Einige ERP-Hersteller leisten sich eine dedizierte Forschungsabteilung für neue Technologien über ERP hinaus. In einer Master-Vorlesung im Juli hatte ich das Vergnügen, den Director der IFS Labs per Skype zu ERP und Blockchain aber auch anderen technologischen Trends befragen zu können und wir haben die Session direkt für den ERP-Podcast mit aufgezeichnet.
Herausgekommen ist ein interessanter, englischsprachiger Einblick in die Entwicklungslabore von IFS. Viel Vergnügen.
Foliensatz Beyond ERP – Blockchain und andere technologische Trends,Bas de Vos IFS Labs
————————–
Wenn Ihnen unsere Folgen gefallen, dann freuen wir uns über eine 5-Sterne-Bewertung auf Ihrer Wunschplattform, damit auch andere auf diesen Podcast aufmerksam werden und wir das Angebot weiter verbessern können. Zeitaufwand: 1-2 Minuten.
In diesem Sinne: keep connected.
Herzlichst
Ihr
Axel Winkelmann
Transcript:
ERP-Podcast, episode 24. Beyond ERP, blockchain and other technological trends. Interview with Baz Defoos, director of IFS Labs.
Some ERP manufacturers provide a dedicated research department for new technologies beyond ERP. In a master lecture in July, I had the pleasure of interviewing the director of IFS Labs via Skype on ERP and blockchain, but also other technological trends. And we recorded the session directly for the ERP-Podcast.
It turned out to be an interesting, English-speaking insight into the development labs of IFS. Enjoy! Welcome to the ERP-Podcast. The podcast for those who want to actively engage in the use and design of corporate software and the resulting changes and potentials in companies.
With this podcast, I would like to accompany you with your own thoughts and interviews on the design of modern IT concepts, such as walking or driving. With this, I would like to offer you information in this time of technological change, which cannot be found in magazines, textbooks and scientific articles in this form and which cannot be found in everyday life. My name is Axel Winkelmann.
I am a professor of business economics and business informatics at the University of Würzburg. What I would try to do today, and this is getting a little bit used to, because this is the first time I’m actually giving a guest lecture through Skype. So that’s a little bit, yeah, finding my way here a little bit.
If you guys have questions in between, just raise your hand and I’ll try to pick it up or just shout out and we will just talk. So what I’m actually trying to do today is try to, let’s say, tickle your minds a little bit to see beyond ERP. What is ERP beyond ERP? And to do that, I would like to explain a little bit first what I do myself in my day-to-day life.
Like the professor introduced, my name is Bas de Vos. I’m actually director for IFS Labs. And to explain what IFS Labs is, I’ve got a couple of cartoons for you guys that I would like to take you through.
Because what you see today and everywhere out there in the day-to-day business is that there’s a lot of technology out there. There’s a lot of trends out there. There’s buzzwords.
There’s real things. There’s things that might be hype today, might be a real product tomorrow or the other way around. And you can ignore it.
But basically, if you ignore it, you might be done a couple of years from now. I personally believe that for companies to survive what’s happening to us, to survive the disruption that’s out there in the market. I mean, who says Amazon is not going to be the next house builder, right? Nobody thought that Google was going to be a car manufacturer.
And actually, they’re building cars, right? So that’s something that the automotive industry might see as a big disruption to them. And that happens to IFS as well, right? There’s startups. There’s new companies.
There’s the big tech firms that are out there that are trying to innovate and try to look at these all these trends and technologies out there. And it’s happening to our customers. Our customers basically also have news ways of doing business.
We have a customer in Sweden, for example, that’s in the pest control. So basically, they are very good in capturing rats, basically. And they now have a new service out on the market that is what they call the smart trap.
Basically, rat traps that are connected through IoT to the Internet all the way down to the ERP. With that, they’re disrupting the pest control business. So for us as IFS, it’s very important to stay ahead of that, to make sure that we actually understand what’s going on.
And that’s why we have IFS Labs. We’re basically a small team within IFS R&D. So basically, the research and development department that’s responsible for creating the IFS products, the ERP, for example, that you guys have been exposed to a little bit.
And the funny part about IFS Labs is that we try to be a little bit of a startup within a large enterprise. I’m a firm believer in the creativity of my people. And the only way we actually are able to get that creativity from them in the full sense is by just letting them do their thing.
Not steer them too much, but basically just allow them to creatively look at new ideas, new opportunities, and to build their own, let’s say, mini products. So we have kind of a startup within the larger IFS organization. And very important, and I’ll come back to that at the end as well, we’ve got a license to fail.
You can’t succeed in life. You can’t succeed in IT if you are afraid that whatever you’re investing your time in is not going to return its investment. If you fail, you fail, you learn, you continually step up.
And that would also be my advice to you guys if you’re going to look for a job later on or you’re going to start your own company. Yeah, have the guts and basically start what you think is a great idea to investigate. The type of projects we’re looking in as IFS Labs are things like digital twins, machine learning, where sentiment analysis is an interesting example.
