Industry 4.0 uses a host of new technologies to optimize manufacturing efficiency. In this podcast, SME Media senior editor Steve Plumb talks with Per Treven, director of business development-manufacturing for Ericsson, about how advanced 5G networks and “limitless” connectivity enable next-generation systems and maximize ROI.
Introduction - montage
Aerospace and defence manufacturing, machines and automation, additive manufacturing, smart manufacturing, global manufacturing economy.
Speaker 1
Welcome to Advanced Manufacturing Now, the podcast for manufacturing professionals powered by SME. From the design screen to the shop floor to final assembly, we drive the conversation about making manufacturing smarter by connecting people who are passionate about manufacturing. Visit us at sme.org.
Speaker 2
Today's episode is sponsored by our guest.
Steve - Hello and welcome to Advanced Manufacturing Now, the podcast for manufacturing professionals. I'm your host, Steve Plumb, senior editor for SME Media. I am joined today by Per Treven, Director of Business Development Manufacturing for Ericsson. Welcome to the show Per and thanks for joining us.
Per - Thank you very much.
Steve - Well, our topic today is smart manufacturing and connectivity and some of the key challenges and opportunities. But before we get started, can you tell us a little bit about yourself Per?
Per - Yes, I'm in Ericsson's North American market area, and my focus is on business development for manufacturing. And the idea is to look at how we can enhance industry 4.0, manufacturing 4.0 with connectivity, which use cases are good for enabling with connectivity. So, my job is to identify that and help our customers and our partners to articulate that value to an end customer, a manufacturer.
Steve - Very well, well smart manufacturing is a hot topic, industry 4.0 as well, and connectivity with a lot of moving pieces, literally in the case of smart manufacturing. Per, can you tell us what some of the drivers are that you see that are accelerating the move to smart manufacturing and what some of those strategies are?
Per - Sure. If we look at manufacturing 4.0, it goes some time back maybe almost ten years ago, the key themes were brought up for manufacturing 4.0 such as autonomous robots, simulation, IoT, additive manufacturing, augmented reality and so forth. And what we are seeing now is that as manufacturers are trying to improve further, their operations, reduce cost and so forth, we see that these themes are now becoming more and more important.
Per - So, we see that there is a clear pull from the industry. And we also see that the technology has become better and better to be able to realize some of these themes. And we see also that some of the old improvement methods are not possible to use to address today's challenges. So, we're seeing that we're coming to a bit of a sort of ropes end in terms of the old methodologies and that is where the new manufacturing 4.0 themes they can bridge that gap and a manufacturer will be able to continue to improve.
Steve - Well, there's a lot of definitions out there about smart manufacturing. Won't get too deep in that. But how far do you think the process is in terms of implementing some of these technologies and strategies?
Per - I think we are still fairly much in the beginning. There are, of course, some that have come farther than others that I still think that we are a lot in the beginning. What I've seen now is that we have a number of manufacturers that are building new factories and that is where they are looking at what new technologies, what new techniques can we implement from the start rather than maybe trying to improve the old factory. But there is a little bit of a mix, but definitely we see for new factories that there is an additional interest in making sure to be at the forefront of technology.
Steve - Do you see more opportunities then, from your standpoint, to help companies when they're launching a brand new factory opposed to retrofitting an existing one and are there still opportunities for the older factories to enjoy some of the benefits?
Per - No, definitely I think that when we are looking at new factories, we are looking at Perhaps looking at things like, I want to reduce my cabling costs, so I want to install a 5G system from scratch. Maybe I can reduce some of the cost that I'm having to do if I have a regular implementation of my factory. There are also other considerations, maybe I am building my assembly line or my production line a bit different. I may want to increase my flexibility and I can do that by using a AMRs or AGVs. And to be able to run those efficiently, I need a really good network. So therefore, they go hand in hand to invest in that network technology, to be able to take the biggest advantage of the AGVs or AMRs and thus have a more flexible production.
