Podcast #4: Making Industrial IoT a Reality

Hello everybody this is Ed Maguire, Insights Partner at Momenta Partners, and today we have with us Ralf Schultz who is an Industry 4.0 and IoT expert. We’re going to discuss a bunch of different topics around Connecting Industry, some used cases of emerging technologies, and some of the potential out there. I think what Ralf brings to the table is a real knowledge of where the rubber hits the road, when it comes to Connected Industry.  

It’s a pleasure to have you join us Ralf. Can you provide a bit of background on your experience, and what brought you to work in Connected Industry? 

Thank you very much Ed for the opportunity to speak. What brought me to the Connected Industry was basically the customer, and I guess that’s the best thing to start with. I spent a couple of years in consulting, and have been consulting a lot of customers on mostly supply chain and after-sales service topics, and doing that I had my first encounters with machine to machine communication tasks back in 2011, that was mostly due not to the fact that there was specific technology, but mostly there was a specific customer need and some pain points of customers which brought us to the attention of machine to machine, or as it is called to day IoT, can IoT help us with doing that? 

Above and beyond I’ve spent a good portion of my career working in Auto ID, and especially RFID and stuff like that. I’m currently working for a large vendor of mobility, Auto ID, IoT technology, as an industry expert over here in EMEA, so Europe, Middle East and Africa. 

Could you talk about how you’ve seen this evolution of Auto ID and RFID, into what we’re thinking more broadly of as Internet of Things. What were some of the initial challenges in that with the uses of those technologies, and how have they matured in a way that’s allowing the value to come through? 

In order to answer that question, to be very detailed you would have to break it down by industry, so for instance RFID has now a quite successful track record in retail, for instance in the fashion industry, and in that technologies have matured to quite some extent. There are other industries where the real value of Auto ID and RFID in particular are just coming up, and I’m thinking about automotive, and aerospace in Defense in particular. Probably the reason why it took so long in these industries in to bring something up which is now valuable and useable, is the fact that we’ve been lacking standards, and standards are very important if you introduce such technologies to larger corporations, and by any means especially in these high-tech industries, these are larger corporations. 

So, standards help you to justify the investment, and secure the investment in these technologies. Just very recently the European norming body for the Automotive Industry, named Odette, has put out these standards for the RFID usage, especially in the supply chain. Having said that, it has been a real boost to the RFID technology, especially in the automotive sector. Before that, we had the situation very often that there were some great use cases out there, but for almost every use case you would not only need an individual RFID solution, but also an individual back-end solution. With that, business cases were hard to justify, so the more standards you have, then the more platforms you have in these types of technologies the better the business case outcome is, the more likely it is to get these things through your part. 

That’s really an interesting point about standards, I know certainly in connected industry when you look across verticals, there are a lot of different standards for different industries, maybe not one overarching standard; but what I’m interested in is to understand a bit of how these technologies early-on had a lot of promise that took time to materialize, and now we’re starting to see the RFID technologies being used in many productive ways, I think it would surprise people. What are some of the use cases you found which jump out at you, or stick out in your mind for RFID which maybe the average person in industry might not even appreciate? 

First of all, we still have to differentiate when we talk about RFID between at least active and passive RFID, there’s different type of use cases depending on when you talk active or passive. Above and beyond that, there’s also a couple of neighboring technologies such as NFC and even probably Bluetooth Low Energy which will boost this. I guess in the end that helps the entire industry, because the more choice you have on that end, and the more standardised choice you have, the more use this will come up. 

But, back to your initial question. In the passive field it’s still the topic of asset management which is probably the one big use case; so, do I need information about the current whereabouts of my asset in terms of stage and gate, then I’m probably very well-off with passive RFID. That’s what happens in the automotive industry these days, especially if we talk loading devices and containers, so that’s probably the biggest use case in automotive I currently see. By loading devices, I mean all these containers you use to carry components for your car assembly; for instance, from the manufacturing plant to the final assembly line, component manufacturing plant to the final assembly line, and back, so for those you use special containers and load carriers. In order to track these, because they’re highly important to the logistic processes of automotive companies, passive RFID is just a great technology. 