Blockchain, which is basically going to be, let’s say, the underlying subject of my presentation today to you guys. But we also look at things like autonomous assistance, wearables and devices, Microsoft HoloLens, a very interesting device, right? We’re investigating how things like that can actually make sense to ERP. And not so much all kinds of cool 3D models, but really how can you take a end user of an ERP application like IFS applications, who’s out there in the field and needs to do a repair, how can you use technology like this to actually help him do his job? Drones, another interesting example.
But like I said, today I actually want to talk about blockchain. So for that, I got a question for you guys. Who knows what blockchain is? Well, about a 40% score, I would say.
So next question. Who knows what Bitcoin is? Oh, that’s way more fingers, I would say. So the answer to what blockchain is, is actually quite simple.
Blockchain is a technology that sits behind Bitcoin. So that’s the end of the lecture. No, that’s just a joke.
So some history behind this, right? Let’s start in the middle, because as you guys said already, you know what Bitcoin is. And Bitcoin, a new digital currency, right, was first conceptualized in 2008 by a guy called Satoshi Nakamoto, who we still don’t know whether he exists. Actually, there’s four guys out there in the world who say that they are Nakamoto, but no one has ever been able to prove it.
But he did that in 2008. And in 2009, he launched Bitcoin. And we all know how that’s developed, with some ups and downs.
Basically, the value of the Bitcoin evolved from eight cents all the way to almost $3,000, as it is today. And some interesting part in that as well is, if you look at the number of transactions that are actually done with Bitcoin, so not only the value, but how many people are actually transferring a Bitcoin from one person to another, you see that there’s been quite an exponential growth as well, which is also a concern for Bitcoin, because what’s going to happen if this growth continues, right? Will the backbone be able to actually deliver this? Actually, this guy, Satoshi Nakamoto, was not the first guy talking about something like blockchain. Back in 1991, Mr. Haber and Stornetta, they were actually the first guys in IT that described what they called a cryptographically secured chain of blocks, which blockchain chain of blocks is sort of the same thing.
And people have been writing about it quite a bit since 1991. But actually, Mr. Nakamoto was the first one that actually achieved that. And the interesting part about blockchain is that not only for money, and that’s where it’s starting to get a little bit of interesting for ERP, not only money, but there is since then some other use cases that have seen the light as well.
There’s people investigating blockchain as being able to capture a digital identity. So instead of having a passport, you actually have a sort of a digital identity that sits on blockchain that goes with you through life, that can be used by suppliers online to basically verify that you are who you say you are. And Airbnb is, for example, investigating in that or HomeAway, which is sort of a competitor to Airbnb.
Electricity trade, right? If I want to trade the electricity that I capture in my house with my solar panels, if I want to trade that back into the world, maybe I could use a Bitcoin to actually track all those purchases and sales in there. There’s a company called Grid Singularity that’s quite actively involved in that. Home properties.
Today we need to go to, the Dutch word is Cadastre. I’m not sure whether that’s close enough to German. It’s basically a government agency that you say, OK, I transferred the ownership of my house from myself to somebody else that has acquired my piece of land.
Instead of going to the government and to a notary, can we actually do this online as well through blockchain? And companies like Bitfury and Factom are investigating that. And the interesting part is, I said it’s a little bit more interesting for Europe at this stage, but these cases are still what I would call a little bit financially let’s say ownership oriented. So with other words, it’s still about I have something and I give it to someone else.
And that person wants to give it to somebody else as well. And that way we build up the chain of law. We, I think basically that for it to really impact Europe, we have to look a little bit further outside of the box here.
But before I do that, I actually want to explain a little bit about the basic working of blockchain and its current state of technology. So for that, I sort of have a one pager that I typically use, right? The basic principle of blockchain is actually quite simple. I’ve got a transaction.
That transaction, I basically, you could represent that as a block, but I basically upload that transaction into my chain, into my blockchain network. And every participant in that network, we call them miners in the Bitcoin world, that basically makes a registration of that transaction. He does that by basically creating a hash of the transaction and stores that transaction on his own PC as well.
And that way we’re building a block of transactions that is validated by all the participants of the chain. And then everybody can see those transactions. And that’s basically the basic workings of a blockchain.
Reality, of course, is a little bit more complex. I’ll go to that in a second. But the basic idea or the use cases that you could achieve with this are all circled around these four terminologies.
It’s all about creating consensus. We’re all in agreement that this is actually true. A single picture of the truth.
It’s about proponents being able to trace the transactions all the way back to the origin. It’s about immutability because it’s a distributed ledger. It’s distributed over multiple systems.
It’s no longer in a single database. I can’t change it that easy. And it’s about finality.
This is it. The world is what it is. This is it.
So, if we take one step deeper into the technical workings in here. So, who’s familiar with hashing in the room? That’s not a whole lot of numbers. Okay.
Basically, what hashing is, hashing is a one-way cryptographic way of basically encrypting your data in a one-way direction. So, basically, what happens? I take a word. I hash that.
So, basically, I would use a key that’s called the salt against it. So, with that key, I can basically transform it into something that’s called a hash, which is basically something that reads like this. And the interesting part about hashing is I can’t reverse it.
The only thing I can actually do is take the initial word I want to hash again. So, in this case, Bilbo. So, I can hash it again by using the same salt, but I can’t take the actual hash and reverse it into Bilbo.
