Manufacturers facing labor shortages and supply chain challenges are increasingly investing in smart factory technology to remain competitive. The good news: software providers have made crossing the digital divide easier with tools that are more intuitive, compatible with legacy equipment, and bring a quick return on investment.
“More and more people are saying, ‘I really have to automate and I need to learn about the technology,” said Ed Cuoco, VP of strategy and solutions at PTC. “If you don’t start to learn about this technology and start to apply it to your operations, you can be sure you’re not going to get any better. When people apply this technology correctly, the results are unbelievable. I’ve seen half a million-dollar investments pay off in six months.”
“Most of our customers are small to medium manufacturers,” said Jim Finnerty, product manager of Wintriss Control Group’s ShopFloorConnect OEE and Data Collection software. “The capital outlay is relatively modest and payback is very quick. One customer paid off the whole system in six weeks.”
Smart factory technology finally is becoming viable for small- and medium-sized manufacturers. In some cases, manufacturers are focusing on incremental improvements instead of complete overhauls that require extensive replacements and investment.
Blame or thank the pandemic for creating new problems, shining a light on ongoing problems, and pushing manufacturers to seek solutions.
“Over the last 24 months, we have seen unprecedented supply chain challenges exacerbated by all things connected to the pandemic,” said Paul Wellener, vice chairman of U.S. Industrial Business at Deloitte.
“[Lack of] labor is the other big thing,” Wellener said. “There was a significant drop in manufacturing jobs at the beginning of the pandemic. Now growth is back to nearly where we were at the start of the pandemic but we’re still several hundred thousand jobs short. There’s a greater demand for manufacturing jobs. There’s lots of competition for entry level jobs, which is slowing down the pace of change. Manufacturers are really struggling to get the right talent into their organizations.”
“The demand for well trained operators in manufacturing has continued to grow and grow and grow,” Cuoco said. “This has nothing to do with Covid. There’s just not enough labor for manufacturing.”
“We now see companies investing in the digital spaces in a greater degree than in the three to five previous years, especially since the summer of 2020,” Wellener said. “We’re seeing more investments to improve productivity. The pandemic is definitely an accelerator.”
For example, the pandemic increased the use of Wintriss Controls Group’s more advanced features on its ShopFloorConnect software, notably a feature that enables remote monitoring and programming of network controllers, Finnerty said.
“This was one of our features that whenever we talked about it with a customer, they always liked the idea of it and said, ‘This is something we want to take advantage of eventually,” Finnerty said. “Covid knocked these things off the back burner.”
Although some small and medium sized companies may not have the ability to invest in digital factory tools, eventually the initiatives will make their way downstream, Wellener said.
“Just as LEAN production systems 30 or 35 years ago started at OEMs, then went to first- and second-tier suppliers, we’re going to see digital investments follow the same path with larger companies first, then trickling down to medium and small companies,” Wellener said. “It won’t happen overnight.”
Based on Deloitte’s research, early adopters of digital factory technology are continuing to invest in smart initiatives while manufacturers who experimented a few years ago with one factory, one assembly line and/or one pilot program are now beefing up their investment, Wellener said.
One Wintriss Controls Group customer in Georgia was typical: the manufacturer started by connecting 20 machines to the system and eventually increased to 60, Finnerty said.
“There is a lot of history of creating islands of automation in organizations,” Wellener said. “There are always going to be some islands of automation because people are going to pick use cases where to make a surgical investment to apply smart manufacturing.”
Manufacturers are now building bridges among these islands of automation, integrating smart factory technology across entire enterprises.
“More and more companies are breaking down these islands of automation, connecting things, and creating an entire value chain,” Wellener said. “Smart factory tools and software packages are helping to connect those islands of automation.”
One key differentiator: some software companies are focusing on tools that don’t require manufacturers to replace expensive legacy equipment.
“We’re seeing startup tech companies as well as creative process engineers figuring out ways to extract analog data and turn it into digital value,” Wellener said. “We did one project where we used video to read factory gauges, took a picture of the numbers, and then used a character recognition system to convert the numbers and plug them into an enterprise system.”
Instead of trying to figure out a way to program a manufacturer’s existing controller to send the necessary information to its the software, an often-difficult task because every machine is different, Wintriss Controls Group opted for a simpler solution, Finnerty said. “We install our machine interface box that monitors uptime and downtime,” Finnerty said. “All we need is an electrical signal to tell us when the machine is running. When a machine stops, we know it.”
Before installing ShopFloorConnect, many manufacturers weren’t completely aware of production rates on each of their machines, Finnerty said. In some cases, manufacturers wouldn’t update and raise standards when they improved machines or wouldn’t lower standards when machines experienced problems. ShopFloorConnect provides the necessary information to revise standards and processes, he said.
“We can tell them how they’re running, compared to how they’re normally running,” Finnerty said. “We can do job scheduling. We are well aware of the current rate of production. We can tell our customers with a great deal of accuracy when a job is going to be complete. We also have an historical record of how the next job runs in the machine and when that job will end. Users can also plug in a deadline. In many cases, they may be losing money when making certain parts. We can track that.”