Steve - Oh, whether it's smart manufacturing or traditional methods, the key things that most companies are looking at are reducing costs, improving quality, improving efficiencies, improving safety. What role does smart manufacturing and more specifically, some of Ericsson's technologies have in that environment?
Per - Yeah, it can play a very big role in that. And if we go back also looking at some of the challenges that we are seeing today in terms of worker shortage, supply chain interruptions, production availability uptime, sustainability. That is where we can play a big role as well. So, if you are looking at, for instance, worker shortage, right, that is where automation comes in, where we can look at automating processes.
Per - We have AMRs, AGVs, we have other ways of doing things. Then if we are looking at supply chain interruptions, that is where it goes a bit outside of the factory, but just to be able to track material in a more efficient way and understand how material is on its way when it's in the warehouse, when it's arriving and so forth, that that is where the connectivity technology can play a big role.
Per - And of course, we've really looking at production availability, additional sensors and better collection of data, and on the sustainability side, we are looking at monitoring energy usage and all the items that go into the factory in terms of water, energy, air quality and those types of things to reduce the carbon footprint.
Steve - It really all comes down to information and data, both having it in a timely fashion, real time about having quality information and being able to share it with other people in the organization. Ericsson, of course, is known for its cellular technology. What types of things are you working on to better enable smart manufacturing?
Per - So again, when we are talking about our products, we have our private network solution, that is a very compact solution for somebody to install in their factory with all data on premise, very secure. Your data stays in the system and does not go outside. And what we are seeing also is that with brownfield factories, for instance, also greenfield but brownfield, we are looking at when you are deploying another network, you can start putting your OT data on this new network.
Per - And in some instances, you don't really mix that data. You don't integrate with IT data until you collect that data into a data warehouse. So that is where you can do all your AI, machine learning to better understand how your processes are performing, how you can improve them, how you can automate, how you can change your settings, etc. So, by deploying a private network, it gives more flexibility and more security to the OT data.
Steve - And what should manufacturers consider when they're identifying and selecting smart factory use cases to deploy?
Per - So, when we're looking at the factory, we're looking at multiple areas and not only inside the factory as I mentioned before, when we're looking at supply chain. But we're also looking at warehousing, materials management, the connected workforce, environmental monitoring and so forth. When somebody starts to implement a private network and in our case it's called our version is called EP5G, it's important to think about what are the sort of some of the easy use cases that you can implement and as you implement some of these and then you gather your return on investment on the early ones, you can continue to build it out.
Per - But if we start in the in the beginning, I think you start with visibility, increase the information and knowledge about what is happening. As you have more information about what is happening, then you're looking into what is changing. You compare with what you had. Now you can see what is changing. Then leading into items such as what is the cost, what is the rule or the recipe, what will happen, what steps to take and how to optimize.
Per - And as you go along that chain, you will need additional data to be able to do that. So just getting visibility is something that is fairly easy to start with. And if we are looking at a couple of use cases that make sense, some of them are related to just being able to relay a problem in the factory.
Per - For instance, something is broken, or you need help with a particular item that you haven't experienced before, you don't have the expertise. So, one of the things that we have used in our factory is this halolens, and you can then look at an item, somebody else can also communicate with you effectively helping you to solve that problem, and that can help you to centralize just your experts.
Per - They don't have to fly in for something that can be solved, and they halolens is good, because you free up your hands, you don't need to hold something. That's an example of in the use case that is fairly easy to implement and another use case that it also is fairly easy to implement this with AMRs or AGVs to enable them to go on a private network.
Per - And the reason I'm saying that is because you don't need a whole lot of data collection per se to make those go more efficient. In our smart factory in Lewisville here in Dallas, we are seeing that the AMRs were performing so much better on a 5G private network than what they did on Wi-Fi.
Steve - Can you elaborate a little bit more on why it's better to be on a private network and what some of the benefits are?
