Shaping Ideas into Shoes: A 35-Year Journey into Digital Shoemaking | Design&Develop | PI Apparel

I'm Alberto, and today I want to talk to you a little bit about our journey into digital transformation with regards to footwear for the last 35 years. But first, I want to introduce ourselves a bit. Our company was founded in 1990 by my dad, of course, in good Italian fashion. This gentleman over here doesn't look like that anymore, but used to. And we've always been a research center focused on footwear. And we have, from the get-go, liked digital technology quite a bit. And this is where we are located in Italy, nearby Venice, in Asolo, which still is a pretty active district for footwear innovation and footwear development. Our business model has always focused on innovation. Innovation as a— throughout the— from the concept to the design, throughout the process for the footwear development. And we've always liked to be supported by the latest and greatest technology. But all of this we have with tradition back in our DNA. I put two pictures here. On the right-hand side, you see a picture of Scarpa, which is a famous shoemaking brand nearby us, and one of those gentlemen over there is my granddad. Again, my dad has always worked in footwear and shoemaking, my brother, my mother, all of them. All in— I've always worked in footwear. And of course, I am a computer scientist and kind of the black sheep of the family. My background is not in footwear. It's more in compiler technologies and computer graphics. But 10 years ago, I joined Design Developed, and we tried to make somewhat of a connection between the two worlds. So the word software and technology in the world of footwear. And today I want to tell you 3 stories of digital transformation, of what we have seen and we have helped and supported in the last 35 years. So the first one I want to talk to you about is about ski boots, something that maybe here in the area is not that much developed, but in our area still is quite a driver for business. And especially in the '90s, there were way more volumes. And it was a very, very big driver for innovation from the hard shell and plastic and aluminum buckles, all of that. And we thought back then that the 3D— sorry, the hard shell of the ski boot was a very good candidate for 3D process to create directly in 3D. And so we wanted to implement this 3D technology inside of ski boots design and manufacturing. The question is, how was it done before? Before, it was done with modeling, with pallets, and glue them together. And then this was then cast, and you would make a mold. So very time-consuming. And of course, if you made a mistake, you had to remodel the whole thing by hand. And it was very time-consuming. And for context, we have to go back to the mid-'90s, and 3D tools were very rough. This is the kind of stuff we had to deal with. So everything on Unix. Windows was not a thing, and very like number-heavy interfaces, not really the most user-friendly interfaces. And so the feedback we got from this was, well, it's faster if I do it by hand. So very much a pushback from the traditional, let's say, wood modeler that they didn't want to switch to 3D. They'd rather keep on doing what they were doing. But after a lot of convincing, we were able to convince the management on the savings, of course, of 3D technology. And today, it would be unthinkable to do by hand wood modeling and such. It gives you much more control on the thicknesses, on whatever you're going to do. On such a complex product. A little bit the result of this. We had a very good reputation after being able to deliver on 3D ski boots. And we gathered a lot of experience early on. And again, we're talking about the mid-'90s. And we were able to upskill our workforce quite a bit to handle this sort of complex data. And because we were early adopter in the footwear industry, this opened a lot of doors with technology providers. So whether it was 3D scanner, 3D printer, we being early adopters of this technology allowed us to have direct communication with the various technology provider. The kind of what we get out of this is that being able to clearly demonstrate how the improvement of the process was done really helped us push through this change, especially with the management. And that kind of bring us to the second story I want to tell to you. And it's about the transition that we supported of a fairly large organization from an old pattern engineering solution that they were using, Crispin namely, to our own pattern engineering solution, which is Javara. And the big challenges we had here were that we had a A lot of users entrenched in this old solution. They were forced to move to the new solution because the old one was discontinued and not supported anymore. So they were not really willing to move. And plus, there was a very large amount of files. And you know it really well, probably. When you're dealing with a supply chain that spans all around the world, you have also linguistic and cultural barriers that are really important, that makes a lot of sense to do some consideration on that as well. But the management supported us quite well in this. There was the will to unify all the process under the same umbrella and all the engineering process. So the tooling engineering, the pattern engineering, all under the same platform, which is Rhino. And we had a strong mandate to change. So the old pattern engineering solution was out the door. The new one, Javro, was the de facto standard. So that helped us quite a bit in pushing this change. And what we did to help as much the changes, trying to listen to the users. This is something that is not always super easy, but to ease the transition, we tried to implement shortcuts and icons that were similar to the old tool. We tried to also go down and make specific feature for specific users because they would have a big voice in the community and then And those people, once they were converted to the new solution, they became evangelists for our own. And everything for, again, having a frictionless transition, the most important thing was doing trainings and again and trainings. And again, this is something that I truly believe and that The trainings are not just a moment for teaching a new tool or a new process, but it's just a great moment of exchange. We call them training, we call it workshop, whatever. And it's really good to get like a feedback and an idea of what's the process today, how can I improve it, how can I get rid of some bad habits, and and to teach to the users that the software is not written in stone. So if they have a problem, they can come back to us and then we can improve that and change the tool. And not always this works, but that's what we're trying