So, the only thing I can actually do is take it and do it again to actually make sure that this is what it is. So, I can verify whether my data has not been tampered with, because if I would change a number in here. So, let me take my mouse.
If I take a number in here, if I would change this three into a four, I can validate whether this encryption is still correct by basically taking my original input, hash it again, and I should have the same outcome. I can’t take this number and reverse it into the original world. And this is an underlying principle to blockchain, because how does it work? I basically take a new block in my blockchain.
And what I do is when I take this block, I basically put it in my blockchain. But part of this, let’s say, set of data is that it actually contains the hash of the previous block in the chain. Do we understand that? And I use that to basically generate a new hash.
So, what I can do, I can basically say, okay, I have this, let’s say, bits of information, all these transactions. I can use my salt to basically encrypt it again and therefore verify that the hash that I see here is correct. So, whenever somebody is going to tamper with this data and change something in one of these previous blocks, it will therefore change the way that this hash is built up because it’s every time it’s built on the previous discussion.
And therefore, I can validate that my data is tampered with. So, the only way that I could actually hack blockchain, that’s at least a theory, right, is go into all these computers that are sitting there and that are part of the blockchain network is basically by hacking them all and everywhere replace the entire blockchain, not just a single record. Because if I replace a single record, my hash that I can recreate over my entire chain is no longer correct.
And that’s the underlying, let’s say, technical principle of blockchain. That was also the first model, and that’s the model that’s used for Bitcoin. Since then, the world has evolved a little bit further.
So, today we are in what we call blockchain 2.0. It’s what we call smart contracts. And assuming you’re a little bit familiar with programming, if you compare it to PLSQL, you could basically say what PLSQL is for Oracle DB, the ability to add code into the database. That’s also what smart contracts are for blockchain.
So, instead of just having the transaction in my block, I’m also adding the business logic inside of my block. And it has some benefits and some drawbacks. The benefit is, for example, that everybody that’s using the blockchain is using the same business logic.
The drawback to it is, I can’t just easily change my code anymore because whenever I change my codes, I therefore change the hashes, and therefore my blockchain is no longer, let’s say, dependable. It’s broken. So, I would basically create a branch and create a new blockchain whenever I change my codes.
There’s also a next step that’s called blockchain 3.0, and that’s basically what we are investigating today. This is not really in production or available that much yet. It’s a little bit in an experimental phase still.
It’s called Project Bletchy within Microsoft, for example. And this is basically adding cryptlets to blockchain. And cryptlets are a little bit sort of secured entrances in the blockchain.
So, by taking a cryptlet, I can say, okay, everybody can see this blockchain, but only if you know the codes to the cryptlets, you’re actually able to make a transaction into the blockchain. So, if you add these models up, blockchain 1.0 would basically be just a simple state machine that registers all transactions. Blockchain 2.0, you could add business logic into the blockchain that allows you to actually make more advanced transactions.
And then blockchain 3.0 is actually the entrances in the blockchain are also protected, which basically allows you to limit, authorize whom is able to enter the blockchain, enter a blockchain transaction and who isn’t. So, slowly but steadily, you see that the actual blockchain principle evolves as well. And that leads us to basically this picture.
In here, I try to sketch a little bit the different models that are possible than today. And to the left side, we see, of course, what we have today, right? It’s ERP. It’s IFS systems.
It’s our competitors. It’s everybody that just has a single database on site doing all kinds of transactions. And all the way to the right, you basically see Bitcoin, the most open implementation there is.
It’s fully public. Everybody can see it. Everybody can make a transaction, as long as you own, of course, the codes to the Bitcoin.
And it’s fully connected. So, there’s a couple of models in between. And those are more interesting for enterprises to be used.
Because, let’s face it, no enterprise, no bank is actually going to expose all its data to a full public ledger that everybody can see it. So, with these technologies I just described, like cryptlets and adding Go to it, you’re going to be able to do more. And the models that sit in between are basically ranging from, okay, I can see it, but I can’t change anything.
Or I can’t even see it, and I can’t change anything unless I’m part of this landscape, right? So, if you look at this, where’s my mouse? The second model basically says, okay, let’s see, what are we? 25 people in this room. We create our own blockchain. And only the people inside of this room have access to the blockchain.
And the type of transactions that you can do, I basically, a little bit like traditional database, right? The first person can do these type of transactions because this is the Kryptlet that you have access to. The next person has access to other Kryptlets. So, it’s basically fully permissioned private blockchain.
And then there’s something that sits in between as well. I can still give only certain people access to the Kryptlets, and therefore, they are able to make changes. But it’s a more public one.
It’s still out on the internet. It’s still distributed on the entire world. And then you also have models that are basically sitting in between and that are what we call hybrid connected ledgers.
So, quite some few tastes in there. But you already see that different types of use cases are existing or servicing in the different types of ledgers. Financial institutions that are looking and researching whether their backend payment systems can be replaced with blockchain.
They would basically look at this technology or this model. While digital identity, that is something only specific persons are allowed to make a transaction, right? I don’t want anybody else but myself to make a change to my digital identity. But as the idea is that I can use it everywhere to identify that me is me, you still want to have it public.
So, there’s different models in between. So, very interesting. But where does ERP come into play? And for that, I would like to present sort of a use case that we are talking about with our customers, right? If you look at IFS, IFS is focusing on product and asset centric industries.