Per - Yeah. So, for instance, going back to the AMRs, we are seeing that they are stopping less frequently. Now we have a much better coverage. A private network comes with all the functionality that you have in your macro network. You're using the same type of products that are being used by the big operators and scaled down, you have your 5G network in your factory just serving you.
Per - And what we are seeing is that the coverage is so much more effective. We look at, again going back to our factory, we can see that we cover that whole factory with two of our street macro radios. And that would require many more access points for Wi-Fi and what we are seeing also from that network is this improved latency, no disruption to where the AMRs that is running on this network.
Per - So, we can see that it's superior in that in that sense.
Steve - So better, more uptime, better efficiency, and just higher quality overall.
Per - Yeah. And what we are seeing too is for use cases that require mobility. That is where it really comes in as a real game changer. For machines that are just stationary, it may not make so much sense to enable them on a 5G network. Those could be cables if they are not going to move. But for the use cases that require mobility or where you just don't want to pull additional cabling, that is where the 5G private network comes very handy.
Steve - Well, speaking of 5G, everybody's familiar with the terminology and knows that 5G is better than 4G is better than 3G. But can you talk a little bit just how significant the move to 5G is and what types of things it enables?
Per - So, if we're looking at both 4G and 5G, they are both based on the same standard 3GPP. So as the standard evolves, new functionality comes into these different networks and technologies. But if we are looking at 5G, 5G was really designed for the industry. It wasn't so much designed for you to stream faster video on the street as a consumer.
Per - But it was targeting industry. And what we are seeing now is that new features such as time sensitive networks with defined latency, and we are seeing location services being implemented in the coming releases here in this 3GPP standard. We are seeing that those are going to be really important for the industry to be able to do some of the use cases that they are not capable today and with built in functionality in your network.
Per - So as the technology advances and the standard advances, there are new features that somebody can take advantage of to implement new use cases. And if we're looking at 5G as well or we're looking at these new technologies, we haven't identified yet all what we can do with it. So, in a sense, it's a bit like the Internet, that now you can start doing a number of things that you weren't capable of before, that we don't have the ability to come up with all those use cases and ideas ahead of time.
Per - There will be a lot of innovation as more and more manufacturers and industries start implementing these networks, they will be able to innovate and come up with many use cases that we are not able to think about today.
Steve - Are you already starting to see that with some of your initial customers, where they're implementing and discovering new things that they can do with the technology?
Per - Yeah. One of the things that I find interesting is, again in our own smart factory where we have implemented our own technology is where we are looking at the AMRs and we are looking at the initial installation. We are now seeing that there is an iterative process with the software developers for the AMRs and the AMR management system.
Per - Now you can take advantage of real time information to the AMRs, real time map date that can be downloaded. The AMRs can talk to each other, telling somebody that, hey, there is an obstacle in this route, take another route, and that means that you can improve the efficiency, you can reduce the number of AMRs that you need, and you can do things like faster docking and speeding up the whole process by allowing this new technology be iterative with the developers of the software for the application.
Per - That's an example of how we are seeing that the new technology gives them the ability to use it for improvements.
Steve - They're kind of acting like a real time traffic controller in the plan.
Per - Correct. It's like the cars would speak to each other on the on the road. A bit like, now you have things like Waze, but you need to upload information to some extent to be able to do the re-routing, that similar bit in an automatic fashion.
Steve - Now, I would imagine there's a safety component in there too, to prevent the AGVs from contacting people or other vehicles.
Per - Correct. If we are seeing now, if they drive by a person, they need to drive very slow. And the way that the initial software was developed also was to reduce the use of camera, it uses a lidar, but reduce the use of the camera to preserve some of the bandwidth. Now you can use that bandwidth that you have on the 5G network, which is much more abundant.
Per - And you can then identify, is this a person? Is this something that is sort of I really need to avoid or is it something that I can still go fairly fast around? So again, that shows how the better bandwidth can improve the routing and speed of the of the AMRs
Steve - Now there is another term called limitless connectivity that's helping to enable smart factory. Can you tell us a little bit more about what you mean by limitless? I mean, it sounds fairly straightforward, meaning no limit. But what does that mean in real terms?
