Software automation A robotics company that wants to make T-shirts. “We want to make one billion T-shirts a year in the United States, all of which are tailored to demand,” said Palaniswami Rajan, CEO of Software.
The company entered into an agreement with Darper in 2012 with the support of the Georgia Tech Advanced Technology Development Center. Two years later, a prototype was launched and running. Work began in 2017 to build a production line that could mass-produce shirts. That same year, the company struck a deal with a Chinese clothing manufacturer to set up a large manufacturing facility in Arkansas. Although that deal was broken and the software is now focused on opening its own garment factory.
The time it took to get to this point is not surprising. Machines have proven to be proficient in many steps in making fabrics, from printing fabrics to cutting and folding fabrics and packaging finished garments.
But sewing is notoriously difficult to automate, because they work as textile bunch and stretch. The human hand is adept at keeping the fabric organized while going through the sewing machine. Robots are usually not efficient enough to handle the task.
Software’s robots have overcome that obstacle. They can make T-shirts. But making them as cheap as workers in places like China or Guatemala, where workers earn a fraction of what they can make in the United States, would be a challenge, said Sheng Lu, a professor of fashion and clothing research at the university in Delaware.
The software calls its robotic systems cellbots. These are basically wide work tables that combine sewing machines with complex sensors. The company diligently protects the details of how they work, but here are the key points: the fabric is cut into pieces that will become part of the shirt: front, back and sleeves. Those pieces are loaded onto a work line where, instead of a person pushing the fabric through a sewing machine, a complex vacuum system expands and removes the material. The cameras track the threads on each panel, allowing the system to adjust while the garment is being made.
But no two batches of cotton are exactly the same, often varying from crop to crop; Variations in fabric and color make matters more complicated. Each variant can be used to recalculate the system, disrupt operations, and train the software to respond accordingly. “The biggest challenge we face in a manufacturing system is being able to operate at high speeds 24/7 and more than 98 per cent quality,” Rajan said.
Garment factories churn out more than 20 billion T-shirts a year, most of them outside the United States. In order to make T-shirt production possible in the United States, it must be cheaper than import. But eliminating shipping costs and import duties is not enough to cover the cost of sewing garments for U.S. workers. The Bureau of Labor Statistics says the average U.S. sewing machine operator gets just $ 28,000 a year. That’s about $ 13.50 per hour অনেক a lot more than the countries that currently make a lot of t-shirts. Professor Lu of Delaware said wages in China for such work are about one-third of wages in the United States, while in Guatemala they are less than one-fifth of US wages.
Concentrating on T-shirts allows software to keep away from another problem with automated sewing systems: switching from one type of clothing to another. A skilled team of people can sew short sleeve men’s shirts one day and women’s jeans the next day. Such changes are even more challenging for robots. The way a cotton polo is sewn together is significantly different from making a pair of polyester pants. Creating a new work line for cutting different fabrics and sewing different stitches is complex and expensive. Once production for T-shirt making begins, it will be difficult to quickly reconfigure cellbots to make anything else.
Since the initial funding, the software has raised 30 million in venture investments and grants – including a মার 2 million grant from the Walmart Foundation. Rajan said it would take a few more million to produce 1 billion T-shirts every year. To reach that goal, the company will need multiple facilities, each with its own cellboat and skilled workers for their maintenance. A Sewbot work line can make a T-shirt every 50 seconds, Rajan said. At that rate, if run continuously, a work line could produce only 620,000 T-shirts per year অর্থ meaning it would take 1,607 cellbots to reach 1 billion a year. Rajan said the more realistic number is closer to 2,000; So far, the company has done less than 50.
Robots inevitably arouse suspicion of displacing people and destroying jobs. Rajan acknowledges that the software will hire fewer people than a traditional thematic T-shirt maker, but he believes his company will create higher-paying jobs for people who will maintain the machines. “You want to develop the staff, and you want to train the staff,” he says. “Our aim is efficient labor and fast, agile production.”