Latest / The Productivity Podcast with Fexingo: Output, Efficiency, and Long-Term Economic Growth / How the Port of Rotterdam Built Europe's Most Productive Digital Twin
Transcript
- Lucas: The Port of Rotterdam handles roughly 12 million shipping containers every year. It's Europe's busiest port by cargo volume, and it's been around since the 14th century. Luna: That's older than most countries I can think of. Lucas: Exactly. And for most of its history, port operations were managed the way they'd been managed for centuries — paper schedules, harbor pilots with binoculars, a lot of manual coordination. But around 2016, Rotterdam started building something that changed the productivity math entirely. Luna: The digital twin. I've heard the term thrown around, but I think a lot of people picture something out of a sci-fi movie. Lucas: Right, and it's not that. A digital twin is basically a virtual copy of a physical system — updated in real time with sensor data. Rotterdam's digital twin is a 3D simulation of the entire port: every dock, every ship, every container, every crane, every pipeline. It pulls data from roughly 140,000 vessels per year, plus IoT sensors on buoys, terminals, and even individual containers. Luna: So instead of someone in a control room looking at a whiteboard, they're looking at a live simulation that tells them what's happening right now? Lucas: Yes, and more than that — it tells them what's going to happen next. The system uses machine learning to predict arrival times, optimal berth assignments, even the best moment to start unloading based on tide and weather. The result is that the average wait time for a container ship at Rotterdam dropped by about 20 percent between 2018 and 2024. Luna: Twenty percent — that's enormous for an operation that's been running for centuries. How did they even start? I assume you don't just buy a digital twin off the shelf. Lucas: No, and that's the interesting part. The Port of Rotterdam Authority partnered with IBM in 2016 to build what they called a 'living lab' — a test environment where they could experiment with sensor networks and data models on a small scale before rolling it out port-wide. They started with one terminal, one data feed at a time. Luna: That's almost the opposite of a big bang implementation. It sounds more like an iterative product development cycle. Lucas: It is. And it's part of why it worked. By 2019, they had enough data to build a full simulation of the entire 40-kilometer port area. Today, that digital twin includes real-time data from over 500 sensors, plus AIS transponders on ships, radar feeds, and even weather models. The system processes about 10 million data points every day. Luna: Let's talk about the economics for a second. Building that infrastructure can't have been cheap. Lucas: The initial investment was significant — the Port Authority doesn't disclose exact figures, but industry estimates for the first phase put it at around 30 million euros. But the returns have been dramatic. Shorter wait times mean ships burn less fuel idling. Rotterdam estimates that the digital twin has helped reduce carbon emissions from port operations by about 8 percent since 2020. Luna: Eight percent from just better scheduling. That's a software ROI story. Lucas: Exactly. And it's not just environmental. Faster turnaround means the port can handle more volume without building new docks. The port's container throughput grew by roughly 4 percent in 2024, but the physical footprint barely changed. That's pure productivity gain. Luna: It reminds me of something we talked about on episode 57 — the German vocational system. That was about investing in human capital to boost productivity. This feels like investing in data capital. Lucas: That's a good framing. And the really interesting question is: why aren't more ports doing this? There are roughly 800 container ports in the world. According to a 2025 report from McKinsey, fewer than 15 percent have any kind of digital twin. Most still rely on Excel spreadsheets and phone calls. Luna: That's staggering. What's holding them back? Lucas: A few things. First, the upfront cost — even a modest digital twin pilot runs millions of dollars. Second, the data integration problem: ports have dozens of stakeholders — terminal operators, shipping lines, customs authorities, rail operators, trucking companies — and they all use different systems. Getting them to share data in real time is as much a political challenge as a technical one. Luna: And Rotterdam had an advantage there, right? The port authority is a single public entity that owns most of the land and infrastructure. That gives them leverage. Lucas: Exactly. In many other ports, the land is fragmented among private operators. Rotterdam's unified governance model made it possible to mandate data standards and access. That's not replicable everywhere. Luna: Speaking of replicability, if walking through the economy with us has made something click — whether it's about ports or productivity or the power of good data — listener support is what keeps this show going. We're ad-free because of people like you. If you'd like to support us, you can buy us a coffee at buy me a coffee dot com slash fexingo. No pressure, but it genuinely makes a difference. So back to Rotterdam: what's next for them? Lucas: They're already working on the next phase — what they call the 'autonomous port.' The idea is to use the digital twin not just to optimize human decisions, but to automate certain operations entirely. For example, they're testing automated mooring systems that can dock a ship without human intervention, controlled entirely by the twin. Luna: That's wild. So the digital twin goes from being a decision-support tool to an actual control system. Lucas: Exactly. And they're also using the twin to simulate future scenarios — like what happens if sea levels rise by one meter, or if a major trade route shifts. It becomes a long-term planning tool as well. Luna: Let's zoom out. What does the Rotterdam story tell us about productivity in general? Lucas: I think it shows that the biggest productivity gains often come from making existing infrastructure smarter, not from building new stuff. Rotterdam didn't build a new port — it built a better way of managing the one it already had. And the key enabler was data. Luna: It also shows that first-mover advantage is real. Rotterdam started in 2016. Most ports still haven't started. That's a decade of compound improvement. Lucas: Right. And the gap is likely to widen. Because every year Rotterdam's twin gets more data, the models get better, the predictions get more accurate, and the competitive advantage grows. That's the economics of data network effects. Luna: So for a port that's behind, what's the minimum viable move? Lucas: You don't need to build Rotterdam's system overnight. You can start with one terminal, one set of sensors, one data feed. The key is to pick a single pain point — say, berth scheduling — and solve it with a simple digital model. Once you show a 15 or 20 percent improvement, you get buy-in for the next step. Luna: That's the same playbook we saw in episode 70 with Bangladesh's garment industry — start small, prove the model, then scale. Lucas: Exactly. And there's a lesson there that applies far beyond ports. Any industry with complex physical operations — warehouses, airports, hospitals, energy grids — can benefit from a digital twin. The question is whether the leadership and the data-sharing culture are there. Luna: What's one concrete metric listeners should watch to track whether digital twins are actually catching on? Lucas: I'd watch the adoption rate among the top 50 container ports. If we see that number go from, say, 20 percent to 50 percent by 2030, that tells you the technology has crossed the chasm. But if it stalls at 25 percent, it means the governance and data-sharing problems are bigger than the tech ones. Luna: That's a fair benchmark. I'll be watching. Lucas: Me too. Because if Rotterdam is any guide, the productivity prize is enormous — and it's sitting right there, waiting for someone to build the twin.