The other thing is active technology. Active is a different price-point, but it also adds a lot of additional functionality to this. You might look at active when it comes to not only looking into the question at which stage for instance introduction in stock, or in transition is my current asset right now? But when you want to know exactly where it is, and we want to locate that. There are a lot of examples out now with active RFID technologies, and that’s the field of what we call RTLS (Real-Time Locating Services), and that has some great use cases in automotive than anywhere else, especially when it come for instance to locationing of tools and a lot of functionality you can put behind exactly that information.  

You were telling me there’s a high-end auto manufacturer on the assembly line who was using RTLS to make sure they would get the right screwdriver for their MES system. Could you talk a little bit about that use case? 

That’s technology used by Porsche the car manufacturer in Germany, on a couple of their lines. They use an active RFID-chip, or tag on their cordless screwdrivers. So, in the final assembly line they have to use cordless screwdrivers, in order to avoid any scratches to the already painted cars. But as you might figure, a cordless screwdriver can go anywhere in the plant. To keep it quite simple, you have certain operations where you need dedicated screwdrivers, rather than just a generic one. Those dedicated screwdrivers are much more expensive and have a lot more care effort than just regular ones, so in order to avoid somebody taking a regular screwdriver to be certified fitting, you need to make sure the right screwdriver is used for the right operation. 

Porsche is using that technology of RTLS to locate the screwdriver. They have a map of all the different working cells in the final assembly line, so they know what is going on in that working cell of the final assembly line; what is the operation to be done, and which other screwdrivers I can use for that, so I know which screwdrivers at the working cell, I know what needs to be done, I can match the information, and I can see this screwdriver is certified. If so, it can go along as planned, if not you can block the screwdriver from performing that operation through the PLC. I guess that’s a great use of RTLS, it’s a solution made by Zebra Technologies, and there’s even a video out on YouTube if you want to look, enter ‘Porsche’ and ‘Zebra’ as search terms, you’ll very soon find the video on that and its pretty interesting use. 

That’s amazing, if you think about the ability to really enforce adherence to a specific process in manufacturing or warehousing. Just recently it came out in the media on GeekWire that Amazon has applied for a couple of patents that have their warehouse workers have wearables which will help them track and optimize the location of the warehouse workers in the warehouse. Where do you think this is all going? Are we going to attract just more processes, or do you think this could work its way into other disciplines, or other industries? 

It definitely takes the step-in other industries, but one of the things you have to see is not only on the field of IoT, RFID, Auto ID we are moving into the next level, but also, we are moving on the next level when it comes to a couple of other technologies. That’s why we come and speak of what is the fourth industrial revolution, or Industry 4.0 That’s just a toolbox of a couple of very interesting and very promising technologies, to be used in order to digitize the next plateau of manufacturing excellence.  

So, one of the things I would add to what you’ve just said is, yes, it is able – or at least technologies are able to track workers, but when we look just a little bit into the future we will see workers are accompanied by a lot of autonomous vehicles in the warehouse, so, not necessarily does the forklift truck require a driver. So, we see human beings working together with machines, working together with robots and even drones etc. In this environment the information of who is where and when, and real-time visibility into a location of any given asset is not even the wish, but it is a necessity. 

This really leads to a vision in the future where we’ll be able design semi-autonomous systems, where humans will have judgement and will use semi-autonomous or autonomous tools and processes, where the judgement or the capabilities of humans will be augmented by all of these real-time tracking systems, and a bit of AI. Where do you think we’ll go with this, are there some risks that people are not thinking about? 

First of all, the question is where do we want it to go? When we put it from what’s possible in technology point of view, we can already achieve a lot more than current realities, but there are multiple views on the question of what we want to achieve. It’s very similar to where we started this conversation, first of all there’s the value in putting so much technology into our processes. I definitely see that we have a lot of processes these days where not only IoT and RFID can be of help, but where all the other technologies like AI, AR (Augmented Reality), Artificial Intelligence, and probably also RPA can add a lot of value, and we will see that. We have a couple of technologies, and we will be able to in the future add these technologies to a stack of technologies, plus human-being and human judgement to create value. 

For the foreseeable future the only risk is that we speak too fast in these technologies, and try to put in too many just because they’re there. Sometimes we just forget the question of is it worth it to do so, on a larger scale. Fortunately, enough of the market regulates itself on that trend, because it takes a lot of time for technologies to really kick-off. Just think of IoT as an example, as I said, it’s been out for quite some time, and there have been certain pockets of productivity coming by IoT, but only now as the ecosystem of IoT is really evolving, it comes to a broader use and a wider adaption since, it is a secure investment at this point in time. 