But solving a labor shortage and/or achieving a higher state of smart is not simply a matter of throwing money at a problem or having a bunch of robots delivered to a factory. Rushing too quickly to buy and install technology can lead to rusting robots and senseless sensors.
“In many cases, a labor shortage is what starts the discussion,” Cuoco said. “But they don’t make the investment in their time to understand the solution. A manufacturer says, ‘I can sell more widgets if I can build more widgets but I can’t find more people to build the widgets. Maybe I’ll get a robot.’ Because many people don’t have experiences with robots, [they] think that solves it. They have no idea what makes a successful robot cell. I know of a major organization that bought 50 collaborative robots. A VP sent emails to 15 plant managers and said, ‘They’re coming. Put them to work.’ But it’s not a matter of getting an industrial robot and setting it on the floor. In defense of these people, it’s just desperation.”
Clearly, manufacturers need to know the problem they’re trying to solve and their budget for solving that problem. They also must get a handle on the basic need, the entire process—not just where the problem lies—as well as the minor but critical decisions a human operator makes without even thinking but that a robot must be programmed to do, Cuoco said.
Another key lesson: factories don’t have to be 100 percent smart. Even becoming a little smarter can make a big difference, often with little investment required.
“Companies are looking for downtime reduction” Finnerty said. “Regarding possible downtime reasons, there’s a lot of low hanging fruit in every factory we go into. A lot require zero investment to fix.”
“Sometimes it’s good old-fashioned industrial engineering,” Cuoco said. “If you can solve the problems the operators are having today, you can get a good return and it may not cost you anything to fix those problems.”
For example, at one factory in Texas that makes seat belt parts, a manufacturer discovered that one machine operator had significantly lower setup times than his coworkers, Finnerty said. “They were measuring setup times, but they weren’t measuring correctly,” he said. “The missing piece was who was doing the setup. By reviewing the SFC data, a need was identified. They told him, ‘Train the other operators.’”
At another factory, each machine operator typically ran four machines, Finnerty said. Using the software, the manufacturer discovered that some jobs were more complicated, requiring more intervention from operators. With that knowledge, the manufacturer reduced the number of machines an operator was responsible for when overseeing the complicated jobs and giving other operators more machines, Finnerty said.
Although downtime is certainly to be avoided, it’s possible to create additional problems by seeking to avoid downtime at all costs. A better plan is to predict and manage downtime, Finnerty said.
“That’s one of the problems with downtime,” Finnerty said. “Everyone remembers the catastrophes. People remember, ‘The drive burned out last year. It took us three weeks to get the part and customers were yelling at us.’”
But downtime, if planned for, isn’t necessarily the only driving factor. At one manufacturing plant, the head maintenance foreman dictated the speed of operations—to a slow pace—with the goal of avoiding maintenance downtime, Finnerty said.
“If they ran a feeding device too quickly, it would break down and have to be overhauled,” Finnerty said. “He put a speed limit on the device. When they put our software on the device, they found they could afford to overhaul this equipment four times a year because it could run three times faster.”
As it turned out, the device needed to be overhauled only every 18 months—not four times a year—and that maintenance could be scheduled in advance during planned shutdowns or holidays, Finnerty said.
“At least it allows you to make a decision: this is how many more parts we can make if we run the machine faster,” Finnerty said.
Further, manufacturers can decide how long of a stoppage matters, for example five minutes, and program ShopFloorConnect so that the machine operator restarting the machine is required to select a reason why the machine was down, Finnerty said.
Another manufacturer identified short but frequent and costly periods of downtime caused not by equipment breakdowns but by waiting for operators and quality control specialists to sign off on job changes, Finnerty said. The manufacturer’s high-mix, low-volume production requires setup changes on each machine up to eight times a day, with each operator running several machines.
For each change, the operator must find the quality control person to verify the new setup, then the quality control person must find the operator, who then can start the next job, Finnerty said. That process took 20 minutes for each changeover. “They had to find them,” Finnerty said. “They had to walk around the plant until they saw whoever it was.”
With more than 20 machines and eight changes a day, that added up to more than five wasted hours a day.
ShopFloorConnect now texts quality control and machine operators as soon as they’re needed, then escalates with a text to the boss if there’s no response in 10 minutes, Finnerty said. The average wait time at machine changeovers has shrunk from 20 to five minutes.
Since the system is easy and runs the same on different machines, training and use also is easy, Finnerty said. Across the board, tools that are easy to use are increasingly in play, making smart technology more accessible.
“An inhouse electrician typically spends about half a day connecting each machine,” he said. “The software is installed remotely in just a couple of hours. The machine operators need just a couple of hours to start using ShopFloorConnect.”
“A lot of times, our customer will say, ‘We want training,’” Finnerty said. “We will do the installation, set it up remotely, and get it running. Then we will intentionally schedule the training for three weeks out. At that point, the customer says, ‘We’ve got it. We don’t need training.’”
Connect With Us