And one of our key markets is aviation. We actually got customers all over the globe. Emirates Airlines, KLM in the Netherlands, Southwest in the US as a customer.
So, we’re quite big in aviation. So, this is interesting for us. And as being IFS Labs for me, it’s always like blockchain, really cool technology.
We should investigate in it. Yes, of course. But what does it mean for my customers? Yes, we have got a license to fill.
Yes, we can try out. Yes, we can investigate all kinds of cool stuff. But what’s in it for our customers? In the end, if there’s no business case, if there’s no value, you investigate and you stop.
So, for blockchain, let’s have a look at aviation. And for that, setting the scene a little bit. If you look at modern aircraft, if you take a Boeing 777, then that consists out about two to three million parts, depending a little bit on how I count.
And in aviation provenance, so basically who manufactured this item, is really critical. Is this a part that I’m installing in my aircraft that was produced by Boeing or Airbus? Or is this a part that somebody acquired on Alibaba.com? And therefore, it looks the same, it feels the same, but it can’t use the same stress or the same number of rotations as the original part can do. And therefore, my aircraft falls out of the sky.
So, that’s quite critical. And at the same time, you also need to make sure that your entire maintenance history is kept and stored. And of course, this is highly regulatory.
There’s so much regulation in aviation. I think there’s more regulation in aviation than in any other sector there is. So, how do people do that today? Basically, by taking care of their own turf.
Every party in a chain, whether it’s the manufacturer or it’s the operator all the way at the end, they all look at their own piece of the garden, their own turf. So, they have their own system. Best of breed works perfectly, but it’s not connected to any other system in the chain.
There’s a lot of written and verbal feedback in the chain, which leads to incomplete records. There’s a lot of stuff that’s still done on paper, because at least then we have a paper file that we can send with the assets. There’s a lack of standardization, therefore.
It’s even a fact that sometimes the serial number needs to identify the parts that was assigned by the manufacturer is not the same serial number as the operator all the way down uses because he coincidentally has a different asset or from a different manufacturing with the same serial number. So this is all quite disconnected. And in the end, it leads to high costs for compliance.
So in something like that, if you would like to set up blockchain in a world like this, what are those different players we see in there? And how would you need to look at it? Because I don’t believe that a single airline could say, okay, now we’re gonna do blockchain. And this airline basically sets up a blockchain. And then what? Blockchain is useless and doesn’t make any sense if you look at it from a single company point of view.
The only way that blockchain actually makes sense is if you look at it from a network point of view. So around something like blockchain, you have different type of parties that all need to work together in able to create a production worthy blockchain. And we’re talking about the technology consortiums, Microsoft or the consortiums like Hyperledger and Ethereum Alliance that are basically looking, okay, we’re gonna be working together to make sure that we’ve got the backend technology.
You’ve got the participants. In my example, that would be KLM, that would be Emirates, that would be Southwest, that would be a Boeing. Those would be all the parties that sit in there.
The regulators, the FAA, the JAA, which is the European version of it, basically saying, okay, we need to regulate this. And software vendors, which is us, which is IFS and some of our colleagues. And together, if they would work together with a purpose specific, so for a specific purpose point solution, if they would cooperate, collaborate on that together and are willing to invest to gradually build something that works towards a solution, and it should be focused around single picture truth, then I believe that this actually could work.
So, and of course, we as IFS, we are sitting here, we are a software vendor, so we would look on how should we contribute towards something as blockchain. And to do so, let me clarify an example a little bit more. Today, as I said in my previous example as well, many companies, all companies have a single organization, large IT system.
And let’s face it, it can always be better, especially if they’re not a customer of IFS, of course, but it can always be better. But today this works, as long as I stick within the boundaries of my organization. A matter of fact is that this is not how the world looks like today.
As in the example, there’s gonna be a lot of talking, there’s gonna be a lot of interaction, there’s gonna be a lot of working together. So in practice, it’s not one single IT system, most likely not even within one organization, it’s only one IT system, right? But okay, let’s simplify it for the sake of the argument. But if you would look at multiple parties together, it’s multiple IT systems, they are connected, sometimes via phone, sometimes talking, text batch-wise interfaces, some of them are more advanced through a web service or whatever, but it’s all interfaces.
And this basically creates issues. There’s many single points of failure, right? There’s many different systems, many places, many interfaces where it can go wrong. It’s difficult to create a single picture of the truth.
There’s a high cyber vulnerability. Cyber vulnerability is quite interesting in these days, right, with WannaCry, and now the most recent Ukraine successor to that. If I hack anywhere in this chain, I can actually influence the entire chain.
And this comes with high cost. So how could blockchain help a situation like this in aviation actually going forward? Well, what if the world looks like this? You still have that chain, right? You still have the supply chain of an aircraft assets. We have the manufacturer who produces it, for example, Boeing or General Dynamics or any type of organization.
Then somebody puts it on a ship, brings it to Europe. There we have somebody that installs it maybe in another assets. We have the inspectors or regulators, the JA that want to do an audit or want to do an inspection for whether it’s correct.
Then we have somebody that’s maintaining the assets. So every 10 flying hours, I need to inspect this, whether it still works or every two years, I need to replace the battery. I mean, it’s all inspections, it’s maintenance.
