Machine tool heavyweight, MAZAK has launched a partnership with Big Data integrator Cisco Systems to change the way factories make things.
MAZAK recently introduced a data collection and analysis system called MAZAK SmartBox, which uses MTConnect along with software and a hardware adapter from MEMEX and hardware from Cisco to connect machine tools securely and intelligently.
The novel system addresses the resistance to true factory automation among many machine tool operations: Fear of excessive cost, complexity and security.
Currently in use at MAZAK’s Kentucky manufacturing plant, the system works independently of the machine’s electrical system, can be retrofitted to older legacy equipment and can use existing off-the-shelf sensors.
“The next Industrial Revolution is digital manufacturing,” says Bryce Barnes, manufacturing IoT senior manager with Cisco. “What does it mean when you connect everything together? It’s about the value of moving information. Think about what a smartphone does. No one predicted a device that can access almost all of the sum of human information. The total number of machines in the world is enormous, about 65 million. Most are operated and managed locally. Moving data from those machines is a challenge.”
The benefits of integration are many, but the bottom line is literally about cost reduction: “Ford plans to reduce their energy consumption by 25 percent, but they can only do that with connected, optimized machines, adds Barnes.”
The numbers reported by Barnes seem difficult to believe. Unplanned downtime can be reduced from 10-11 percent to as little as five percent.
Defect rates lowered by almost 50 percent, inventory turns up 35 and new product introduction cycles shortened by over 20 percent.
Are machine tool users leaving that much productivity on the table? According to Barnes’ data, yes, but not because human-based systems aren’t optimized. Rather it’s because there are limits to what humans can do to control production processes.
Cisco defines IoT as real time awareness, such as ultrasonic monitoring of machine vibration that operates beyond human perception, but then analysing the data in ways more complex than possible in human systems.
But the gap between what people can do and what the machines can do is growing. Can manufacturers fully use the capability of new machine tools?
The gap between operator knowledge and the techniques needed to take advantage of hybrid machine tools, new tool geometries and cutting path algorithms is widening, quickly.
“The ability to take control of the data offline lets you use the information (to optimize processes) in a new way,” says Bryce. “You can go to bed as a manufacturing company and wake up as a software company.” It’s evolving in ways similar to the Ethernet/computer connectivity wars of 20 years ago that led to massive connectivity in computer systems.
The technology has been in place for decades, but engineering conservatism, security issues and the lack of industry standards has been an effective brake on data-driven process efficiency.
MTConnect has addressed the standards issue, but Big Data analytics are still constrained by several factors:
- Insufficient IT resources in-plant;
- Operation don’t understand IT;
- Lines/plants operate independently;
- IT is Ethernet based, while machine use Ethernet, Profibus, Fieldbus, etc.
IoT promises to be the unifying force. “It’s where the two worlds come together,” says Bryce. “If you have a factory you have to be honest: Do you really understand your production process?”
Many factories implemented data acquisition decades ago, but with proprietary systems that didn’t talk to each other, the ability to process the large amounts of data that pour from real time manufacturing processes was severely limited.
“We need investment that are not one-offs,” Bryce adds. “It’s important to decouple data processing from results consumption.”
Current systems require users to dive deeply into the information processing analytics, instead of focussing on the end goal: results that drive corrective action.
The new connectivity matches data patterns, analyses, histories and processes the information into forms useful to the production manager directly, replacing the IT middleman with a simpler, faster and unequivocal results delivery system.
But what does knowledge really mean in manufacturing?
“I’ve seen a 200 percent variability in machine productivity in manufacturing operations, with no way to understand why this is happening,” said Bryce. This is common among smaller manufacturers, who lack the in-house resources to develop custom solutions that carry an ROI that’s minimal over short production volumes.
Off-the–shelf solutions like SmartBox and MTConnect pull the cost down into the range where analytics are now available to thousands of lower tier part makers in both the OEM supply chain and the stand alone manufacturing community.
One key is to build associations between parts and processes. It’s a new way of thinking about machined parts; currently, the machine determines the part making strategy.
In the new paradigm, the part carries with it its associated cutting tool paths and control points, which are then matched to a manufacturing process. It’s as if the part “shops” for a machine tool and production process that’s best suited to making it, rather than the other way around.
The potential isn’t restricted to faster, cheaper, better parts. The decentralization means that machine builders will have access to the same information as their customers, with one major difference: the machine tool maker’s ability to see more data from more machines.
If Mazak or another machine tool manufacturer can see that a hydraulic cylinder manufacturer is consistently lower in machining productivity than their competitors, how much would they pay to find out why?
The potential for machine manufacturers to become production consultants is obvious, especially as the ability to gather machine performance data without compromising customer data security eliminates the understandable reluctance of part makers to let machine makers access their equipment in real time.
And then there’s the next logical step: will machine builders simple take over the manufacturing process themselves and change manufacturing to a service industry? Maybe, but the same advantages that IoT offers will likely democratize the data analytics function, making small players as efficient as the heavyweights.
Smaller owner-operated part makers are a stubborn and strongly independent community. Whether they can stay afloat in the tsunami of Net-connected data may depend on their willingness to connect, and ultimately share their knowledge.
For more information, visit www.mazakusa.com.