A New Kind of Co-Worker
When executives at the Earle M. Jorgensen Company decided their Schaumburg, Ill., plant needed a major upgrade, replacing people with machines was high on the agenda. Manually stocking and retrieving steel products for the company’s three-shift central hub was holding back the firm’s ability to deliver inventory to satellite service centers—not to mention customers. The solution: a 60-foot-tall, custom-designed Kasto robotic inventory system that automatically loads, weighs and delivers pipe and bar to the machine that needs it.
Or rather, the Kasto system was part of a much larger solution. Kasto, though it has plenty of developing competition, seems to be the market leader at this point. Like many service centers, EMJ Schaumburg had grown piecemeal for decades, with saws and storage racks added when and where space could be found. The company spent $40 million and more than five years upgrading a facility that acts as a central warehouse for more than 30 locations across the United States. The process required the company not only to add advanced machines, but also to rearrange the entire layout and flow of the facility and install new, integrated software systems.
The result: Trucks that once drove an entire mile within the service center’s walls now might move only 1,000 feet, and an order entered by a salesperson in Dallas will send the Kasto system into action, delivering tube or bar to an automatic high-speed saw in Illinois. The Schaumburg location moves two-thirds more inventory than it used to. Yet the plant still employs 300 workers, down just 100 from before the process began in the late 1990s, say executives at Reliance Steel and Aluminum Co., which acquired EMJ in 2006.
“We didn’t put that system in to cut jobs,” says James D. Hoffman, senior vice president of operations at Reliance and EMJ’s former operations chief. “It was a service play. We’re in the service center business.”
Lights Out, On or In Between?
While EMJ reaped huge rewards from its technology investment, for many a prognosticator the robotic future has a ways to go before it lives up to expectations. The “lights out” factory—the kind in sci-fi flicks where machines run the machines without a hard hat in sight—is only now starting to appear in some specialized areas. In the metals services industry, the idea of a worker-less service center now seems misguided, if not downright naive, say executives and vendors. Squeezed between customers and suppliers that thrive on large production runs where robots are making inroads, metals processing is going the opposite direction for many, toward smaller orders of more-customized products and more value-added services that please clients.
That means the more likely future for service center automation, they say, is incremental improvements—faster bundling on this machine, reduced setup time on that one—rather than processing facilities that run themselves. Cloud-connected saws may relay data to operators that helps them reduce wear and tear, cut downtime and improve performance by avoiding costly repairs. But investment in technology will vary dramatically from one service center to another, based on location and size, product mix and the financial resources of their owners.
In the States, we don’t see the same levels of automation. It’s expensive. In Europe, there’s more justification.”
The competitive pressure to upgrade technology may also be helped by continued consolidation in the industry. Reliance, for example, has acquired 56 service centers since 1994, says chief financial officer Karla R. Lewis, and prides itself on giving acquirees the resources to upgrade.
“If you can provide the customer with a better product, we think you’re going to pick up some market share,” Lewis says. “If you’re a smaller company that does not have the financial resources to invest, you’re at a disadvantage.”
In part, the slow pace of technological change reflects an industry that perceives itself as following trends, not creating them.
“The service center isn’t a leader in technologies,” says Jim Sloan, operations director at Klein Steel in Rochester, N.Y. He rates the industry “a three or four” out of 10 on technology adoption (Klein gets a five or six, he says).
That assessment smarts at a time when a new wave of automation technology—led by advanced robots, machine learning and 3-D printing—is shaking up the broader world of manufacturing and enabling a return of production to the United States from overseas where once-cheap labor costs are soaring.
Though the U.S. manufacturing workforce’s percentage of total employment has plummeted from the mid-20th century (from 28% in 1962 to 9% in 2011), it has by and large held up in recent years while productivity has increased. Metals services have benefited from that trend as well. Employment in what the government considers primary metal manufacturing, which includes some metals-processing activities, dropped from around 500,000 workers in 2003 to fewer than 400,000 in 2013, yet industry productivity rose almost 40% in the decade ending in 2011 (the most recent for which Bureau of Labor Statistics data is available).
“Automation advances are becoming cost competitive with the alternative of high labor staffing,” consulting company Deloitte said in its 2012 report, “The Automation Evolution.” “The next evolution of manufacturing automation may stem the outflow of manufacturing from developed markets.”
That evolution may result in some completely automated manufacturing, but more prevalent will be robots that work alongside humans to aid in complicated tasks such as assembly and welding, says David Bourne, principal systems scientist at Carnegie Mellon University’s Robotics Institute, who has worked with the Japanese sheet metal machine maker Amada.
“The 30,000-foot view is all about flexibility, all about maximizing customization, minimizing setup time,” says Bourne, who cringes if people ask him when “lights out” will finally arrive. “How do you make machines flexible enough to go from one run to another? The batch sizes are going down and down and down.”
He cites a project for the Defense Department as an example of what future automation might look like. Assembling a Hummer frame involves welding together around $400 worth of square steel tube, a job that was taking 89 man-hours. Bourne set up a robot to run alongside an assembly worker but, contrary to how things typically work, the robot was asking for pieces and the human was bringing them over to be welded.
“So the person is doing the part handling. Think about that,” crows Bourne, noting that his Hummer was put together in 10 hours with a single worker.
For U.S. metals processors, robotic buddies are a long way off, though. In fact, when it comes to investing in automation, U.S. service centers badly lag behind even their European counterparts, says Kasto’s chief executive, Werner Rankenhohn.
