a prototype?
We are 100% convinced that a prototype can be realized. We have done so much research that there is no longer any doubt about it. However, this prototype will not yet have a semi-autonomous operating system. There is a greater risk for us in the development of the operating system. Because that has never been done before in the format as we plan it. Theoretically it should be correct, but the question is whether we can actually realize it. We will never find an answer if we do not try. The solution we want to work out is simple, but that immediately begs the question why no one has tried this before. The danger can arise that we overlook something, creating a blind spot for ourselves. If we manage to do this, the world will be amazed at how simple the solution is. As we have already described, we can only find out when we have a working prototype available. In simulation we have a total solution for the operating system, but simulation alone is not enough.
But what if we fail to develop the desired operating system model?
That could be a major setback for the project. We have to be honest about that. But at that point, nothing should be considered lost yet. Because if the project failed, it would mean we gambled everything on one horse. That would be terribly stupid and a waste of all the effort, time, money, and energy we put into it. Unfortunately, there is no fast way to the finish, if necessary, cutting corners, because nobody walked that way before us. We will have to get to work and figure out how to achieve the goal we are trying to achieve. We think we can go very far and have a very reasonable chance of success. Fortunately, we can make use of age-old concepts from mathematics, the polar and Cartesian coordinate system. Forward and inverse kinematics help a lot with this. Yes, every movement will have to be fully analyzed and calculated, that is a hell of a job, but that’s why we attach so much importance to collecting data from our production employees.
This just does not answer the question, what if. Imagine that we are completely wrong and cannot develop the desired operating system model. Then we have the module of simulation and manual operation. This can be a solution, but for production purposes it is only possible to take advantage of this if several robots can be controlled at the same time. Yes, this limits the use, but the work and investment involved should not be considered lost. It will be a longer journey to market the cobots commercially for a broader user group. But sooner or later we will get there. The great thing about innovation is that what is not possible today can be a solution tomorrow. As long as we keep looking and be creative in the work we do.
Are there other revenue models if not everything of the project can be realized?
Yes, we can make the data we collect from the production movements available to third parties. We only do this if we cannot make full use of it. This is not our preference, but if necessary, we can at least recoup the costs associated with collecting the data. But again, only in extreme necessity.
Then there are various parts of the cobot that we are sure are popular with other parties. For example, the flexible forearm, a kind of gripper we work on and joints. Chances are we will trade those models on a commercial basis regardless of the success of our own project.
In addition, we can also have the design and development department carry out assignments from third parties. Especially in the field of prototyping and small-scale production for collaborative robots, there is a great demand for specific knowledge.
Whether all this will be enough remains to be seen in the future, but we are very optimistic. The whole project is set up in the form that we invest our own money. Doing business involves risks, but unnecessary risks are certainly not one of them. That is why we have thought a lot about how to overcome the potential risks and we are convinced that we have taken care of them.
Future expectations
To conclude this article, we will express our expectations for the future of this project. We expect to be able to present a first prototype within a year. At the same time, we will also have to manage data collection. There is more to be gained from that data than you would expect, which is why we think it is so important that this is done very accurately, because in the longer term we will only benefit more from it. It will benefit the production of clothing; it will be possible to make a big step forward in terms of quality. Because by properly processing the data, we can rule out weaknesses in production. The same goes for efficiency in production.
We expect that with the introduction of the first prototype, the project will be split off. That is a suitable moment for us, there is also a product to show, it is no longer an idea that only exists on the drawing board. As we are going to publish the prototype, there will also be criticism of the techniques we use. This will only benefit further development; we are absolutely not afraid of it. The same goes for anyone else who wants to surpass us. Competitors making the product better, otherwise we do not deserve to market a commercial product.
In our opinion, the split will not yet be the time when the conversion will take place. We will advise participants to be patient. A prototype is not yet a product, the valuation will still be low. However, a larger funding round will have to take place. As long as this is done under the right conditions, it will not be disadvantageous for the participants present. The money is needed for the development of the overall operating system. This is the moment when we will have to fight to be able to control the cobot semi-autonomously. The further development of the cobot to a commercial model will be many times faster than the development of a general operating system. Because with the operating system we are dealing with an empty sheet of paper, while with the cobot we have a very clear frame of reference in the prototype. We will have to fight on 2 fronts, the hardware side and the software side.
On the other hand, this is also the moment when we can show the outside world where the power of this cobot lies. We expect that in this phase there will be spin-offs of components that we can market commercially. Do not forget that the development of industrial robotics is very fast, but people do not always look at solutions creatively. So, if we come up with that bit of creativity, it will have an effect that we can reap the benefits of. This can cover possible costs for further development, although we believe that this may also be a time when we, together with the participants, choose to have the conversion to shares take place. Because in addition to a product, there is also a revenue model that guarantees a stronger basis for the future. This does not necessarily have to be an IPO, but instead of raising further funding through convertible bonds, we will be making shares available within the project company.
As soon as there is a commercial model of the cobot, there will gradually be more emphasis on the software side. The further development of the cobot continues, specific parts are produced in-house, but it is not necessary to do the entire production of the cobot in-house. Test and research center has been there from the start and will always be there. The production of non-critical parts can just as easily be done by others. It would be inappropriate to preach that clothing no longer has to travel 10 thousand kilometers before it reaches the consumer’s wardrobe, to ship parts of the cobot around the world while production can also be done locally. Production costs, which are therefore more expensive, are likely to be offset by transport costs, which will be many times lower. If this is not the case, still we will not invest in a production line yet, because then we will outsource the production of parts locally. As long we keep the quality control in our own hands.
It will not be hundreds of thousands of cobots per year, so we do not have to set up a complete production facility and focus on it. It is better for us to pay attention to what we are good at and that is to continue to improve the product, to continue to assemble clothing ourselves, so that we know what is going on in the workplace and that we serve the users well. We at P.T. Emas Cemerlang Bersama will always be involved in the project. There is always something to improve. For a long time, there has been no profound innovation for the apparel manufacturing sector, and we want to help ensure that this changes permanently. Because we continue to produce clothing ourselves, we know where the defects are, but we also have insight into where production should go. The time of mass production has come to an end. We ourselves expect that production will be closer to the consumer. So more small-scale production houses will be created locally. Our cobot will certainly help with that, but we must therefore continue to pay attention to what our customers want from us. As we have written before that the cobot will be a joint development with our employees, the customer will also play an increasingly important role in this. Consumers will demand even more for faster turnaround times for collections, and we expect consumers to pay more attention to personalization of clothing. Then the sector is almost automatically obliged to have production take place closer to the consumer.
We are already studying this possibility and it should not be surprising if within a few years we have several production sites on all continents, or at least collaborate with different local producers. Designers and e-commerce platforms will be working more closely together, which means that there seems to be a greater role for us if we can again offer our services to these e-commerce platforms by producing for them instead of for the clothing brands. This production can be controlled centrally, but can be carried out locally, as long as we have access to the cobots everywhere. Our developments will mean that more than enough jobs will be lost in production, but more jobs will be created in the creative process or in the management of production. We will have to prepare our people and future staff for this, and this can only be made possible if we continue to pay attention to education and training. Nobody should be left behind in our struggle for innovation. We also hope to find the readers in it and to jointly build a sustainable future for the clothing that we will wear for a long time to come.
We hope this document has provided more insight into what we want to do and what we are trying to achieve with it. It has become a general story, but we have tried to cover all the parts in it. In subsequent articles, we will take a closer look at the various components and the progress of the innovation.
If you have any questions or comments, you can reach us as stated below. Thank you for your attention.