We really are starting to see these proofs of concepts and these ideas come to fruition. I wanted to ask a bit about your experience working with manufacturing, and particularly the auto industry which has been an early adopter of a lot of connected technologies, but over time this vision of moving toward a transport as a service model, or everything as a service model has really in a sense maybe it’s been the Shangri-La of a nirvana of IoT! What are your thoughts about what needs to happen for us to be able to truly enable ‘x’ as a service business model, and are there any industries that are leading which may have lessons for other industries? 

Well, as a service model this is already widely adopted across many an industry. For instance, if you are to produce newspapers or books, and this is another industry which has heavily suffered from digitalization, you have to buy large machines helping you to print your newspapers/books, and there’s a couple of manufacturers out there who already offer you these machines as a service, so you pay by page. There are aircraft engine manufacturers who offer power by the hour to the airlines, and of course there is the automotive industry, as your question indicated. With automotive I always think about two totally different areas, one being the whole process of manufacturing a car, and this very large ecosystem or this large automotive network coming from the dealer down to tier 1, 2, 3 suppliers including the OEM, not to forget the after status. 

Then on the other hand there is this whole notion of the connective product itself, so  connected car. You’re right, you’re pointing out that this as a service market is ramping up quite heavily; especially when we look into automotive and to the whole notion of connected car, it is worth taking a broader look? And if you nowadays look into the strategic intents of virtually all different automotive or car manufacturers, you’ll find very similar strategic guidelines for these car manufacturers. The evolution of the power train and the switch to electrical compulsion systems being one, the other thing is the connectivity of the car as such, so some might still call it a car, others already call it a hotspot on wheels, and along with that there goes this question around, is it really useful, and is it really necessary to own a car just to consume mobility? Or, is it enough to consume this mobility, and the ownership remains with the OEM? 

If there weren’t external forces pressing this, the automotive industry would not necessarily go in that direction, but with all these digital technologies and all the other trends, electric, connectivity and so-on, being there one of the things which happens is there’s quite some radical change in the market, and there’s quite some new players coming into the market because these times, especially for the automotive industries are so disruptive. Along with it goes the fact that if something is disruptive you have the ability of a fairly easy and fairly likely to be successful market entry as a new player.  

We already see that with companies in China, like Geely and others, they take these technologies and trends of no ownership needed, electrification, connected car, autonomous driving I haven’t mentioned so far, they take these and have the possibility of attacking the traditional car industry as such. So, there is a strong force, and a strong need for the traditional car industry to move themselves, that is a big boost to things like cars as a service. We’ve just recently seen that, at least in Europe, the two big providers which are owned by Daimler and BMW are staring to get closer together, and I guess we can expect some interesting news over time. 

It’s really interesting the relationship between a manufacturer and their clients or their users, or their end-customers as we say, has really changed. In our prior conversation you’d mentioned something that Tesla was doing with battery life, after the hurricane. 

Yes, that was after the hurricane and was all over the press. Tesla has some cars out in Florida and what they did, they had an online update to all of their cars, allowing the battery to exploit its full capacity in order to give enough range to become safe, and get out of the area of the hurricane. So, for instance, battery capacity is something which is a software defined feature in the car; which brings me to another point, traditional manufacturers regardless of which industry are currently facing, the products are becoming more and more intelligent. I know Momenta does quite a lot, if you talk about the Edge, more and more devices and more and more things in the Internet of Things become Edge devices themselves. So, a fair portion of the value to the customer is not defined by mechanical or electrical systems, but is software defined, and this was the case with Tesla. Tesla was able to give this capacity extension to everybody for a limited amount of time, and after a certain period they took it back. 

It is pretty amazing that you have this ability to touch your customers, to collect information about how they use the products in a way that this was never available before, and then be able to deliver a better experience. From the standpoint of manufacturers and using connective industry to create connected products, what are some of the considerations and really different ways of thinking that you see as really important priorities for manufacturers, and those who are designing the next generation of products? 