to do. And the result is that after a couple of years, most of the users were transitioned to the new solution, so to Javero. And we still have some nostalgic, but that's a fact of life. That's a generational thing. You will never get rid of everybody using an old solution. But we had great support from the brand side. And that helped us quite a bit because that was the clear mandate of where we wanted to go. And this is, again, I can't stress enough how much a vision, an idea of where we want to go, is very important when enacting this sort of transition. And for the last story, I have— we had a fairly smaller group now. Of 3D creators that wanted to improve their process. And we went to them and proposed our solution, Bocha, to improve this process and speed it up. And again, also in this case, we had a lot of strong leadership. This is again, very important when transitioning to a new solution, when implementing a new process. And they had the strong will of implementing the tool throughout the process. So their old upper creation process involved scanning an upper, retopologizing it, and remodeling on top of that. Which is good because it's fairly quick. But it doesn't allow you to have perfect control of the geometry of what you get out of this. And one big problem for them was the fact that you cannot have layers on top of layer because you're just modeling the outer shell of the scanning. And if you have transparency, like a plastic material with transparent features, you need to model all the layers. So the workflow we suggested and the approach we took is how do we then take this and make it faster? Instead of doing a retopology of an existing scanning, why don't you directly build the stuff in 3D? And that's what we— propose, and that's what we arrived at. And we gave them this new way of working that they're going to build their models directly either from the sketches or from the pattern. So they don't go through this scanning and then retopo. But they rather build up all the shoe parts, again, potentially It depends on the sort of output that you want to achieve. And with the tools that they have at their disposal, they can add a lot of rich details with ease, like stitching, lacing, all of that, and having weird effects like welding and all of that very, very easily. And we saw a speed-up of the modeling process of 60%. So what before took a week, 5 days, then went down to 2. So you can more than double the output of the same amount of people, which is quite, quite impressive. And now they also are not just that, but some more complex structures are now enabled and you can build all the— overlays and all the intricate structure that you want without having to model and be a painful journey to model this. And last but not least, because all of our tools are sitting on top of Rhino and most of them have a connection with Grasshopper, they were able to build custom automation for their specific process, for their specific workflow. On Grasshopper. So without asking us, without asking anybody, they did build these automations to, again, speed up even further some repetitive tasks. And a bit the takeaways of these three stories, but not just this, all the experience that we Agatha with transformation. So the question is, how do we succeed? That's what you want. How do we transition properly from one process to the next? How do you implement a new tool? How do you search for a new tool also? And we have seen that, again, search for innovation is hard, but at the end, You may stumble sometimes, of course, but when then you find a way that suits your brand, your company, then the output is always rewarding. And this is something that we have seen time and time again, that the most important thing is the strong vision where we want to go, and everyone needs to be on board with that vision. If we have that, I would say that 60% of the work is done. Because if everybody is saying, we're going into that direction, and everybody is pushing for that direction, things will happen and you will move forward. Whether— you will always have detractors, but those will be left behind. Something that I truly believe in is to be The integration, and again, we have built our whole business model around this to try to integrate everything into less system possible because the interconnection between system usually is where you have the most friction. So this is a real good thing to do to have like an integrated digital pipeline that where all the parts and all the actors in the pipeline can access the data, can contribute to the data. That's a very valuable thing to not just gather speed, but also to enable processes and products that you wouldn't be able to do without a good integration. And of course, last but not least is the training and giving the people space for learning. On new process. Of course, when you change a pattern engineering solution or 3D solution, whatever it is, you're going to have loss of efficiency in the short term. That's a fact of life. People need some time to learn to adjust to the new solution. But if you give them the space, then they will— and you train them and you support them, you can accomplish quite a bit of stuff. Like one of the interesting thing that we notice is getting rid of some bad habits that people have since, I don't know, the beginning of time. That because one software is built one way, somebody builds like a habit, which is wrong habit. During this transition, you have that chance to iron them out. So you can say, okay, this is wrong, don't do that anymore. In the new tool, you're going to do it this way. So this moment of training is very, very valuable. And to finish this, I talked about integration. What we believe for integration and what we offer for integration is a number of plugins on top of Rhino. We call it altogether the Footwork Kit for Rhino. Which includes Javro, which is a pattern engineering solution that allows you to make patterns, grade them, export for cutting machines, and has quite a number of features for all sorts of different workflows. Then we have Javro Cost, which is more for production planning. Assessment of costing and to prepare sort of your FOB and your cost. And the last piece is Botcha, which is a 3D tool for footwear dedicated on top of Rhino, which allow you to work in 2D, have results in 3D, connect them. And create all sorts of rich details with stitching, lacing, last grading, digital texturing. There's all sorts of things. And all of this is integrated within Grasshopper so that then if a computational designer is interested in using this tool and put them in a pipeline to generate a new product, you have that capability. With that, that's all. If you have any question, I'm here. Thank you. Cool. Sure.