And then you have the guy actually flying the aircraft, the operator that does stuff with it. I’ve flown two hours, I’ve flown 25 hours, I made five landings, I had a whatever. It’s all information that’s critical about the assets.
And what if all these different parties based on smart contracts, based on cryptos are able to actually take a transcript of a blockchain transaction and put it into a connected permission ledger. So something that is available through the entire chain, but people can only contribute, change data that are actually allowed to through a cryptos. What you then get is basically a situation where this entire asset history.
So I take a single item in an aircraft all the way from when it was initially produced by, for example, Boeing or a third party vendor to Boeing, that was the first party that actually entered this asset into the blockchain. And therefore I already got the first benefit, which is provenance, remember? I know that this asset was uploaded by Boeing or General Dynamics and not by an organization like, like I said, some fuzzy party somewhere abroad. I know it was transported by that party.
I know who installed it. I know who inspected it. I know who maintained it.
I know who operate it. And I’m able to build up a complete life cycle history of that single asset in there. And I think that could potentially, if all these parties work together, would have very interesting benefits for aviation.
It would lead to improved data quality. We would have an accurate maintenance history throughout the entire life cycle of the assets. The Holy Grail in aviation, the single traceable records.
I have a serial number that I know that this, it contains everything that belongs to this asset. Today that’s impossible to achieve. I would have an increased trust in the chain.
So in other words, the JAA would put less effort on me to prove that I actually did this. And also the parties that are working together would get increased trust that actually, well, the information they have on the asset is actually true, which would all lead to lighter administration, therefore lower cost and a higher system utilization. And if you look at aviation today and you guys all follow the news, right? I mean, there’s two things that are important in aviation that is lower cost.
It’s always about lowering costs and it’s about flying more with a single aircraft because that is basically allowing me to get a higher revenue out of the same asset that I’m flying out for 30 years. And something like this would, I think, really help aviation. But like I said in the beginning, it also requires a lot of parties to working together, which is basically why we as IFS are now talking about this and talking with our customers and investigating and researching and putting our little stone into the brick wall that we have to build.
Is this gonna work today completely already? The answer is no, unfortunately. There are still a lot of challenges that need to be overcome. The sheer amount of computing power and eco footprint for today’s blockchain technology is off the charts.
If you look at blockchain, remember the chart I showed with the transactions, they say that the amount of energy you need to actually compute today’s blockchain is the same as the energy consumption of irons. So it’s insanely high. The transaction latency is quite high as well because it takes more time for all the participants in the blockchain to actually process your transaction.
There’s a challenge about compatibility and trust and the existing IT landscape, how we’re gonna bring that together, which is specifically why, where we as IFS basically wanna lead the way as well. It’s about knowledge. It’s about the fact that not a whole lot of people know about this already and know what it can mean.
And also within our customers, we see this, like they’ve heard about blockchain, or they’ve heard about Bitcoin, but they’ve heard about blockchain, they’ve read articles online, they know it could mean something, but what does it mean for us? How can it help us? Quite a very important item as well. Laws and regulation, quite important in that as well. And one of the biggest, but that’s a little bit more, let’s say something we have to work out together.
It’s like, who owns the data? Who can see it? And who can change it? I mean, that’s in any implementation, but that’s in something like this, it’s very important to resolve questions like this. Do you guys have any questions so far? You talked about transaction latency and one of the problems I see with blockchain is that there’s a large amount of data standing behind these transaction IDs. So how do you expect to handle the companies with all the data that they’re actually sharing over the blockchain? Well, that’s a very good question.
And my answer would contain two parts. Part one is basically, if you remember the slide with the landscape, I mentioned purpose-specific in there. So I don’t think for us as industry, for us as everybody collaborating on this, to make this a success, we should start small, right? Because the data volumes can go quite high.
And also I don’t think that something as blockchain is going to replace your underlying business systems like ERP. It’s actually, what are the data sets that are gonna be critical to maintain in a landscape like this? And limit the data that you actually wanna upload to the blockchain to only the minimum that you require. At the same time, that doesn’t resolve it the same way.
So it would also require, I would say, different ways to reach consensus. So there’s, for example, some investigations around, maybe we should not have everybody creating consensus, but maybe you need to have some sort of a critical mass in there. If 55% of your blockchain agrees, it’s also right, right? So there’s different ways of looking at it as well.
And the last part of the answer would be clouds. Cloud has given us an incredible increase of scalable computing power. And what you will see around the blockchain as well is that it can be quite, how do you say that? Quite fluky, right? So it can be, hey, I’ve got a lot of transactions now and now I’ve got nothing.
Now I’ve got more transactions, now I’ve got nothing. So, and this is where cloud is coming in as an answer. And for example, if you talk to Microsoft, and last March, I was in Redmond with Microsoft Research talking about various subjects, and one of them was blockchain.
And their answer was as well, for us as Microsoft, I mean, our big drive going forward is Azure computing. So for us, this is why blockchain is an important area for us to research as well, because if we can get all the blockchain transaction done on a scalable Azure environment that still has all the different parties connected, but they’re all using the same underlying Azure backbone, you still get your distributed ledger with the security of Azure in there, but you get your scalability of cloud computing that sits behind it. So, yeah, to summarize, it would be decreased the number of computing power we need, and therefore transaction latency by creating smarter technology, limit the amount of data that we actually put on the blockchain by letting the ERP and the other systems play the role, and I would say cloud.
