LabsCubed develops ‘lab of the future' with Industry 4.0 testing

AKRON—What makes a lab Industry 4.0? It might not be what you think.

"Most people think the introduction of robots and computers is Industry 4.0," said Khaled Boqaileh, co-founder and CEO of LabsCubed Inc., a manufacturer of tensile, tear and flex testing equipment in the rubber and plastics industries.

In his presentation, "The Automated Lab of the Future," at the International Tire Exhibition & Conference, held Sept. 13-15 in Akron, Boqaileh said the world has experienced four industrial revolutions.

Implementing robots and computers in the lab or on the manufacturing floor is merely the third.

"Industry 4.0 is actually making those systems smart," he said. "Smart meaning they can make predictions, they can change things on their own."

Without human intervention.

"There's some really exciting work that's happening out there around fully automated labs," he said.

And the work happening at LabsCubed is helping those labs.

Boqaileh co-founded LabsCubed alongside Chief Operating Officer Jeffrey Petracca in 2016. He said the Kitchener, Ontario-based company started with the questions, "Can we improve this process? Can we help to make materials faster?"

"(I have) a master's in material development. I did my undergrad in mechanical engineering, so I didn't really get into the materials science side of things until I did my master's," Boqaileh said.

"I didn't realize how arduous it was, how long it takes when you're sitting in a lab, doing development work, testing work, how much of that is very manual."

Boqaileh said LabsCubed met with over 100 professionals in the rubber lab and testing industry and found that the industry could benefit from faster sample testing methods.

After six prototype iterations and over 10,000 samples tested, LabsCubed developed the CubeOne in 2019, "a high-accuracy and high-consistency tensile testing machine," according to the company's website.

The CubeOne, Boqaileh said, is fully automated and handles ASTM and ISO standard samples.

The CubeOne can test up to 36 samples per hour with less than five minutes of setup, offers increased consistency and can automatically upload test data to any device via the cloud with video backup of each test to analyze elongation.

"It uses vision systems, it uses a whole bunch of very intricate, high-end systems inside to measure everything as accurately as possible," Boqaileh said.

When preparing for the "labs of the future," Boqaileh says there are two key points to hit—increasing automation and improving data collection.

And these two points make up the first two stages of developing an Industry 4.0 lab, where the third and final stage is applying automation and data to artificial intelligence systems.

With automation, Boqaileh said, labs can produce cleaner and more consistent data at a much higher rate.

Referencing consistency and accuracy tests LabsCubed performed with its CubeOne with over 1,000 samples—both manually and via automated systems—Boqaileh said the automated systems are about 40 percent more consistent than performing tests manually and save about 85 percent of technicians' time.

"It's not because people are bad or are not doing their jobs," he said. "It's just automation can do the same thing over and over and over really well. It's kind of just what it's made to do."

Boqaileh told ITEC attendees to consider one key question when it comes to automation in their labs or factories.

"Of any of the tasks that your employees currently do, are they repetitive, bottlenecking or prone to human error?" he asked. "If you're answering yes to those, then you have something that you really should think about automating."

As for data collection, Boqaileh said, companies need to find better ways to collect and store valuable—and often critical—data.

This is important when dealing with A.I., as the data output is dependent on its input, he said.

"We still see a lot of people with papers and pencils," Boqaileh said.

"The metadata, for example, as we call it—which is the data that underlies things, where your chemists are writing very important pieces and points down on a piece of paper that's not collected anywhere else, but makes a huge difference when it comes to predictions and needing to understand what just happened—that all needs to be collected in a much better format."

He said A.I. systems, while critical components of an Industry 4.0 lab, are merely prediction systems. And while some are more complex than others, at the end of the day, poor or incomplete data put in leads to poor or incomplete data put out.

"(It's a) trash in, trash out kind of thing," he said.

"This is why the cloud is so important. Organized, accurate, traceable and in the same format," he added. "Centralize (your data), put it in the cloud."

Boqaileh said the CubeOne can predict the hardness of samples within 1 percent accuracy, even though it doesn't measure hardness, because of this centralized data.

"We don't measure hardness, there's no durometer inside the machine," he said. "But we can do it because there's so many different data streams coming off from the gripping systems, from the vision systems, that we can do some cool stuff and have that prediction. And that's just because we have the large treasure trove of data from customers running all the time."

By implementing automation and improving data collection, Boqaileh said, labs could be well on their way to implementing efficient A.I. systems to achieve Industry 4.0 standards.

The CubeOne, he said, could be part of this for some labs.

"What we've built is simply a piece of the lab of the future," he said.

"Starting with one system—it doesn't have to be our system—find something that you think automation will play a big part in within your lab. Plug it into the cloud, start plugging more of your systems in, start finding providers, find services, find people that can help you through that process," he said.

"Then you get to that final stage … where you can then start to use A.I. to help you with predictions and speeding the (testing) process up," he added. "If you start putting those pieces together, you start to get toward a really smart, amazing lab."