Vestas: Improving Manufacturing Sustainability
00:00 All right. Welcome. Thank you for having me. My name is Leslie Lundberg. I'm the head of digital engineering at Vestas. And I will be doing a presentation today on the topic of digitization in manufacturing for a sustainable future. So to get things started, I would just like to to take a look at how is it that businesses have been traditionally run and what has been the main focus on running a successful business in the past?
00:32 And typically you will see things like revenue, earnings, market share and of course, great products in the center of all of that. But in order to be successful in the future of business, you have to bring in the element of sustainability. And it's simply something that's becoming a a requirement for a successful business in the future. And it's also something that is gaining traction with some of the younger employees that we are hiring at today.
01:03 It's something that is important for them to to have as a strategic element for any company as one of the core reasons for why they're joining a specific company. And hopefully what I will be able to show to you today is that sustainability doesn't have to be a burden on an organization. It can also be an opportunity to combine both competitive products and sustainability sustainability in one big cohesive strategy. So basically what I'm trying to say is that if you think about sustainability
01:36 and you handle it in a smart way and you are able to leverage the aspects such as configuration, digitalization and modernization, you can really get the best of both worlds. And hopefully I will be able to inspire you a little bit about what it is that we have done at Vestas to essentially bring forward that strategy. Before we get started, I would just like to tell you a little bit about what Vestas is as a company. I don't know how many are you familiar with our company
02:06 and what we have done in sustainability overall. So before I get into some of the technical details, I'll just give you a quick introduction to what Vestas is and what it is that we have done. So for those of you who are not familiar with it, this was actually voted the most sustainable company last year and this year we have also come in second. And within the energy sector, we are still the number one company within sustainability. And why is that? Well, it's because we make wind turbines,
02:39 which are part of the renewable energy sector. We have a company that is about 29,000 employees, and we have, throughout the course of the last 30 odd years, been putting up around 54,000 turbines, which is corresponding to around 150GW of power at this point in time. We are also servicing a tremendous amount of turbines as well, and we have an annual revenue of around €15 billion per year.
03:12 What is perhaps much more interesting is that with the turbines that we produce and ship in 2001, we were able to avoid approximately 532 million tonnes of Co two, which is the equivalent of 125 million cars, 97 million households, or the carbon sequestered by a really large amount of trees as you can see. So by very nature we are, you can say, a sustainable company.
03:45 So it is, of course, one of the factors that played a vital role in our nomination as the most sustainable company. But it's also about taking on challenges such as how do we deal with the resin and glass fiber materials that we have in our blades? How do we recycle those? And we have actually made some significant progress within that area that has allowed us to now show a strategy for how we can actually recycle our blades, essentially extract the resin from the blades and make a turbine that in 2030 will be fully recyclable.
04:19 And our goal is also to make sure that our entire value chain is also sustainable and meeting our sustainability targets. And we essentially aim for having that in place in 2040. And basically this is the summation of what our goals are as a company. And we are aiming to, as I just said, have fully recyclable blades,
04:49 90% increase material efficiency and a 50% reduction of supply chain waste. And essentially all culminating in a combined goal of a zero waste wind turbine by 2040. And that also means that by that time, we have overcome some of the core challenges that we are facing as a company now and some of the things that we're still being criticized for, namely, how do we deal with, for instance, the the
05:22 essentially the plates that have been leaving us with no other choice but putting them in landfills and so on and so forth, we actually can recycle those. Last but not least, as part of the introduction, I have shown a graph of how is it that wind turbines compare in terms of carbon footprint compared to some of the more traditional energy sources. Just to give you an idea of how much better are we doing overall?
05:56 Having said that, what you guys are probably here for more than learning about what this is, is a company and what it is to be doing. What is it that we are doing specifically to actually achieve some of the goals in terms of what this conference is about in which is of course a topic of PLM and digitalization? And in order to answer that question, I've listed a couple of bullet points that as part of our digitalization strategy. So basically what we are trying to do is we are trying to leverage PLM in such
06:30 a way that we can track and measure the carbon footprint footprint of our products. And I'll talk a little bit about that in some of the one of the future slides. What we also want to do is we want to allow customers the choice to configure based on carbon footprint. So basically, if we are in control of our base materials and we are also in control of our raw materials, just like cost, carbon footprint really becomes about roll up of that total cost of footprint.