Thanks for such a great presentation.

Thank you.

So interesting, and those original wood models were so beautiful.

Yes.

Yeah, how long did they take?

I think anywhere between 1 and 2 weeks.

Yeah, so beautiful.

Yeah.

What does your father think of the change that this company has gone through, right, in the last 30 years?

Well, I— We have always looked at trying to speed up the change. Always been very active in investing and trying new things. Just for reference, for example, we bought our first 3D scanner in '97. We are based in Italy. In Italy, there were 3 scanners. One was like Ferrari, another one was the guy that sold it to us, and then which was us. Amazing. And so, he was always proactive, and that's a bit I think where I also got my passion for computer science because I saw all these computers around when I was a kid, and I decided to go into that career.

That's amazing. We have a question. What's the best way to take the know-how of traditional shoemaking to empower it with 3-day workflows?

That's a hard question. Yeah. Rene wants to ask the hard questions always. Yeah, it's one thing that I think it's very important in this case, it's the mentoring. So if you have like an old traditional footwear engineer, whatever, his role is, and have the new kids that are more and more into the digital things to have some sort of relation between the two people in the company. That kind of then also going through school is a very good, it's a very good, way because it's somewhat easier to get digital know-how because you go on YouTube, you can find, for example, I don't know, Rhino, you find a million tutorial or Blender, same thing. You find like from noob to skilled. While for the traditional shoemaking, it's harder to achieve. So institutions such, I know there's Pencil here, or Italy, we have Arsutoria, for example, another big school. Those are like the best way of getting into, let's say, more traditional. And then inside of the companies, then it's the experienced people that can give that help in training the new generation into the old way. Even sometimes you have to even do the clay modeling or just then to get a feel of what it is live in your hand. And then you can go and build it in Rhino, sure. But as a learning process, that is very, very, to me, interesting.

Yeah, I think all of the big theme that I keep hearing out of today is just the translational knowledge of how do you translate something that's quite complicated, yet small, right? With new technology amongst many different people that have many different abilities, right? And so that's the problem we have to solve. Another question, looking at your users, what are the measurable elements that made the transition from their already integrated sponsorship solutions to yours easier?

