3D Digital Twin for Utrecht Central Station

NS

AMIS Conclusion helped NS activate the real-time data feed for the digital twin of Utrecht Central Station. A digital twin is a digital copy of reality - including all relationships - that changes in real time as the physical world changes. With this digital twin, NS improves safety and passenger comfort and prevents equipment failures at the station.

It enables station managers to automate decisions based on data. This is the first time a 3D model of such a large building has been connected to a real-time data feed. The project has also received international recognition: at the IoT World Congress 2023 in Barcelona, this digital twin of the station was named the best IoT project. Curious about how we achieved this? Read on to find out in this customer case study.

NS is the largest public transport operator in the Netherlands, focusing on easy, fast, safe and affordable rail travel. NS also maintains trains and develops stations and the surrounding areas.

NS wanted greater insight into what happens at the station and lacked a unified view of all assets at station locations. The aim was to create a single plug-in-ready platform that makes technical and operational (IoT) data from intelligent assets at station locations available in a standardised way, using standard Azure Cloud technology.

Before AMIS Conclusion joined the project as a partner, NS already had a 3D model of Utrecht Central Station. Creating this model involved collecting and processing twenty billion detailed LiDAR data points and 540 high-resolution drone images. AMIS Conclusion ensured that real-time status updates and information about lifts, escalators and passenger flows were fed into the 3D model, effectively turning it into a digital twin.

Within NS, Microsoft Azure is the target platform for analysing data that is not customer-facing. We therefore started by setting up a sandbox for a one-off data import and configuring the analytics environment with R/Python. We then created a production environment in which external data streams are ingested automatically through IoT Hub and Event Hubs. After validation by Azure Functions, the data is passed directly to the digital twin model. We also developed analytical outputs that are made available both internally and externally.

The 3D Digital Twin gives station managers immediate insight into the station’s status, allowing them to intervene when needed. The model can also establish relationships between projects that generate data within the building, enabling detailed analyses and predictions. This helps NS improve safety and passenger comfort and prevent equipment failures at the station.

Jean Keultjes, Digital Lead at NS Stations IT, explains: “The model allows us to easily map the building’s current structure in detail, including the real-time use of objects within it. We can now do this without having to consult separate dashboards for a vast number of individual sensors. That is very valuable to us.”

Robbrecht van Amerongen, Head of IoT at AMIS Conclusion, adds: “The model can detect equipment faults and simulate the impact of failures through different scenarios. This allows us to make better decisions about passenger routes, safety implications, maintenance and sustainability.”

The digital twin model greatly improves station managers’ understanding of what happens at a station throughout the day. NS can now also schedule maintenance on lifts and escalators to minimise disruption to passenger flows. In the future, the model will also provide insight into energy waste and peak loads, helping to reduce the carbon footprint. Audio analysis enables NS to automatically detect undesirable behaviour and intervene more quickly during peak periods.

Real-time data processing is essential for understanding the current situation and taking immediate action. Only up-to-date data enables you to make the right decision. AMIS Conclusion stands out with 20 years of experience in connecting systems and processing real-time data. We add immediate value to core business processes and lay the foundation for automated decision-making. Our agile and iterative approach allows us to respond flexibly to the daily dynamics of an organisation and deliver tangible results within a few months.