07:04 And just like you can imagine that you are offering a specific customer to buy a specific configuration of your product based on a preference regarding a certain sub manufacturer that is perhaps a little bit more expensive. You can make those choices available as configuration options. And our idea is to essentially do the same thing with carbon footprint. So for instance, if you have a piece of sheet material or a piece of steel,
07:39 you can track through the manufacturing part numbers or the offsets. Where is this specific piece of steel sheet metal coming from and how was it produced? And that allows you to do a segregation of what are the carbon footprints from each of these manufacturers based on their choice of production methods. Similarly, you can do the same thing for the logistics. So are you if you are, for instance, as we are investors producing one of our towers or nacelles or blades
08:11 or whatever it is at our home country of Denmark? Well, if we are importing a piece of sheet metal from Sweden as opposed to, let's say, China, that's of course also a different carbon footprint associated with where we are transporting them from. But that's all part of the calculation and that's something that you can just build into the logic of how your PLM system works. And essentially that brings me to to my two final bullet points
08:42 on this slide that essentially we want to treat the role of carbon footprint just like cost. And we want to enable the above through configuration of our supply chain. And that essentially boils down to our catchphrase here at the bottom that we want to create sustainability through transparency. Because if we can bring the transparency in our PLM systems about how all of this is working, we can make this a conscious choice and something that can actually be
09:12 a configurable option as opposed to just the passion of you can say, an individual person or whether it's a designer or a customer. It's something that becomes a selectable choice or or a logical choice. So what are some of the core elements of how we actually did that? So first and foremost, if you want to have a strategy for sustainability,
09:42 you got to get in control of your base materials. Your base materials are really at the core of what you actually put into your design. So that's step number one. And the way that we have done that is essentially just by well, not just but but by enabling or integrating Ansys into our PLM landscape. And that's where we have put in all the base materials. There's also an option to put in your raw materials into grants.
10:13 It's capable of handling that. But we have made the conscious choice of not putting our raw materials into grants that we have instead put that into Windchill and then instead link the base materials to raw materials. And we had a lot of internal debate and also discussions with Ansys and grants about whether we should go with one strategy or another. But the reason for why we wanted to put our raw materials into Windchill was because Windchill does really well through partnering in terms of enabling
10:46 the description of your raw materials with classification and dimensional attributes, which allows you to better reuse those raw materials. And secondly, we also wanted to make sure that when we implement MPM link, we can also say, well, these raw materials go into MPM link and then our manufacturing units can be held accountable for actually the amount of waste that's associated with each raw material. And that's the reason for why we have made a very conscious choice about segregating how we handle both base materials
11:19 and raw materials as data objects. Similarly, as part of our overall MPD strategy, we also wanted to transition from 2D to model based definitions as a strategy in itself, but also because we wanted to get away from actually showing our material designations on the 2D drawings as what I usually call dead text, but an actual data attribute that's just coming directly from our Windchill system,
11:50 such that we can also have the option of changing materials on the fly if we wanted to, or again handle the raw materials from different vendors as something that is a possibility to configure in our value chain. Again based on our choice or our goal for carbon footprint and. Last but not least, the base materials. Are connected as data objects both to parts, semi-finished parts,
12:23 raw materials and Windchill. So we have the whole connectivity and as I said, we use parceling for classification and assignment of attributes and hence reuse of the the base materials. And that's of course something that is also associated with our standardization and our modernization strategy at Vestas. And it's also, of course, something that allows us to simplify the whole setup in terms of making sure that we have the same standard base materials, same standard parts. It just makes
12:54 the whole management of our entire product structure and all the data that's associated with it much simpler. So in a perhaps more intuitive and graphical form, I tried to sketch, you know, high level how the whole setup works. So on the left side, you see the centralized base material database, as we call it, aka the grantor database, which is where you can say we have the the central database for all
13:25 the base material information. And we have migrated both base materials, but we are also in the process of migrating surface coatings and in that sense also enable change management of all of those data objects so that once we have put in a base material, there is only a dedicated amount of administrators that actually have the authority to change those base materials. And also the same thing applies for our surface coatings. Similarly, we also putting in test data both our the steel we use for our towers
13:59 and castings. And the goal is to essentially put all test data that are related to base materials into the granted database. Moreover, we are also connecting what was previously a somewhat disconnected sustainability database that keeps track of our carbon footprint. It's essentially a database that we are using at the moment to keep track of our carbon footprint for the entire product. But the idea is to essentially get rid of that one in the long run
14:31 and essentially have everything run in Windchill, as you can say, the long term strategy. But it's there as a temporary solution. And we have, as a first step, just connected the two databases together. Essentially. We then connect everything else through various gateways and APIs, depending on what is necessary to various simulation tools. For instance, we have our own proprietary simulation design tools for our blades,
15:01 and whenever they then designate a specific glass fiber material or carbon fiber material to our base design, it's coming directly from the granted database. Similarly, we have also connected other tools like Ansys, which of course use a lot of different materials such that you are not keying in in your usual fashion, or essentially just selecting for whatever you can say material library that is already built into Ansys, but you instead picking from
15:32 the base material library that's connected through grant. And as you can hear from the conference, it's perhaps not coincidental that Ansys eventually bought grants and that grant is now part of the Ansys family. Last but not least, we have connected the granted database to both Windchill and Creo. So whenever you align a material in Creo, it's essentially connected through a gateway to grantor.