Per - Yeah, limitless, maybe a little bit of an exaggeration. There is some limit to it. But what we are meaning with it is that you are not limited by the connectivity to think about your use cases. You have the bandwidth, you have the latency requirements fulfilled by your network. So, your constraint is not in the network, it is on your applications side.
Per - How you want to use it, then how you want to take advantage of it, but the connectivity is not your limitation.
Steve - So, the next question would be what types of use cases are employing level of connectivity are the most right to benefit from it?
Per - Of course.
Per - What we are seeing is a lot of data collection overall from sensors that may be monitoring vibration, or it may be monitoring some condition of the machines. Also, environmental monitoring. And what we're hearing from multiple customers is that this number of sensors will increase with 10 to 20 times because now you have the ability to collect much more data.
Per - You can implement these sensors in an easier way when you don't need to wire them and it's just going to explode. So that is one of the areas that we see a lot of need. The other one is on the video side and video inspection. Video inspection can be of goods we see. You can look at your bill of material making sure that everything has been received. You do that with video. And that, of course, uses a lot of the bandwidth that you have in your network. But those are some of the use cases where we see that we implement additional eyes, so to say with video and that can be analysed and it can be used for additional inspection of something. It can be an inspection on your production line, it can be inspection of the goods we received.
Per - We are also seeing implementations where you are looking at manual processes and you can follow those with video to make sure that all steps have been fulfilled. For instance, you have at somebody at the workstation needing to put items together. It may be a pick and pack type of process. And you are seeing that the person has been in each of the bins to make a complete package.
Per - That is one of the ways that you can use the technology.
Steve - And what type of feedback are you getting from customers, both in terms of the performance, but also on the human interface with these systems? And is there much upfront training that is needed?
Per - There is some training needed, if we are looking at it from a couple of perspectives, if we are looking at it from an IT perspective, it's often the IT department that will manage the private network. And what we have done is to make it really simple. The interface is very simple to use, somebody who is familiar in IT would feel familiar with our management system of the private network.
Per - So, we have done a lot to reduce it with telecom terms in the early versions. That was something that was making it a bit harder to adopt because you had to learn a number of telecom terms. Now somebody who's working in IT will be able to manage this system effectively.
Steve - Are customers are already starting to see benefits and is there an ROI associated with these applications.
Per - Yeah, definitely. Definitely. And just again, going back the example that I had before with the AMRs is where we can directly see a return on investment on the speed and the number of AMRs that we need to perform the activities. And some of the other findings we have is that the environmental monitoring is reducing a lot of the energy use.
Per - We have looked into some other use cases where we look augmented reality for remote support, we can see that it's decreases equipment downtime and reduction in vendor service cost, reduction in plant maintenance labour.
Steve - The energy savings has a direct result of being more efficient.
Per - Yes, being able to monitor everything and understand there may be items that you can turn off when you are not using them. Since you have this sort of a central management and understanding of these different variables that that you are measuring. Yeah, I had to one more example that we are looking at, which is very promising, which is what we called the digital thread.
Per - So, we are using that to be able to collect all parameters around our production line. So, for each of the radio base stations that we are producing in our factory, we can trace back all the variables around that production. So, if one is returned, we can understand if there was some problem at the production of that radio base station, and that saves us money in terms of repair and machine downtime.
Per - So just by monitoring and co-relating that with any type of problem reduces the amounts of problems that we are seeing with our products.
Steve - So, it really encompasses the entire manufacturing process from design and development as well of the facility itself and the tooling, the operation as well as the running, the repairs, the output, pretty much everything.
Per - Yeah, I would say so. And as I said before, the amount of use cases and the innovation, we will see that continue a lot. One other example, that we have is where you do drone cycle counting in your warehouse, you do it off hours. The drone will be able to go on the private network and autonomously during the night, it will be able to scan all your items in your warehouse, which means that you always have an updated understanding of what's in your warehouse and you can make better decisions in terms of your production by having that data available to you in a much more frequent way than regular.
