Textile manufacturer rolls out automation, one line at a time

When people think about tech companies, they most likely think about hyperscaler cloud providers or social media platforms. But increasingly, they should probably think about manufacturers.

“I consider us a tech company,” says Adi Nov, chief information officer at Delta Galil Industries, a global manufacturer and marketer of private-label apparel products as well as the owner of leading apparel and fashion brands such as 7 For All Mankind, Schiesser, P.J. Salvage, and Eminence. “We’re collecting data from machines and getting them to talk to each other. We are a data-driven company.”¹

Nov considers automation to be an ongoing process with no specific beginning or end point. As such, Delta Galil has been undergoing a series of transformations that are helping make the company more efficient.

On the traditional IT side, there’s robotic process automation . Delta Galil has been using robotic process automation to automate many back-office processes that previously required a significant number of staff hours. For example, the company’s clients provide sales forecasts that determine how many units of various items need to be produced. In the past, someone at Delta Galil would have to log into a company portal, download these figures, compare them with production capacity at its manufacturing facilities, and then allocate the order. Now this process is largely done by robotic process automation software.

But that’s just a start. On the operational side, Delta Galil is rolling out tracking and automation tools to help boost quality and efficiency on manufacturing lines. It recently started equipping manufacturing lines with radio frequency identification scanners, which help facility managers track the efficiency and the quality of the production lines. Scanners read tags as items move across the assembly line, and progress is tracked on big screens on the production floor. This fuels friendly competition between the lines, which helps boost productivity.

“This lets us identify areas on the line where a problem is occurring rather than looking at the entire line as a whole,” Nov says. “It gives us better efficiency.”

Nov and his team are also starting to work with more machine data. Manufacturing equipment produces huge amounts of data, but in most industries, it’s not yet tied to enterprise data platforms. Nov changed that at Delta Galil, which is now capturing and storing the data from equipment such as knitting machines that produce socks and other seamless items.

Right now, Nov’s team is mostly focused on collecting the data of quality and efficiency issues. He says it can take a few more years before the data starts pointing toward more insights. But with historical data in hand, many kinds of learnings emerge. For example, Nov expects to be able to tell if a knitting machine is running slowly because its needle needs to be replaced, or if the external temperature affects equipment efficiency.

“I don’t know what insights we’ll get because you have to have years of data, but I expect it will help drive efficiency on our lines,” Nov says. “The first step is to have the data.”

Unlike digitally native companies that were founded and matured entirely around digital processes, the textiles and manufacturing industry was mechanized during the Industrial Revolution and has been evolving since the 18th century. It will always involve physical operations. For this reason, Nov acknowledges that achieving greater automation is going to be a long process. Much like digital transformation, automating manufacturing operations is a continual effort rather than a one-and-done project.

“In the textile industry, people are still sitting at machines cutting and sewing like they did 100 years ago,” Nov says. “The challenge for the future is putting automation into the industry. We’re looking at it not as a destination but as a journey.”