Firstly, nowadays you have to look not so much into your product, but what you first have to understand is how can I fulfill my customer’s requirements? There is a common saying that comes out of the sales motivational pitches, and it goes like… ‘If you want to sell a product, or want to sell to your customers, don’t sell on the specifications of your product, but sell around the understanding of your customer’s problem’. All these connectivity tools can help a lot, because they change the relationship being a one-time relationship, and then probably some service sales after that between the manufacturer and your customers, to an all-over and on-going relationship. Because, yes you probably still have the model of selling your products, or you already are at the stage where you give your products as a service.  

That changes, and the constant exchange of data and the ability to offer additional services such as a temporary battery upgrade, such as directions for your car, such as predictive analytics and predictive maintenance for your products, it changes the relationship between the manufacturer and the customer base to be an on-going relationship. As such, you have much more interaction with your customer which allows manufacturers who are building on this, to better make sure they have a long-term relationship with a customer.  

Again, there is this other number, it’s about five times the effort to win back a customer, rather than to keep the customer. In that sense IoT in particular offers some great opportunities as a technology, in order to establish this relationship, this ongoing relationship between the customer and the manufacturer. 

As we look to the future when we have all of these new sensors, autonomous systems and critical data which is flowing across networks, what are the concerns you may have about data security, data ownership, and ways in which we can ensure that we can trust these complex systems? 

First of all, our networks still have to cope with loads and loads of data, i.e. I just recently learnt that cars might produce gigabytes per mile, or per kilometer they actually drive on the road in terms of data. So, the first constraint at this point in time definitely is the network, and in many an industry, not only in the car industry, it’s the network and the ability to even get this information from the Edge to central systems. You recently had a webinar which I found very interesting about what does the Edge mean to the IoT world? I strongly do agree a lot of things are happening on the Edge because of the constraints of the network. 

But of course, we’re already in a world where you might be the integrator of a product, and again for the sake of simplicity let’s take a car; Chrysler just to pick one, might be the manufacturer of the car, but in fact a lot of the systems which build these cars are coming from large-scale suppliers. Of course, they interest and have a stake in the data as well. So, the first question is, who are the players in this data game? And that’s quite different to just pure ownership models too. Who are the players, and how can we distribute the burden and the value of data collection, data transfer, and data security between them? How can we make sure everybody who participates in that, is getting a fair share of the value created by those data? That’s the first thing. 

One of the answers to that definitely is going to be around the blockchain because it probably is the one technology being able to have a secure ledger, and a secure way to transfer and have the necessary transactions when it comes to putting value to all of your data. So, in that sense it is technically to be solved, but definitely it is a system where we need to put in a lot of process work and, put in this whole business model first of all to be able to ensure that you know who owns the data is also the one who gets a share from it. 

You brought up blockchain and distributive ledger technologies, and I’d love to get your thoughts on where you think the potential use cases are, if there are any projects which jump out at you is as particularly useful? And then some of the views on whether this current wave of hype might be overdone in some instances? 

Well, first of all the hype is not so much on the technology, the hype these days is more or less on some of the applications of that technology. I see some great examples especially for instance in the food manufacturing chain, to secure and to have product verification and secure logistic lines, and the same would then apply to even more critical value chains such as in pharmaceuticals and stuff like that. For that I guess the technology is already is great, where we see they hype is more on things like Bitcoin these days, and cryptocurrencies, and again that’s an application of an existing technology. So, where do I see values? I definitely see them in supply chain and logistics, and there’s as I said already applications out there which I think they’re great.  

The other thing is again in connected systems where you have to clear the data, and the value of the data across different partners, and you have to distribute efforts and value of data collection and data usage across different industries. So, to give you a concrete example, how would you remunerate a driver or an owner of a car for giving away the data, if a break manufacturer for instance, or a breaking system manufacturer actually uses that data to improve his product design? So, there is value to that and it’s a large number of transactions, and all of these transactions have a value. In that environment blockchain and distributive ledgers will have a greater impact. We see companies like ITOA already being very hyped on that one, I don’t know whether this is going to be that company or that system, but it’s definitely going to be the technology to be used in that area. 

A lot of really interesting work going on around these distributed acyclic graphs or the dags that IOTA and a few others are coming up with.  

