According to the National Institute of Standards and Technology (NIST), “smart manufacturing” is an umbrella term for “fully-integrated, collaborative manufacturing systems that respond in real time to meet changing demands and conditions in the factory, in the supply network, and in customer needs.” Already, this use of real-time data and technology to solve existing and future problems in manufacturing, while at the same time speeding up and streamlining business to create advantaged value, is being called the next Industrial Revolution. But what does “smart manufacturing” entail, precisely?

To find out more, Quality spoke with John Nesi, Vice President of Market Development at Rockwell Automation, and Bryce Barnes, Senior Manager of Cisco Systems’ Machine and Robot Segment globally under Cisco’s Internet of Things Manufacturing Solutions Group.


Bryce Barnes, Cisco Systems: Smart manufacturing, as embodied by 4.0 and Internet of Things, is really a digital transformation. If we look at the history of manufacturing, there are a few major transitions. The first was an energy transition that happened right before and during the turn of the century, where you had widely available power from wind or steam. And that absolutely sparked its own industrial revolution. Then you had mass assembly, as embodied by Henry Ford. And then you had automation, which really came into form in the 1970s and 1980s in a big way and continues to evolve with robotics. 

And now we’ve reached this transition point where almost everything starts to have a digital facet, a digital face to it, and that digital transition is important. When you talk to manufacturers and machine builders, they’ll say quite openly that with each new mechanical innovation today, the period of time that they get to benefit from that innovation is shorter. They’re getting shorter and shorter returns for each mechanical innovation. Their competition is catching up quicker.

The mechanical side of manufacturing is being commoditized in many ways. Everyone is reaching for a digital twin, a digital embodiment, of process. 


John Nesi, Rockwell Automation: There’s a couple of things in play here. One is that you do have an aging manufacturing infrastructure, and at the same time you have what I would call a confluence of technologies that are occurring that are coming up from the bottom of the sensor world. When you hear IoT, it’s kind of a deceptive phrase; but the way at look at it is that things that used to be static devices that had a job to perform on a piece of machinery are now actively transmitting their status and some metadata that goes with their situation. Now as you replace parts and pieces, you’re replacing with smart transmitters, with maybe some minor computational power; and if you wanted to be ambitious, you could feed that information into different scalable platforms that allow you do more analytic work on those sensors—and the data, the metadata, and the machinery that they sit on—so that you can drive higher levels of productivity. 

When we talk about the Connected Enterprise, it’s really about a game of competitiveness and productivity for manufacturers, and the need for them to upgrade their fixed assets. Now, some of those fixed assets have been in place for a long time, and nobody wants to touch them because they run, but they’re not necessarily competitive with newer plants that are going in to the global environment. So when you hear about smart manufacturing initiatives, a lot of the time what you’re hearing about is a public-private partnership of sorts that involves a coalition of vendors like Rockwell and Cisco, or government agencies and academia, to try to foster reinvestment in the manufacturing infrastructure. 

For example, in the United States you have the Advanced Manufacturing Partnership, which involves the Department of Defense and the Department of Energy along with academics and companies like Rockwell. In Germany, you have a phenomenon called Industry 4.0, which is the German government trying to export German technology for smarter manufacturing and revive their export economy for machinery. And that has a ripple effect in companies like China, Korea, France, and others. So they all have some level of these public-private partnerships to reinforce manufacturing on their shores—because as a result of years of global trade expansion, we’re seeing the ability of supply chains to be more localized again, and people want to manufacture close to their consumers. Where there are vital consumer markets, there will be manufacturing; but if you’re going to that, it needs to be modernized. Contemporary manufacturing technology needs to be in place to serve those customers who require more SKUs and flexible response to demand; and any manufacturer today has to be a lot more cognizant of the demand pressures, and also be efficient with supply-chain management. 

So when you hear about Internet of Things, you think about smarter things transmitting a lot of data. But it has to go somewhere; it has to be processed into something that’s contextualized, and it has to have value to the productivity equation of the manufacturer, or it’s a waste of time. And that where [Rockwell Automation]’s Connected Enterprise strategy allows you to converge information technology and operation technology to make better decisions faster for the benefit of the business and to ripple that upstream into your supply-chain management. 


Barnes: Manufacturing, if you survey it, is really a patchwork of communication technologies. Machine makers and automation venders want to take advantage of this, so typically they’ll deploy a purpose-filled communications infrastructure to support a particular type of machine or automation. What’s happening now is that everyone over the past five to seven years has converged on Ethernet, and that really establishes the groundwork for a digital transformation, once that is in place. And that’s been the heart of Cisco’s partnership with Rockwell, to not only integrate the network into automation so that everyone is part of the same system, but also to allow the operators to think of the network as part of their control platform, or PLC. So from a PLC perspective, much of the innovative work that Cisco has done with Rockwell has been to make the network a part of their IO tree—so that you can be able to manipulate, visualize, and control the network right through your factory’s automation platform. 