Let me put another question in this area. The informatic guys, they always talk about technical issues with data, and then they see we need to exchange data, we need to exchange functionalities, and then they come up with something like service-oriented architectures. It all seems to work until it really works, which means in the business world, it’s not only about the technical aspects and the technical aspects of the semantics, but it’s about business power.
It’s about who is the strongest guy in the industry, who can force some standards, who can ask or require the data, who can demand the data, and stuff like this, so it doesn’t have to do with technical issues rather than with power issues, with semantic business issues, and I’m not quite sure whether the technical guys who are in the blockchain area at the moment see this part of the whole problem. Yeah, I think this is a very good remark, and there were a couple of questions in your question, basically, so I’m gonna see if I can digest them out here. Well, first of all, if you’re talking about the power of the companies that sit behind this, I would say there’s quite a lot of power behind the blockchain.
If you look at Everledger and, sorry, Hyperledger and Ethereum Alliance, and if you look what companies are all contributing to something like blockchain, and also if you talk to the analyst community on what blockchain could mean, you see there’s a lot of drive towards blockchain at the moment. At the same time, to be very honest, I’m not sure whether blockchain is going to live up to its promise it has today, but I also see that, and this is also a little bit of license to feel right, I also see it as a tremendous opportunity if we can get it working, and that’s why we as IFS Labs are investigating and investing in it, and if it doesn’t work a year from now, we will just pull the plug and do something else. That’s fine, but there is a tremendous opportunity, and therefore we have to sort of have the, I feel, the responsibility to invest in it.
With regards to the business angle to technology, I would even like to take it one step further. Successful implementation of technology starts with humans, and a very often overlooked, yeah, challenge with introducing new technology is the human change it requires, and by definition, humans don’t want to change. We did a survey with about 750 decision makers recently where about more than 50% of the companies we interviewed actually says, well, the biggest risk for us going forward with introduction of new digital transformation trends is the inability and the unwillingness of the organization to change.
So you can start thinking about cool stuff like HoloLens, helping the end user, or robots to make your flow in your factory smoother, but the only way you can successfully implement that is if the people in your factory, your workers, your staff, your management is actually embracing these technologies, and the only way they actually do that if there’s something in it for them, and something in it for them, part of that is the business value. What’s the business problem? What’s the business challenge? What’s the business thing we solve? But part of them is also just their benefits, right? You need to convince them that this works, and to a certain extent, this is blockchain as well, right? We see an opportunity, but you’re right. There’s a lot of questions that need to be answered about who owns data ownership, and what kind of business problem do we actually solve here? And is it worthwhile, the investment of doing it? And well, you name it.
It’s a little bit the data ownership, privacy, security, that ties much more into underlying business problems than the actual technology. I’m convinced that if IT wants to resolve this and keeps on developing on the cryptolite technology that we’re now talking about, that we can actually resolve this quite well. If we put enough power around it, we can resolve this as well.
But if you look at data ownership, and if you talk about transparency in the chain, there’s a very big underlying fundamental issue. Do I, as KLM, want to share my data in a chain that also has my competitors, or my customers, or my suppliers? Data transparency is not a technical problem. It is a business problem.
So yeah, you’re totally right. Did I answer all of your questions? Because there were quite a few hidden in your story. Absolutely answered.
You agreed with me, which is good. Lesson number one, guys. If you want to have a good mark, agree with the professor.
I think the informatics guys always forget about the business aspects that are behind it. Standards in the car industry always work very well because there’s a strong car manufacturer who can force all the other supply chain members to do what he wants. But in most chains, we don’t have this strong leader who can force everyone, or we have large and very strong competition, as you said, and the people don’t want to give away any data because they fear that their competitors can do some predictions about the data, can estimate about some moves that you will do.
And this is one of the problems I still see. And the more we open data, the more these business problems come to life. Which is a very nice, let’s say, bridge to an example I can share here.
I was, to be very honest with this, I’m not completely up to speed to this story myself, but Renault is actually investigating whether they can use blockchain to have all the repairs and therefore also the mileage of the car that are done through the Renault car dealerships, that they can actually put it in blockchain so that you get your maintenance history of your car into blockchain, because one of the biggest drivers for the value of a secondhand car today is the mileage, correct? And did this car get its full maintenance history every so many kilometers or every so many years? That really drives the value of a secondhand car. So the idea of Renault behind this is that they actually wanted to say, okay, if we can put a system in place that guarantees that this secondhand Renault, well, nobody wants to buy a secondhand Renault, right? But if you want to buy a secondhand Renault, that you can actually guarantee that it has this mileage and this maintenance in the right timings, it increases the value of your car. And therefore the value of a Renault in general, and therefore the fact that people maybe buy a Renault quicker than they would do in the past.
And so automotive, and then also with self-driving cars and Google investigating in that and the upcoming of Tesla and this entire world, I mean, automotive is actually an industry, I would say, that is very vulnerable to disruption today. And car manufacturers have to do it. And if you don’t do it, look at the upcoming of the electrical car.