Steve - The really 24x7 lights out capability to some extent, at least.
Per - Yeah, yeah, to some extent. We will maybe never go completely there, but we will still need humans to, to do some work that that's an application that is very promising and can improve your production in ways that you didn't think about before.
Steve - It touched upon this a little bit earlier. But how important is flexibility in terms of allowing different manufacturers or the same company to make changes on the fly and adapt?
Per - Yeah, if we are looking at manufacturing, then we are looking at the production line. What we have done in our factory, again, going back as an example is to reduce the length of the production line, to be able to increase our flexibility in production. Our production can be modified if there is a supply chain problem with a specific component, we can change to manufacture another radio base station than the one we had planned.
Per - And what we have done then is to by reducing the length of our production line, we need to increase the material handling and the increase in material handling we have done with the AMRs. So, they go hand in hand in improving your flexibility, you need more material handling that you don't need more people to do that. You can do that in an automated way.
Steve - Another thing that you mentioned earlier, and as a very important topic, not only in the business world but on the consumer world is cyber security. With so much more connectivity, how can you safeguard your data?
Per - Yeah, if we are looking at it from a networking perspective, there is a lot of built-in security already in the private network. Each device that connects to the network is provided its own SIM card, meaning that devices with only SIM cards associated with the network will be able to connect. In addition to that, you are able to section off different connectivity items, if we call them that with APNs, you can assign APNs to different areas of your factory or to different devices, meaning that you will limit the amount of devices that or you are in full control of the devices that are connecting to your network.
Per - So, from a networking perspective, there is a lot of built-in security when it comes to data security at rest, you have all the technologies at hand that you have in your IT environment and you will apply that when you have collected your data. But from a networking perspective, the private network is very secure, and you are in full control on the devices that of the devices that are connecting to your network.
Steve - I see. So, you talked about being kind of a learning phase and implementing in stages, starting with visibility. How far out do you see it going in the future as things like AI and virtual reality take machine learning become more common place?
Per - I think what we are going to see soon is that we are having the machines also being connected to the network the PLCs you will soon start having a closed loop type of functionality in your network. And I think that the 5G private network will enable that. When it comes to AI, ML that is already in place in a number of factories.
Per - And the way that they can use that is to reduce maintenance, understand how the process works and improve your automation and overall efficiency. Now we are seeing that AI, ML is easier to adopt, easier to use because of a number of new innovations in that field. Even if you are going to do some programming yourself, there are libraries that you can use that provide a number of the approaches and algorithms to adopt and apply to your data.
Steve - Sounds like the future is going there's a lot in store for the future in terms of tomorrow's factories and manufacturing technology. We have covered a lot of ground up here. Anything that we missed that you wanted to add?
Per - Yeah, the last thing maybe to add is that the premise is a lot about collecting data, and that is where we see a lot of the explosion happening, gathering data from a number of points that you are not able to gather data from before that your wireless network now opens up for and that data then you can use to improve in number of different processes and automate with. Private 5G is the answer to many of the problems that we have seen before.
Per - And by enhancing your connectivity, your throughput, your latency, there are endless opportunities to innovate and come up with new use cases to further improve and enhance your competitiveness as a manufacturer.
Steve - Great. Well, Per, thanks again for joining us. You've shared a lot of great information, very informative. It's an exciting time in the industry with the advent of smart manufacturing, and you certainly help shed some light on these key issues. I know for myself, I'm certainly smarter for having talked to you. We appreciate your time and for our listeners, if they want to learn more about Ericsson's efforts in this area, where should they go?
Per - Go to the ericsson.com/manufacturing website and Ericsson is E R I C S S O N. There you will find more information and of course you can look me up on LinkedIn and communicate with me there.
Steve - If people want to learn more about the latest manufacturing trends and hear more podcasts, please visit us at sme.org. Thanks again Per.
Per - Thank you for having me.