Nesi: If you look back at the last 20 or 30 years, manufacturing has been looked upon as a labor intensive activity; and at the same time, the world has started to globalize. And so there’s been a move to markets where labor is cheap and the middle class is growing. So a lot of influx went to China, and now it’s moving to Vietnam, the Philippines, Indonesia, Thailand, places like that, where there’s more consumer growth. Labor rates start to grow, and at the same time, you can’t equalize the cost of shipping goods back to the United States; if you make them in China, you’re going to give away all your labor benefits in terms of just shipping costs. You can look back to when oil was $100 a barrel, and that pretty much diminished any labor advantage you had in China. 

Things are a little bit different now, but at the end of the day, it makes much more sense, logistically, to handle the logistics of serving customers with localized manufacturing. And what we’re seeing now is a lot of major companies that serve, say, the consumer products area looking for plants with smaller footprints that can be put into population centers where they can respond pretty quickly, and that requires technology that’s very responsive in terms of being able to receive an order, process an order, get the suppliers to ship on time and spin that around so consumers get what they want when they want it. And when you stretch your supply chain out, while it’s good for global trade and it’s good for labor arbitrage, we’re kind of past that point. Now we’re at a point where it makes sense to shrink your footprint and shrink your service area to serve as locally as you possibly can. 

And to do all of that, you’ve got to have more flexible mechanicals as well as more flexible electrical and information systems. In the older factories in the United States, you can’t just suddenly make a switch and say, “You know what, I want to start manufacturing in the United States all over again,” because if the plant is 50 years old and the gear has been there forever, it won’t be able to keep up with the instantaneous changes and demands from customers. So the logical course would be to start putting in information systems and automation systems that allow you to handle a more complex mix of products and a more rapid response with the products.


Barnes: Security is one of the fundamental questions that has to be addressed before you can have an open and interconnected network fabric. Think of networking as a utility—as a platform on which all of these industrial applications, hardware and software, can run and interconnect with each other—and the manufacturing plant itself becomes something that you optimize holistically. But in order to do that, you can’t just have openness and move toward openness; you have to have a secure connected environment.

Historically, plants have just isolated themselves. In fact, the extreme version of that is machines being isolated within the plant; they are part of a small cell, and very often, that cell is not interconnected. So in order to move forward and embrace this digital revolution with outcomes of quality by design, outcomes of predictive maintenance, and advanced streaming analytics at the edge that can detect conditions and possibly even enable new modes of quality control, you have to secure the environment.

The best example of this is the work that Cisco Systems has done with FANUC Robotics. In this case, FANUC is offering something that they call zero-downtime services. And zero-downtime services is a way to continuously monitor the mechanical systems within the robot and be able to predict failures. Now in order to do that, you have to connect the robot to an application with analytics capabilities that can intelligently analyze behavior—torque forces, force curves, acceleration curves that are generated every second that the robot is operating—and then begin to see minute changes in those patterns. 

In order to do this, you have to secure not only the robot but the communications channel from the robots wherever the analytics application is. In this case, it’s in the Cloud. And Cisco’s opportunity in this space is to provide that secure communication and connectivity to give identity to the Cloud, to the applications in the Cloud; to give identity to the robot itself; and to create a platform approach for security that’s unique. It’s an integrated technology view that will allow the security to occur and enable [a company like] General Motors to adopt predictive maintenance as a Cloud app.

Another way to look at it is that security is actually the business opportunity. If you can create a way to connect on a very secure foundation, it opens the door to new service models, new service models and new partnership models.

For example, the starting point that FANUC chose is predictive maintenance. And the predictive maintenance is not just getting analytics; it’s actually taking FANUC’s domain knowledge and applying that to solve problems in the factory on behalf of GM. But all of these machine makers and automation vendors, Rockwell included, can now—with this new security framework, this Internet of Things connectivity model—actually participate in the business of manufacturing things and help provide service level agreements around that. So, I think what you’re going to see, once this IoT framework is established as a normal way of operating, is a completely new partnership model in manufacturing.


Barnes: We want to help the world digitize the machine base. To give you an idea of the scale and scope, there’s approximately 55 million machines that are actively producing all of the things that make up our lives today; and probably every year, two to five million more machines are produced. And the question is, with tens of thousands of machine builders, how do you solve the problem of normalizing them and getting data off of them in a standards-based way? Because they all have different standards, and they all have different control systems. So we’re in a great position to normalize, connect, and secure these machines, as well as to partner with some of the leaders in the machine space as they embrace this digital revolution. The manufacturing industrial world wants to make the digital leap, and you can see the leaders emerging already. Q