BIG M 2014: Big Data Looms Large Over Manufacturing
By Patrick Waurzyniak
Big Data seems to be everywhere these days. The vast amounts of data compiled by manufacturing operations—and how to collect, analyze, leverage, and secure this invaluable data—pose a dilemma for manufacturers of all stripes. Manufacturing executives and users discussed the latest software solutions used to grapple with this vexing problem in “The New Digital Space: Future of Manufacturing Intelligence” panel discussion at The Big M conference at Cobo Center in Detroit.
“Manufacturers face immediate challenges from global competition, collaboration and profitable growth,” said Dennis Bray, interim CEO of SME. “How can product reliability performance be improved while increasing productivity and profit? Turning mountains of data into business intelligence can be the key to increasing efficiencies and exceeding customer expectations.
“We want to learn how leading companies large and small are working together, understand that big data doesn’t have to mean supercomputers, and how the future will be changed on how to predict it.”
In this new digital space, manufacturing operations are attempting to capture and better leverage shop-floor metrics like Overall Equipment Effectiveness (OEE) and other key performance indicators to take advantage of the wealth of digital data coming off the factory floor. “We’re trying to take from that space data that in the past may have been time-sensitive and paper-based,” noted Jay Monahan, director, SAP Shop Floor–Global Lead.
Taking a cyber-physical approach to factory floor intelligence is key, added Franz Gruber, CEO of Forcam (Friedrichshafen, Germany). “This is a prerequisite for world-class manufacturing,” he stated.
At every turn, software seems to be the hub for dealing with massive data, with new software solutions that work in concert with machine controllers on the shop floor to capture, route, and analyze great volumes of data. Roughly 10 years ago, management at Siemens surveyed what was most needed in manufacturing, said Helmuth Ludwig, CEO of Siemens Industry Sector USA. “The next best value is combining manufacturing with design,” Helmuth said, noting that Siemens invested $3.5 billion when in acquired the former UGS (now Siemens PLM Software).”
“If you think about what’s driving US innovation, it’s all software-driven,” Ludwig stated, pointing to 79% growth in Siemens’ software business. “There’s something incredibly deep here in the US about software.”
Better use of shop-floor data in real time is beginning to happen, noted Josh Davids, CEO of Scytec (Greenwood Village, CO), a developer of the cloud-based DataXchange shop-floor collection and monitoring system. “There’s a trend toward getting more data in real time. You have to access it quickly to be able to leverage it,” he said. Many recent advances are due to the cloud-based systems and to companies embracing factory-floor data-exchange protocols like MTConnect, Davids added. “It was kind of stagnant for about 10–15 years,” he added.
Intelligent sensors also play into the equation, beefing up sensor on manufacturing systems that in the past perhaps didn’t have enough reliability. “I’ve seen the robustness of sensors that now have gotten to where you can rely on them,” noted Brian Dods, executive, Manufacturing Technology for the Global Supply Chain, GE Power and Water (Greenville, SC). “It’s to where you can model your manufacturing on them.”
New applications for this digital space hold great potential for building better, more reliable products. “There’s a lot of creative technologies out there that you can beg, borrow, or steal for improving your business,” Monahan said. Manufacturers stand to greatly reduce the amount of power consumption for manufacturing with the software available to intelligently power down workstations, saving as much as 24% in power usage, added Ludwig. And the latest simulation technologies allow testing different prototypes in product development many times before finalizing designs, Ludwig said, so you can improve them.
Enabling the Smart Factory
In another session at the show, “Smart Factory: Achieving 4.0,” Mohamed Abuali, chief operating officer, Forcam USA Inc. (Cincinnati), demonstrated that the use of intelligent shop-floor management software tools can effectively raise factory productivity by as much as 25% within a year or less. The Factory Framework is a manufacturing execution system (MES) from Forcam that offers users a way to leverage OEE metrics and other KPIs that show real-time status of machine tools on the factory floor.
To deal with the flood of data, manufacturing systems have to be capable of handling large volumes of data in a real-time capacity. For example, Abuali gave the cases of two auto suppliers, a Tier supplier and a stamping plant. The Tier supplier had to deal with high-velocity data from nearly 300,000 events per hour, and 3.26GB of data; at the stamping plant, the volume was far greater, at nearly 1.5 million events per hour and a daily intake of 16.3GB of data.
“Many of the unmet needs of manufacturing fall under Big Data,” said Abuali. “Whatever system you use, it needs to have data compression.” At IMTS, Forcam’s cyber-physical system will feature a new add-on that will enhance the systems with predictive manufacturing capabilities, he said. “It’s how you can capture the information at the machine control, via MTConnect or OPC,” Abuali said. “You will be able to visualize the data in real time, not only looking at OEE but other aspects. It’s a complete simulator.”
Key to the system’s handling large amounts of data in real time is having a database that can quickly crunch numbers, Abuali said. The Forcam system has employed technology from Wall Street, he said, that includes in-memory complex event processing (CEP) that consumes gigantic volumes of data and processes that in real time.
“A Smart Factory is a combination of cyber-physical systems,” Abuali said. “You’re mirroring the physical production world in real time, and in the cyber world.
Securing Factory Data
While Big Data is on everyone’s mind, nobody has forgotten that there still are substantial worries about the security of manufacturing data in the cyber world. In a Big M panel, “Cybersecurity Challenges to Industrial Networks,” moderated by Katherine Voss, president and executive director, ODVA Inc. (Ann Arbor, MI), security experts, manufacturing companies and users debated the issues posed by hackers bent on wreaking cyber havoc against manufacturing networks. “The adoption of Industrial Ethernet is exploding,” said Voss, “and nowhere is it more so than in manufacturing.”
Security policies are an issue, noted Bruce Billedeaux, senior cybersecurity consultant, Maverick Technologies (Columbia, IL). “When I go into a plant, very rarely am I asked what tools I’m bringing in,” Billedeaux said. “With Stuxnet, it was a third party that brought it into that plant. You need to ask ‘Who are your trusted users?’.”
Of concern is the variance between the expectations of IT and the operations technology (OT) personnel, noted Rachel Conrad, business manager, Rockwell Automation (Milwaukee). “This is hands-down the No. 1 concern,” she said, “and how to you respond to a breach?”
With cybersecurity, there are always catalyst events that seem to kick things into gear, noted Jeffrey Smith, Controls Architecture & ICS Security, American Axle (Detroit). “I always say ‘We don’t know what we don’t know’,” Smith observed. A catalyst could be something like the reports that the F-35 fighter jet has the ability to inject malware wirelessly while flying over a plant, he noted. “What if they decide to test it over my plant?”
Whatever steps are taken to guard industrial networks, security and plant professionals have to be able to live with the risks, Smith added. “The worst case for me is, I’ve got a customer really ticked off about nine miles from the nearest dealership,” he said. ME
Published Date : 6/12/2014