16:04 At the same thing is to be said about Windchill, which means that we have one big connected, you could say, our digital thread for how we handle base materials and how we link it to our parts, our raw materials, our semi-finished parts and so on and so forth, as mentioned in the previous slide. And then the goal is, of course, also to take that information. Now that we have digitalized all the information, we can start to do a lot of fun stuff with it. And we are building currently working on building some Power BI reports
16:37 for our purchase department and also our financial groups, which allows them to say, how can we do hedging on materials? We buy a lot of steel, we buy a lot of glass fibers. So how can we bring transparency to those parts of the organization that have those financial interests and really create some foresight? And now that we have everything is digital attributes, it also becomes much easier to do the crystallization of those materials so that their job of hedging becomes much easier
17:08 and they have to deal with less complexity. And and the same thing, of course, applies for a carbon footprint, as I said at the very beginning. Whether you do a roll up of costs for those hedging activities or you do that for carbon footprint is essentially somewhat of the same exercise is just fulfilling two different purposes. So what does it actually look like? Well, I brought in a very simple example of how our drawings looked in the past.
17:41 So you can see on the left hand side, there's an example of a traditional. Yeah, it's like a sea beam of some sort. And you can see how the usual drawing header would look for. I would assume most companies out there still where you have the actual description of the of the material, you have references to the certificates and what kind of impact toughness, whether it's normalized, all of those types of things that are relevant for both
18:14 the base material, but also the raw materials in terms of how they're manufactured. All of that information is now fully digitalized, either described in the granted database or in Windchill. And that allows us essentially to display that information directly in the Windchill. And you can see through the relevant numbers on the two sides of the slide how we have replaced the dead text on the drawings with digital attributes that are readable directly in Windchill. So what is what are some of the perspectives
18:46 for where we want to go long term? Well, as I said at the beginning of the presentation, if we can then link the supplier parts to the parts or raw materials we have in Windchill, we can make it a conscious choice about configuration. We can take the carbon footprint per supplier and therefore also give both our suppliers and our customers and our engineers an incentive to really work on how they can reduce their carbon footprint
19:16 because they now know that this is a conscious choice of configuration rather than something that we have to chase them manually somehow. And we can configure the value chain as per the customer selected target. So depending on what our customers have as their targets for carbon footprint, we can make it a conscious choice as part of our sales configuration. And if needed, we can also allow price differentiation based on configuration of the supply chain, because it's perhaps not unreasonable to think that at least in
19:48 the beginning, there's going to be an additional price tag on associated with the investments that come with producing steel with electric arc furnaces as opposed to cold furnaces. So of course, some of those investments need to be paid off, but we can help make or create visibility about what is the cost, how much am I paying as a customer to help this transition. And then at least it becomes a conscious choice for you as a company and how you want to brand yourself.
20:21 So in conclusion, what we have been trying to achieve at Vestas is through transition provided by our PLM system. We can empower individuals, both engineers, designers and customers, to make informed decisions that contribute towards a sustainable future. Yeah. So with that being said, and I will open up the floor for questions. On the third slide, you mentioned something about blade recycling and you use
20:54 the term commercially and technically. Yeah. Could you explain what you meant by that? Yeah. Could you repeat the question? I have it. On your third slide. You talked about blade recycling. You said both commercially and technically, I think. And just kind of help me understand what the difference are or kind of what that means. Yeah. So, so that particular side is a is a corporate slide that I did not create, just to be clear. So
21:25 but I can give you the answer as to what is it that we are trying to achieve. And basically when we take an entire turbine and you completely disassemble it after, you know, end of life, most of the steel can of course just be remelted and refurbished for use. But when it comes to our composites structures such as our blades, we've had a real challenge in terms of how do we recycle that? And for those of you who follow Vestas in the news,
21:57 you will you will know that we have just received a number of patents and also worked with a big Swedish company on how we could essentially work out the chemistry for basically separating the resin and the glass fiber so that we could reuse both. And we have actually succeeded with that. So I think the intention behind the somewhat, you know, political choose of choice of words is,
22:27 is that it does still things that need to be worked out in terms of some of the details. But we know we have the technology now in place that allows us to achieve this goal. It's just a matter of when will everything be ready so that we can commit to what is the exact timing. But it's coming sooner rather than later. And we feel quite confident that we have essentially cracked this nut and that we have solved the final piece of the puzzle in terms of how we can recycle an entire turbine.