A nice example in the Netherlands, which, sorry guys, was not a very good for the German car manufacturers, is that a lot of people in the Netherlands drive company cars, lease cars, right? And the big German brands have always been the biggest, especially Audi and BMW. And then the Dutch government says, okay, you know what? If you drive in a full plug-in hybrid car that you can charge at home in the power sockets, you actually, your tax rate decreases from 22% or 25%, even down to 7%. So instead of paying 500 euros per month for a car to the tax authorities, you pay only 150.
Well, we’re Dutch, right? And we’re cheap. That is at least what everybody sells about us. So what happened is that all the people say, okay, we’re gonna buy a plug-in hybrid vehicle.
And what Volvo did is that within a year, they had a car, a plug-in hybrid car, the Volvo V60 in the market into the, let’s say into the brackets that typically gets acquired by people looking for a lease car, where they were always lost to the competition of BMW and Audi. And for four years, they were the biggest, they had the biggest numbers, much bigger than the typical German brands. And then BMW came with their plug-in hybrid version three months before the Dutch government changed the rules back to 22% for these cars.
So you need to invest because before you know it, you lose a big size of your market share because somebody else comes in with a new idea that is disrupting your business. Yeah, absolutely. Any more questions? When you want to create a part in IFS, you have to create three parts, stock part, sales part, and procurement part.
You will have to change it for blockchain, do you? Because it’s more data. Yeah, that depends a little bit. To be very honest, I don’t think I’ve got a clear cut answer to that.
And those are typically, let’s say the reasons we are in IFS labs, who basically investigate a technology like this and have answers to questions like that going forward. So what’s going to be the strategy for implementing? If you look at the speed of change, there’s a very fine balance between investing too early and being too late, right? So the big thing that we in IFS try to achieve is have something become a product the moment it needs to be a product. And a good example like that, a total different subject is IoT.
IoT started four or five years ago as a research project like blockchain is today in IFS labs. And we investigated that and we came up with a vision and we came up with a concept. How should this be implemented? We went to one of our customers to try it out and actually get some practical input.
And at a certain point we would say, okay, now we know what this means for IFS. Now we know how the market is developing because everybody’s talking for IoT for the last five years. But very honestly, IoT becoming mainstream, that everybody’s looking for products and is actually implementing it and is actually doing it.
That’s now, right? That’s not five years ago. Five years ago, it was talk. Three years ago, it was proof of concepts.
Today it’s production. And because of IFS labs starting four or five years ago Asking questions like this and researching what this means for IFS, we were able to actually launch an IoT product earlier this year, exactly at the right timing it needed to be a product. And this would apply for blockchain as well. I don’t have all the answers yet.
I’m very honest with that. We are in the middle of this, yeah, let’s say this journey. We are exploring.
We’re in an old wooden sailship trying to find America, right? We don’t know really where we’re going to end up. And I don’t have the answers to a lot of questions either. And especially such operational questions like this, I mean, they will typically follow.
But it is something that we would need to have an answer on by the time that this gets bigger and actually gets into a product. And don’t miss saying, right, everybody’s talking about blockchain today, but actual large volume product implementations beyond Bitcoin and some alternatives, they’re not there yet. Did I answer your question? You just talked about IoT.
I know that you’re doing some research on drones. So could you tell us a little bit about how and why you do some research on IoT stuff? The answer is very simple, because IT has the potential to basically change the way our customers do their business tremendously. The smart red trap example I just used, I mean, that’s the perfect example on how to disrupt the business.
And the only way that was made possible is through IoT. So basically, we have IoT sensors in their traps that are connected to 4G or 3G or whatever, and sends their signals into the IFS IoT products get analyzed and turned into a work order. I mean, the business potential is huge.
We have a customer called Songa Offshore. They are basically an oil drilling company, right? So they have the big oil rigs down in the North Atlantic Basin. And their business case is quite simple.
They have to shut down the oil rig for a single day to do maintenance that’s costing them 750,000 euros per day. So even if through a relatively simple IoT product project, they can connect sensors on that oil rig to move the model from a preventive maintenance model into more condition based or even predictive based maintenance model. And with that, they are able to reduce the number of malfunctions and at the same time, make the maintenance windows laying further apart.
I mean, if they can save per oil rig, and they have about 15 of them per oil rig, if they can save one day of maintenance per year, well, do your math, right? That’s a lot of money. So and that’s IoT. And to answer your question, IoT is not so much research in IFS labs anymore as actually being a new technology.
I would say IoT for us has become an enabler, right? I mean, seven years ago, we were researching cloud in IFS labs. Today, I don’t research cloud anymore. For me, it’s there, right? It’s an enabler.
I use it a lot in my projects. But for us, it’s not something we do specific research into anymore. The same with IoT.
It has moved on. It has become a product. We have a product team building that further.
But for us, as IFS labs, it’s an enabler. The fact that we have IoT, I’m already moved on a little bit beyond that, although I realize a lot of customers are just starting to implement that. Did I answer your question? Yeah, absolutely.
Maybe I can go a bit further. If you say IoT is there and no novelty for you, as an IFS research project, right? There’s still a lot to do about any part of IoT. But what are your projects, your areas where you focus your thinking on in the ERP area at the moment? Other projects we’re doing.
