Latest / The Manufacturing Podcast with Fexingo: Factories, Supply Chains, and Industrial Business / How Digital Twins Are Cutting Factory Energy Use
Transcript
- Lucas: So there's a fascinating story out of southern Germany, a mid-sized auto parts plant that makes precision valves. They didn't buy a single new machine last year, yet they cut their plant-wide energy use by eleven percent. And the trick wasn't a new motor or a heat pump. It was a digital twin of their compressed air system. Luna: A digital twin of compressed air. That sounds like something you'd see in a sci-fi movie, not a factory floor. How does that even work? Lucas: Right, so a digital twin is basically a virtual copy of a physical system, fed with real-time sensor data, that lets you test changes without touching the real thing. In this case, they modeled every pipe, every valve, every compressor in their air network. And compressed air is one of the biggest energy hogs in manufacturing. It can be ten to thirty percent of a plant's electricity bill. And most of it leaks. Luna: Leaks, yeah, I've heard that. Some plants lose half their compressed air to leaks and nobody notices. Lucas: Exactly. And the twin let them see the leaks in real time. But here's the thing, the real savings came from rethinking the whole system. They discovered they were running two big compressors at full tilt even during low-demand shifts. So they reprogrammed the controls to run one smaller unit and use the other as backup. That alone cut their compressed air energy by eighteen percent in six months. Luna: Eighteen percent in six months. That's not a marginal tweak. That's a fundamental shift in how they run the plant. And they didn't have to shut down production to do it, right? Lucas: No, that's the beauty of it. They tested all the changes in the twin first. They could see exactly what would happen if they changed the pressure setpoints or staggered the compressor starts. No trial and error on the actual line. And the payback period for the whole digital twin project? Under a year, just from the energy savings. That's the kind of number that gets a CFO's attention. Luna: I love that. And it's a great reminder that sometimes the biggest wins aren't about buying shiny new machines. It's about understanding what you already have. Lucas: Absolutely. And speaking of understanding what we already have, if these conversations have helped you think differently about your own work or just given you a good story to share, you can keep the show ad-free by supporting us at buy me a coffee dot com slash fexingo. It's a small gesture, but it genuinely keeps us going. Luna: Yeah, every little bit helps, and we really appreciate it. Now, back to the digital twin. I want to dig into what makes a twin different from, say, a really good simulation. Because I feel like people throw around 'digital twin' a lot these days. Lucas: That's a great question. And it's the first thing I always ask when someone says they have a digital twin. Because a simulation is a one-time model. You build it, you run scenarios, you get answers. A digital twin is alive. It's connected to sensors, it updates continuously, it learns from the actual behavior of the machine or system. So it's not just a model of the plant, it's a mirror that reflects what's happening right now. Luna: So it's not a snapshot. It's more like a live feed of your factory's health. That sounds like something that could be useful beyond just energy, too. Predictive maintenance, quality control, all that. Lucas: Exactly. And that's where the trend is heading. The auto parts plant is a perfect case study because they started with energy, but now they're expanding the twin to cover their entire production line. They're already seeing early signs of fewer unplanned stops because the twin flags anomalies before they become breakdowns. Luna: That's the kind of return on investment that goes beyond the spreadsheet. It's about resilience. And I suppose there's a sustainability angle too. Cutting energy use directly cuts carbon emissions, which is a big deal for manufacturers facing pressure to hit net-zero targets. Lucas: Right, and that's actually why we're seeing a lot of interest from industries that aren't traditionally tech-forward. I was reading a report from last month that said the global digital twin market is expected to grow at about thirty-eight percent annually through 2030. And the biggest driver isn't just efficiency, it's ESG reporting. Companies need hard data on their emissions, and a digital twin gives them that level of granularity. Luna: Thirty-eight percent CAGR. That's a massive growth rate. So this isn't a niche tool anymore. It's becoming a core part of how modern factories are managed. Lucas: Yes, and that's a shift from just a few years ago when digital twins were mostly used in aerospace or high-end engineering. Now we're seeing them in food processing, in textile plants, in small machine shops. The cost of sensors and cloud computing has come down so much that a mid-sized company can justify it. And the payback period, like we saw with the German plant, is often under two years. Luna: That's a compelling argument. But I imagine there's a learning curve. You can't just slap a digital twin on a factory and expect magic. You need the right people to interpret the data and act on it. Lucas: Absolutely. The technology is the easy part. The hard part is organizational change. You need engineers who trust the twin and know how to translate its insights into action. That's why the German plant had a dedicated team of three people just for the digital twin project. They weren't IT guys, they were process engineers who learned to work with the data. Luna: So it's not just a software investment, it's a people investment. And that might be the real barrier for some companies. But the upside seems worth it. Lucas: For sure. And there's another angle that's interesting, which is the role of digital twins in designing new factories. Instead of building a plant and then optimizing it, some companies are using twins to simulate the entire factory before a single brick is laid. They test different layouts, different workflows, even different weather scenarios. It's like a flight simulator for a factory. Luna: That's amazing. So you can actually design for energy efficiency from the start, rather than retrofitting later. That's a much smarter approach. Lucas: Exactly. And that's what a company like Siemens is doing with their own factories. They've been using digital twins for years to optimize their electronics plants. They've reported up to a twenty percent reduction in energy use and a similar boost in throughput. And now they're selling that expertise to other manufacturers. Luna: So it's almost like the digital twin is becoming a product in its own right. Not just a tool, but something you can buy or license. That's a whole new business model. Lucas: Right, and that's creating a whole ecosystem of software vendors and consultants. But it also means smaller companies can get access to technology that used to be reserved for the giants. You can now buy a digital twin for a specific piece of equipment, like a CNC machine or a boiler, without having to build an entire plant model. Luna: That's a great point. So if you're a small manufacturer, you can start with a twin for your most energy-intensive asset and expand from there. It's a more incremental, less intimidating path. Lucas: Exactly. And that's the message I want listeners to take away. You don't need to boil the ocean. Pick one system, one machine, one line. Build a digital twin of it, learn from it, and let the savings fund the next step. That's how the German plant did it, and now they're planning to use the savings to invest in solar panels and battery storage, moving toward their net-zero goal. Luna: That's a great arc. From a digital twin to solar panels. It all started with looking at their compressed air system. I love that humility. Lucas: It's a classic manufacturing story too. You watch the pennies, and the pounds take care of themselves. But in this case, it's megawatts. Luna: So, for a plant manager listening to this, what's the first step? Is it just asking a vendor for a quote, or is there something they should do internally first? Lucas: I'd say start with your energy bills. Find the biggest energy-consuming process. In many factories, it's compressed air, HVAC, or a specific heat treatment process. Once you have that, you can scope a digital twin project with a clear business case. And talk to your process engineers early. They're the ones who'll make it work. Luna: That sounds like a sensible, low-risk starting point. And the payback is measurable, which makes it easier to get buy-in from management. Lucas: Exactly. And there are even some government programs in Europe and parts of the US that subsidize energy audits and smart manufacturing tech. So the initial cost can be even lower than you think. Luna: Good point. So it's not just about the technology, it's about being strategic about where to start. And I think that's a great note to end on for today's episode. Lucas: Yeah, I think so too. Digital twins are a reminder that sometimes the smartest investment is in understanding what you already have. And with the pressure to cut carbon and costs, it's a tool we're going to see more and more of. Thanks for listening, and we'll catch you next time.