22:59 Yeah. So the respect to the suppliers, right, that you mentioned, how do you capture the CO2 emission data or the carbon footprint data from the suppliers? The components that you actually source from the suppliers? Yeah. So it's a really good question and it's a question that I get quite often. So let's imagine for a second that we weren't doing the segregation of suppliers, but just looked at what is the amount of carbon footprint that's associated with
23:32 producing a kilogram of of sheet metal, for instance. You can do some analysis to figure out what that is. But in in that calculation, you also have to make certain assumptions about what are the types of production methods that you're using. For instance, it could be a coal furnace and then you can do those calculations and you can probably verify with some accuracy that your carbon footprint and so on so much. And then you have another supplier that's maybe producing it in the exact same way. And there will be some variation in the carbon footprint
24:04 because the production setup could be slightly different. The same thing applies with the logistics. They could be placed far apart depending on where your factory is, of course. But our idea is to say, well, when we specify in our material specifications for when we purchase it, our intention is to also make certain specifications for how it's being produced. And then you can say there's some inaccuracy,
24:34 inaccuracy in terms of how that number is arrived. But we can certainly say that if we can specify that this needs to be produced with, say, renewable energy or electricity rather than coal, we can have, you can say, a somewhat accurate estimate of carbon footprint, but it's not to me, it's not important whether or not we get it spot on. Obviously, the carbon footprint will be significantly different depending on your choice of production methods, and that in itself should be enough to drive what we are hoping to achieve,
25:08 which is to change how we run our businesses and create incentive for everybody to make the right choices. Yeah. Anything else? It's obviously one thing that for like new parts for replacing you said the dead text with the the graph of material database. But did you have a lot of problems dealing with your carryover parts that were
25:38 currently productive and still in that old format? Was that a big effort to clean those out or how did you deal with that? Yeah, so. So, so to be completely honest with you, we have not updated all legacy parts. So so we have to accept that there are some parts that you could say are not transition yet. But when you take an entire turbine, we have some really big components like the hub structure, the steel, the steel tower and so on and so forth.
26:09 And they make up maybe 90% of the total mass of like our top 50 components are the most expensive and also traditionally the heaviest ones. So if you can crack the nut on those ones and make this transition, we have solved 90% of the problem and the rest is just a matter of prioritization. Where do we draw the line between also doing the right thing for our legacy stuff versus all the new designs that we make? Thank you.
26:45 Question about the material database and how do engineers kind of interact with that and find the material they need and get the information into their design? Yeah. So that's like a web interface to, to, to, to grant. So whenever you have to assign a material, essentially a small web interface type of guy shows up and then you select from the database and at least from from my opinion, it's how the data is structured is really intuitive.
27:17 So it's usually very easy to find the right material. At least we haven't had any body raise, any concerns about the ease of use. So it seems to be pretty straightforward. Yes, I just had a question about how you're representing those objects in Windchill. So you'll have a base level object that will represent effectively the chemistry, Right. The the, the spec for the material. Yeah.
27:47 And then it will have a one to many relationship with your form factors. Is it bar, is it sheet, is it plate and so on. And then you'll consume those in the, the actual part definition. Is that the intent? Exactly. So so so essentially, as I was trying to explain in the presentation, your base material are only you can say carry the information about you can say what you would call the chemical or mechanical composition of your material.
28:18 So whether it's it's hot rolled or cold rolled, that's on the raw material part. But the base material information is actually linked to both, you know, the raw materials, semi-finished parts, the parts so that we have the data connectivity throughout the whole thing. So I guess. Is it is it kind of a linear relationship? So art form factor based material. Or is it kind of a simultaneous relationship ? Our based material. How does it.
28:52 How does it interact? Yeah. So, so, so I would I think I would describe it as a linear relationship in the sense that you have your base material that is then connected to your raw material, which is then connected to your part. Yeah. Yeah.