Okay, so in general, what we have tried to do is we have tried to either tailor workflows, specific workflows with our tool that sort of resembled their old, solution, or literally going down to implement keyboard shortcuts. Somebody is very used to using the keyboard for doing certain things, and so we try to somewhat mimic the motions of the old tool to the new one. And up to a certain point, you can do that, but then Again, if you want to evolve the process, if you want to have something new, sometimes you have to disrupt things. Because you may, again, go back to bad habits of clicking 100 times when you can do just things in 2 clicks. So that's a little bit like our experience with that.

Yeah. Can you explain your experience with people who are new to the 3D space? Space and have an interest in using these tools? And do you offer training to new and experienced users?

Yeah. So in general, we have a, let's say, learning base on— for example, we have on YouTube, we have a number of videos to go through this. And of course, we do a lot of live training or online trainings. And these are always, as I said before, always like good moment for not just like a typical frontal lesson where somebody's teaching you, you do this, but rather like an exchange moment where people will tell you also their problems. We had, this is something that I found it like not funny but interesting. We had in a training, a pattern engineer told us, well, I have to take my parts and move them around in space. Because then the cutter, the guy at the cutting machine, will be able to understand where they are. And she would move things around every time she would export a pattern and then send it down. And we told her, don't do that. So we made a button that when she export, she can lay them out automatically. And that saves like hours. And this could only come up in an exchange. So that's why we very much like it, especially live. Because online it's a bit, I don't know, detached. And when you know a person, a person after, I don't know, a couple of days that you're training with, that becomes like an easier conversation to have.

Yeah, it's so good working with users, right? Yes. Learning all the problems. What do you think is— you presented these three technologies, the pattern making, and the production, and the 3D. What is like the future for you? What do you think is the next missing gap?

So we are trying to— improve and integrate as much our solution as possible so that you can have a clear line across digital creation, engineering, digital design. And at the moment, we are still catching up with the requests that our users are asking us. But one thing that it's amazing to see, at least for me, is that how people take your tools and then do all sorts of crazy stuff. And because we had this thing of building on top of Rhino, there's so many other people building tools, and they can get together. For example, we don't have the bandwidth to include artificial intelligence in our tool. But there is a Grasshopper component that integrates ChatGPT, Midjourney, whatever it is. Whatever you want to do into Grasshopper. And you can connect that with our tools. And that scenario is something that neither we or the guy that did the integration thought about. But then you can arrive there with serendipity. You arrive at this solution. And it's quite good to see then people coming back saying, yeah, I have this small Grasshopper script. And you see this huge stuff that they have built. Telling you what they have achieved. It's quite cool.

Yeah. What is one of the most surprising things that you've seen done with your technology?

One of our users built an automated workflow in Grasshopper to basically— the designer built a sketch. and that would automatically trace the sketch, create parts, and flow it to 3D, and you had like a 3D shoe from out of a sketch. Not 100% perfect, some constraints, but yeah, that was like, they just showed it to us like, hey, look at what I've done, and it was like quite amazing.

That's cool. And then I think we have one last question. Does shoemaking history in your company translate to easier software adoption by more traditional teams?

Yes, another tough one. But I would say what we have seen, again, the younger people that maybe have already like a Rhino background because they either come from, say, architecture school or more industrial design school, and then they get into either footwear making in general, or 3D footwear making, or pattern engineering, they are very easy to convert. The more traditional members of the team are a little bit harder. And that's what I meant when sometimes you have to implement a feature for one user. Because if that user is happy and That feature was not in his previous tool, but that was something that was bothering him. And this, again, happened sometimes with us, with Javro specifically. A pair engineer was asking us, oh, I want to have a feature that makes a radius out of this, and it makes the biggest possible radius without me telling him, the tool. So we implemented that, and after that, That user was super happy. He did not go back to the previous tool because that was something that was bothering him. So that's kind of the only, I say, way that you have. So that specific feature that they couldn't achieve before, but they can with the new tool.

Fantastic. Well, thank you so much for your time today.

Thank you.

What a wonderful presentation.

Thank you.