One point, I guess that’s one of the success criteria’s for IoT in general these days, because we see that around the concept the partner and ecosystem – the ecosystem as such and the system of players is evolving. Also, out of your portfolio there’s that company called Senzi, and I think that’s just one example of where there are services building on the ability to, and the network ability to communicate from devices to let’s say central, and there are services formulating around. So, if I can consume a service as a manufacturer using for instance predictive maintenance, if I can use a proven technology to do all my micro-payments for the data I distribute, then I guess I have some means to put value to IoT, and I guess that’s a very-very important point that this ecosystem is rapidly building these days. 

I’d like to just circle back to the operational technology and information technology convergence, and from your experience do you have any… I wouldn’t necessarily say advice, but suggestions for experts in domains that are trying to reach across from industrial to information technologies, and vice-versa? How can experts and companies that have an expertise in one domain, more effectively reach across the aisle as it were to drive a converged vision and solution? 

It’s probably a very personal point of view but, the path for that is not necessarily through the technology, but the question should be different, especially from manufacturers these days if they want to embark this whole digital journey; they should ask themselves the question of… so they have an established business model as of today, and the question is, ‘What’s going to be the business model of the future?’ Undoubtedly everybody had a gut feeling that probably the business in 10-years is looking rapidly different compared to today’s business model. There are a couple of drivers which can be applied, or which might be applied to an existing business model, in order to evolve it to the future. 

We already mentioned as a service being one thing, and I don’t want to pick on all of them, but the other probably being an individualization of the product. Just before this webinar we had that discussion about micro-breweries, and that’s probably a very good example for especially consumers, but also institutional buyers and customers. They’re not satisfied with the standard product anymore, but they want an individualized solution, and that puts a lot of pressure onto current manufacturing operations, and manufacturing business models, because as you do that traditionally your number of variants, and the number of products you have to bring to market is exploding, and that puts a lot of complexity to your supply chain, and to your whole operation. 

So, you have to identify as a first step what are the change drivers to your business model, and depending on what that brings, you are to choose what are the right operational approaches in order to attack and tackle that, and after that you have to solve the question of what are the digital technologies out of what I call the Industry 4.0 toolbox, what are the technologies I need to get hold of as a manufacturer in order to solve that equation? And, of course I have to make sure that my backbone is digital enough, a lot of people always think about when we talk about digitalization, especially manufacturing, a lot of people are talking about all the fancy applications such as IoT, such as Augmented Reality, such as drones flying all over and delivering, inspecting, and doing whatever.  

But what it is about as well is, you have to have a digital backbone, you have to have the capability to process all this data, and you have to have the capability of having end-to-end processes which you not only have in real life, but also in your digital life. So, that brings up an equation which might be a good strategic framework for your digital operation, and the application of things like IoT in general. 

Fantastic. I always like to wrap up my conversations with people with a request for your favourite resources, or recommendations that you give to people, to friends, that you’re passionate about, or that you find can be really valuable as a recommendation. It doesn’t have to be in technology. I’d love to get at least a favourite resource from you. 

A book I’m currently reading which I find very interesting is a guy called Ray Kurzweil who used to be with Google very substantially, and his book, ‘The Singularity Is Near’. I can sense that it is a little provocative, but just to open your mind and think about technical possibilities that makes sense. Then the second question to that which also makes sense in that respect is, how much of that do I really want as an individual? So, that book might be very inspiring to a lot of people. 

There is another guy, Antoine de Saint-Exupéry who wrote the book ‘The Little Prince’, which I think is pretty famous, it’s a children’s book. He did a quote which I like a lot, and that is, ‘Technology always goes the same way. It goes from primitive, to complex, to simple’. Just think about that and try to digest that a little bit. Ask yourself where are we on that journey towards the totally digital manufacturing, and totally digital enterprise? We definitely left the primitive area, we’re probably quite well advanced on the complex area, and our vision should be to make these processes as simple as possible, and the solution as simple as possible. Only then will any company and any individual be able to succeed in that. If we can’t actually make it simple, it’s going to be difficult. 

That’s a fantastic way to wrap up our conversation, Ralf. Leonardo De Vinci actually said, ‘Simplicity is the ultimate sophistication’, so I think we’re on that road. I want to thank you again for your thoughts, it’s really been a fun conversation, and have touched on a lot of really provocative areas and illuminating insights.  

My pleasure. 

So, with that, we look to follow up the conversation; anyone who’s listening, please feel free to reach out and send any questions, and we’ll look to follow up. 

Thanks very much.