Let’s take digital twins. Who knows what a digital twin is? Two fingers. Very good.
So, if you’re talking about IoT as an enabler and the next steps we can take. A digital twin is basically today perceived by the market as a 3D visualized, not necessarily, but in most cases it is, but at least a 3D model showing the exact behavior, the exact state that the actual physical asset is as well. So, imagine a physical asset with a lot of IoT sensors on top of it, all broadcasting their signals into the 3D model.
So, basically, on your computer, you can see the 3D model behaving exactly as the real asset is there. The promise of this technology or this concept is that if you take the Deep Horizon disaster back 10 years ago in the Caribbean Gulf, there was a lot of things going wrong on that oil rig. And the guys on the actual oil rig were unable to determine what went wrong.
And the promise of digital twins is the guys on shore would have had a complete digital version of the oil rig with all the sensors behaving exactly the same as the actual oil rig, including the ability to rewind and do analysis and do some simulations. They would have been able to prevent the Deep Horizon disaster from actually escalating and therefore would have prevented the Deep Horizon from sinking and therefore the entire environmental disaster. So, if you look about doing it like this, yes.
So, if you look about taking a 3D model for analysis that’s completely representing IoT-based signals. That is something what a lot of people in the industry are doing. And that’s something that’s interesting for ERP as well, right? Because you can serve as a model inside of your ERP system.
And if you’re in your work order planning system, you can actually see the asset and you can do analysis and you can do big data analytics and stuff like that. But for us, that’s not really a digital twin. A digital twin contains basically all the information that a asset should provide.
And then if you look at the promise that ERP brings to that. So, if you look at purchase orders and being able to trace back who delivered a certain part and see the behavior over different suppliers of those parts over different digital twins. Bringing the maintenance history.
Maybe bringing all the health, safety, and environment reports. Bringing all the risk and compliance. And there are all the audit findings I did on the asset, on the work orders, on the people that worked on it.
So, being able to create this one centralized view of all the information you can possibly think of together in a digital twin of an item. That would have a lot of benefits towards analytics, decision making, predictive maintenance. So, that’s, for example, an interesting thing we’re looking at.
A second thing we’re looking at could be, for example, autonomous assistance, right? We all know them by now. Everybody ever ordered a pizza through Siri? Nobody. But we all know Siri, right? And we can order pizza through it.
But nobody is using it. So, autonomous assistance are very interesting because potentially if you can implement it in the right way, you can actually help the end user of an ERP in a very big way. And I’m talking beyond the hype, right? I’m not talking about ordering pizza through Siri.
I’m really looking beyond the hype. What if, imagine that you’re standing in front of a nose gear of a Boeing 777, to stick in the same example. And I want to know all the predictive maintenance orders that are related to ideonics of the nose gear that I look at or hydraulics of the nose gear that I look at that are due in the next seven days, right? If I would say, okay, IFS, give me all open predictive maintenance orders for the next seven days for all the hydraulics of the asset that I’m standing in front of.
That takes me 10 seconds. If you would have to type this into a query or even in a very advanced modern UX, it would still take you at least 30 seconds to a minute to find out this information. And then I actually have to use my hands, even if it’s a tablet, right? I still need to use my hands doing my work.
But what if this autonomous assistance could actually take my speech, translate that into a query and provide me all my augmented reality HoloLens and list of those work orders in line of sight so that I can still have my hands inside of the hydraulics gear. That could potentially be a big disruptor for the way that we interact with our business system, with our ERP. Things like that we are investigating quite a bit, which is our wearable devices and autonomous assistance.
Drones is something all the way around. We as IFS, we’re not a drone company. We do cool stuff, but we don’t build our own drones.
We don’t build drone operating systems, but our customers actually use drones or could potentially use drones quite a bit. So we did do research around what do drones mean for our customers and what does this then mean to a business application. So we have many students here who are looking for a job.
So IFS is looking really innovative in the way we lab. So what do the students have to bring or how can they get to be part of the IFS labs? Well, IFS in general is always looking for good people, both in the R&D departments. Any more questions to Rasmus? It was a fantastic time and I have seen a lot of blockchain presentations so far, but I have not seen an ERP researcher who was capable of presenting it in such a complex and deep way that you did today.
So thank you very much for the presentation. Thank you very much for coming to us in our lecture for your time and all the best with your research. Thank you.
And if I would like to add with one thing, you can follow us if you want to on blog.ifsworld.com. We like to blog there about the things we do if we got the time for it and not actually researching. But just just my advice to you guys in your career. And for that, I would like to quote Mr. Edward Lent.
Edward Lent was the inventor of the Polaroid and everybody had that. And he basically said early in his career, an essential aspect of creativity is not being afraid to fail. And this applies to IFS labs.
This applies to myself. And this would also be the advice I would like to give you in your career. If you got a great idea, if you want to start working for a company or whether you would like to build your own company, being a startup, if you have a great idea, don’t be afraid to fail and just do it.
Thank you. My name is Axel Winkelmann. Ich bin Professor für Betriebswirtschaftslehre und Wirtschaftsinformatik an der Universität Würzburg.