European centers, says Rankenhohn, are smaller and often family owned, tend to have significantly higher labor costs, limited access to land for expansion and a longer time horizon for earning a return on investment. Of the 1,500 Kasto automated storage systems he has sold, only 50 are in the United States. Just one U.S. customer has purchased Kasto’s most advanced system for removing and stacking finished product from his high-speed saws.
“In the States, we don’t see the same levels of automation,” agrees Dean Linders, vice president of marketing and sales at Red Bud Industries, in Red Bud, Ill., which makes slitters and other machines to process coil steel. “It’s expensive. In Europe, there’s more justification.”
A Safer, More Efficient Process
While labor costs are a factor in the United States, technology investment is often driven more by customer demands, competition, lean inventory management practices and safety.
“This isn’t about eliminating people or positions,” says Klein Steel’s Sloan. “This allows us to do more.”
When Klein began upgrading its facilities a decade ago, its aging Rochester plant—a century-old, T-shaped foundry never designed for metals processing—still had dirt floors in some areas. One result: 600 worker-days a year lost to accidents, some of them major, most of them back-related.
“Our philosophy 10 years ago was to eliminate jobs we wouldn’t do ourselves,” says Sloan. That included bending over to move loads of steel, using handheld torches and drills, and pulling sheets onto and off of forklifts.
Klein built a new facility (with significant financial help from a tax-exempt county bond) and sent executives to other service centers, including EMJ Schaumburg, to get a look at the latest technology. The firm bought a Kasto system for bar and tube, and has upgraded saws, cutters and other machines. It still uses sideloaders—albeit fancy ones with remote cameras for inspecting inventory—for sheet stocking and retrieval. As with EMJ, getting the most out of technology meant rethinking Klein’s entire process.
“The biggest hurdle is probably the culture change,” Sloan says. “We took a lot of team members here who haven’t seen a lot of computers, much less controls for a $3- to $4-million Kasto system. We wanted everyone to make the journey with us with this new technology, and most did.”
“The 30,000-foot view is all about flexibility, all about maximizing customization, minimizing setup time. How do you make machines flexible enough to go from one run to another?
The company adopted the kaizen system of incremental improvement made famous by Toyota and later installed computer kiosks around the shop floor so that operators could key in inventory transactions while high-speed saws did their work. Next, the company plans to adopt a fully integrated bar code system for tracking raw materials to finished products.
Klein’s now-expanded Kasto system cut the number of employees retrieving orders each night in about half, while tripling the amount of metal they can move in the wee hours. In 2012, Klein had just one lost-time accident (Kasto’s Rankenhohn says his system saved one Kansas City center $200,000 a year in workers’ compensation premiums).
But the biggest bang for the buck, Sloan says, has been allowing Klein to expand its product mix, cut turnaround times and get creative with customers. Klein installed a machining center with lathes and drills to capture new markets, and works with larger customers to understand their manufacturing process and pick out areas where Klein can do it better, faster and cheaper.
One example: A manufacturer was buying lots of raw steel tube from Klein to construct a complicated housing for the large, shaky machines it sells. Klein employees went to the customer’s plant and learned the manufacturing process. Today Klein sells the company a ready-to-assemble kit of 30 parts with pre-drilled holes for hydraulic and other lines that the client can set up without any additional machining or processing.
“You’re really talking about value, not price,” says Sloan.
Justifying a Big Change
Reliance-EMJ takes a similar attitude toward automation.
“We do $10 billion [in annual sales], $1,600 at a time,” Hoffman says. The Kasto system essentially eliminated inventory errors at EMJ Schaumburg, cut monthly remnants from 5,000 units to 50, and reduced the physical counting process from three days to three hours, say Reliance executives and Kasto’s Rankenhohn.
But Lewis also points out that the Schaumburg location was a special case—the Kasto system required five-foot-deep cement floors—and not all service centers can make economical use of mammoth technology investments. Schaumburg sends at least one truck a day to each of EMJ’s facilities nationwide, reducing their inventory needs and allowing the Illinois hub to buy in bulk at cheaper prices.
A service center needs a minimum of 150 orders a night for a Kasto system to make financial sense, Rankenhohn says. The typical system the company sells costs a little more than $2 million.
So, What’s Next?
As for the next technological advance, look for improvements in getting materials to ever-faster saws—not a big breakthrough. Some parts of the service center don’t lend themselves as well to automation: The packaging and loading area still needs bodies, even if there’s a crane doing the heavy lifting.
“It’s not about making the machine go faster,” says Klein engineer Brad Cooper. “It’s about getting the material to the machine.”
Software, too, will play an ever-increasing role, says Kasto’s Rankenhohn, who likens the next wave of invention to automakers’ systems for monitoring their products post-sale. Kasto’s most advanced saws tell the manufacturer how long they’ve been running and how much they’ve cut. He envisions a service that alerts his customers to problems before they result in a dulled or broken blade or even, perhaps, reveal why a service center isn’t as productive as it could be.
Reliance’s Hoffman doubts he’ll see autonomous machines and workers mingling on the shop floor anytime soon. But, he says, Schaumburg does have one robot that removes six-inch rounds from a high-speed saw and puts them in a box. It was a gift from the manufacturer to entice future purchases. Says Hoffman, with a chuckle: “It’s neat to watch.”
Peter C. Beller is a Los Angeles-based business journalist and editorial director at Ebyline.com. A former staff writer for Forbes and MarketWatch, Peter’s reporting has appeared in The New York Times, New York magazine, the Jerusalem Post and elsewhere.