3 keys to metal fabrication productivity

FIGURE 1. Today’s powerful fiber lasers produce stacks of sheets at the manual denesting station, an area that has become a common bottleneck. Images: AMADA AMERICA

“Productivity is never an accident. It is always the result of a commitment to excellence, intelligent planning, and focused effort.”

Phil Picardat, pointing to his leading slide at the Fabricators and Manufacturers Association Annual Meeting earlier this year, was quoting Paul Meyer, long-time self-improvement guru. The automation product manager from AMADA AMERICA led with that quote, which effectively summed up the message of his entire presentation. For decades, metal fabricators have equated productivity to effort—how to get more parts per hour out of a press brake or cut faster with an ultrahigh-powered fiber laser.

Still, “focused effort” is last on Meyer’s list; “commitment to excellence” and “intelligent planning” come before it—for good reason. A laser might cut extraordinarily quickly, but what if parts are denested haphazardly or, even worse, become lost in work-in-process (WIP) inventory? What if a brake operator speeds through a batch of parts, only to find them all rejected by quality assurance? What if parts wait in front of a welding robot as an operator struggles with excessive fit-up variation?

Most North American metal fabrication occurs in high-product-mix environments. On a given day, a fabricator could be managing dozens or even hundreds of different jobs across the shop floor. That makes thinking broadly easier said than done, but in the quest for true productivity improvements—that is, more products shipped out the door in less time—thinking broadly is now a necessity.

To that end, Picardat described three ingredients that together create sustainable productivity improvement: (1) review the workflow, (2) improve communication, and (3) monitor utilization. Addressing all three can help address the root causes of systemic challenges and, not least, make the most out of the latest metal fabrication technology.

“In the past, you identified a bottleneck, then bought more machines and hired more people to push products out the door,” Picardat said. “Now, that’s really not an option.”

This has to do with the lack of skilled labor and, for that matter, the lack of labor period. The COVID recession (and subsequent stock market rebound) spurred what’s now being dubbed as the Great Retirement. With experienced operators heading for the exits, who will fill their shoes?

“There are about 157,000 brake operators in the United States,” Picardat said, “and the country has about 57,000 open press brake operator positions. Also, nearly 60% of brake operators have less than two years of experience. At the same time, most press brakes in the field are more than 15 years old. … In the past, when a major new job came in the door, what did you do? You hired more people. Now, you just can’t do that.”

Decades ago, a fab shop might have landed a large contract, then “hired up” to meet the demand, bringing on more welders and brake operators. Some had more experience than others, but the shop at least had enough skilled people to bring the novices up to speed. This in turn grew the skilled labor pool to support the fabricator as well as the local manufacturing economy.

The Great Retirement, coupled with increased demand from reshoring, changed everything. Starting after the pandemic, experienced people simply weren’t there. Every new hire needed significant training.

FIGURE 2. Automated part sorting after laser cutting helps free the manual denesting constraint.

Combine this challenge with old equipment and issues with overall workflow, and you have a fabricator struggling to deliver on time, expand customer portfolios, and grow. All this hinders career opportunities for the fabricator’s best and brightest, who soon head for the exits. And so the vicious cycle continues.

To avoid that vicious cycle requires taking a step back. First comes the low-hanging fruit: the basic elements of lean manufacturing, like clear and detailed work instructions, organized workplaces, tools near the point of use, and overall standardization. If, say, brake operators on different shifts change the bend program or even the tooling, consistency problems will persist no matter what other actions a shop takes. In-process inspection and other best practices behind quality-at-the-source methods apply as well.

Next comes an analysis of the value chain—all of it. Picardat described the workflow for a typical job in metal fabrication: engineering and order processing, laser cutting, bending, hardware insertion, welding, grinding, polishing, surface treatment (like powder coating), inspection, packaging, and shipping.

The analysis should reveal how the entire value chain connects. Again, the aim is to increase part flow velocity. If an incredibly powerful laser produces stacks of cut blanks, that’s great, but what then? A denester stacks pieces onto pallets or carts, which then flow downstream (see Figures 1 and 2).

In the age of the fiber laser, part sorting is the true blanking constraint, but as Picardat explained, the implications go beyond the denesting station. Consider part presentation in the press brake department. An assembly might have two similar-looking workpieces, a left- and right-hand component, that part sorters mistakenly stack on the wrong carts with the wrong job travelers. Those carts move downstream, arrive in bending, at which point the brake operators scan the bar code on the traveler and pull up the wrong program. They bend the piece the exact opposite way it should be, send it downstream—and, at long last, in assembly, someone catches the error. Considering all the value already put into those parts, that’s an expensive mistake.

The example shows how everything connects and nothing stands in isolation. Sure, the ultrahigh-powered fiber laser sliced through the material in record time, but the accelerated cutting time left operators and part sorters alike scrambling to catch up. That’s where the mistakes happened.

Picardat took the example one step upstream, to nesting. What if specific parts require a certain grain direction, either for forming or cosmetic reasons? If the nesting software platform doesn’t take grain direction into account, the brake operator might struggle with a precision bend. Or, even worse, a customer might reject a cosmetically critical subassembly. All that value-adding cutting, bending, welding, and assembly goes out the window.

Such perspective shows the real value of automation. A lower-power laser coupled with parts-sorting automation, capable of stacking parts in the correct orientation every time, all within an established and predictable timeframe, might have a greater ROI than a high-power laser with double the cutting speed. Raw speed is just one piece of the puzzle.

Picardat described one hypothetical situation in which a shop with extraordinary blanking and bending capacity sped through a significant project and sent WIP to the assembly area—where it sat, waiting for purchased components. “Supply chain optimization needs to be part of the equation,” he said, adding that this applies to purchased component suppliers as well as service providers like platers, heat treaters, and custom powder coaters. Supplier relationships vary widely, but in some cases, “you might need backups upon backups to ensure supply chain reliability.

“Review your workflow,” Picardat continued. “Every process step that is needed to get a product to the customer is essential to understanding where productivity is improved. This starts with the engineering drawings through blanking, bending, welding, polishing, inspection, and shipping. When it comes to productivity, all of it comes into play. We’re not just talking about sheet metal processing. It’s everything.”

FIGURE 3. In this example, product design comprises only 5% of overall product cost, but the process influences other areas by 70%. This doesn’t minimize the need to reduce cost in other areas, but simply shows the benefits of examining all aspects of a product.

A change in cutting can make welding or assembly more repeatable, speeding the job’s velocity down its routing. But a change in design can eliminate an entire downstream process and reduce the number of parts the shop needs to manage. Design for manufacturability (DFM) can be the real game-changer.

Picardat pointed to an example that illustrates just how valuable DFM (and product design overall) really is. He broke down the “current state” product cost for a typical sheet metal subassembly: design (5%), material (50%), labor (15%), and overhead (30%). Changes in each area, he explained, “cast a shadow” on other processes. That is, an improvement in one area can benefit other areas too.

In the example Picardat gave, changes in overhead and labor influenced other areas by just 5%. Changes in material influenced other areas by 20%. But then there’s design, a process that takes up a small slice of product cost but has a massive influence—70% in Picardat’s example—on other processes. The part design casts the longest shadow by far (see Figure 3).

Picardat described an application in which a 16-part job was redesigned into a four-part subassembly. The effort reduced the number of parts the fabricator had to track as well as the routings it had to manage. The result: The shop shortened lead time by more than half.

He emphasized that this doesn’t mean fabricators should ignore improvements just because a design change isn’t negotiable. After all, savings in labor, material, and overhead still can make or break a major project. But the example does show the importance of communication and the benefits of considering all aspects of a product. For customers reshoring work or launching new programs with the intent of keeping more of the supply chain in North America, that’s exactly what’s needed.

Note also that Picardat didn’t say all improvement needs to happen in an engineering or managerial ivory tower. Quite the contrary. “There are ideas all over your factory,” he said. “The entire operation, from everyone in the office to those on the floor, make up a team that’s trying to produce parts in the best way possible. If you can get your design engineers to communicate with people on the floor, great things can happen.”

Scrutinizing workflow reveals the problems fabricators often can identify with simple observation. Parts aren’t presented consistently to a brake operator, who in turn misidentifies parts or forms them in the wrong direction. Material doesn’t fit up consistently in the welding cell.

Dig a little deeper, though, and direct observation becomes more difficult. A typical custom fabricator processes hundreds of different jobs over a week or even several shifts. Catching inefficiencies for every part simply isn’t practical. Here, Picardat said, is where the Industrial Internet of Things (IIoT) can play an important role.

“IIoT can monitor everything from programming through production,” Picardat said. “It can help you make data-driven decisions and can often find the bottlenecks before you do.”

IIoT reveals when the machines are actually making good parts and when they aren’t (see Figure 4). An operator might clock in on a job with ERP or similar software, yet the actual machine uptime (when the cutting head is cutting or the brake ram is cycling) isn’t actually tracked. That’s where IIoT can help reveal those numerous, barely noticeable events—an unexpectedly lengthy tool setup, a few minutes searching for parts or tools—that together add up to serious delays.

FIGURE 4. IIoT platforms track a machine’s setup and run-times for specific jobs, showing when the machine is making good parts.

With a clear picture of workflow, communication, and detailed utilization data, fabricators can start to realize the true value of operational improvement and new technology investment.

All this sets the stage for automation, with lasers monitoring the cut in real time, sensing cutting issues, replacing nozzles, setting up and centering, adapting the beam shape for the best pierce and optimal cutting. Part sorting automation presents parts in the same way, at the same speed, every time. This ensures press brake operators rarely if ever encounter surprises.

Offline bend simulation aids quoting and prevents machines from being idle for on-machine programming. Foot pedals slide across the bed to guide the unexperienced press brake operator through a complicated bending sequence. Press brakes now actually recognize the part being formed and its orientation, ensuring the correct part is in the press brake, positioned in the correct way. And on-machine sensing detects issues with material and adapts to ensure consistent bend angles.

Then comes automatic tool changes, shortening press brake changeovers from a half hour (or more) to just several minutes, making setup times nearly inconsequential (see Figure 5). Combine that with data from IIoT within the context of overall workflow, and the improvement opportunities abound.

It’s no longer about a single part on the laser, press brake, or welding cell. It’s about how everything fits together and optimizing the entire system. Can a small design change, like a sacrificial tab for gripping, allow for robotic bending? How are those parts best presented to the robot, then transported after forming? Brainstorming ways to optimize the system and bringing those ideas to fruition, Picardat said, will be where the next-generation operator will excel (see Figure 6).

Back at the FMA Annual Meeting, Picardat pointed to a slide with an autonomous mobile robot (AMR) carrying a sheet metal blank toward a robotic bending cell. “Everyone’s talking about the potential of AMRs,” he said. “But it’s not just the physical act of carrying blanks from one workstation to the next. It’s about the data. If you track the location, see where parts are and where parts are going, that data will be incredibly valuable. You’ll be able to see the part flow throughout the entire value stream. It’s not a pipe dream. It’s coming.”