Increased safety and efficiency thanks to 3D digital twin at NS

The digital twin enables the station managers to take automated decisions based on data. This marks the first time that a 3D model of a building of this size has been linked to a real-time data feed, and the project has received international recognition as a result: at the IoT World Congress 2023 in Barcelona, this digital twin station received the award for best IoT project. If you're eager to find out how we made this happen, you can read all about it in this client case study. 

NS is the largest public transport provider in the Netherlands and aims to deliver easy-to-use, quick, safe, and affordable rail travel. Aside from that, NS is also responsible for the maintenance of its rolling stock and the development of stations and the surrounding areas. 

NS wanted to gain more insight into what was happening at the station and was lacking an overall view of all assets at its station sites. The aim was to create a single, plug-in-ready platform that provided access to technical and operational IoT data from intelligent assets at station sites in a generic manner using standard Azure Cloud technology. 

Before AMIS Conclusion became involved in this project as a partner, NS already had a 3D model of Utrecht Centraal. To build this model, it obtained and processed no fewer than 20 billion fine-scale LIDAR data points and 540 high-resolution drone images. By adding the real-time status and information from lifts, escalators, and passenger flows to the 3D model, AMIS Conclusion ensured that it effectively became a digital twin. 

At NS, Microsoft Azure is the target platform for the analysis of non-customer-oriented data. As such, we started by setting up a sandbox into which data was loaded on a one-off basis, before developing the analysis environment in R/Python. Next, we created a production environment, with external data flows entered automatically via IoT Hub and Event Hub. Following a check by Azure Functions, that data is then directly moved into the digital twin model. We also produced output analyses, which are made available both internally and externally. 

Using the 3D digital twin, the station managers get instant insight into the status of the station, enabling them to take action as soon as necessary. The model is also capable of making connections between data-producing projects within the building, meaning comprehensive analyses and predictions can be generated. Using this output, NS is able to improve safety and traveller comfort, and prevent defects at the station. 

Jean Keultjes, Digital Lead at NS Stations IT, tells us more: 'The model enables us to easily obtain a detailed picture of the current structure of the building, including the real-time use of objects in the building. .We are now able to do so without needing to check a huge number of individual sensors using separate dashboards That's proving to be really valuable to us.' 

Robbrecht van Amerongen, Head of IoT at AMIS Conclusion, adds: 'Using the model, it is not only possible to identify any defective equipment, it is also possible to simulate the impact of any breakdown via a range of scenarios. And that makes it possible to make better decisions regarding the best walking routes for passengers, and any impact on safety, maintenance, and sustainability.' 

The digital twin model delivers drastic improvements in the overview station managers have of what is happening in their stations over the course of the day. The system also enables NS to schedule lift and escalator maintenance in a way that minimizes disruption to passenger flows. In the future, the model will also offer insight into energy waste and peak loads, which will help NS reduce its carbon footprint. Using audio analysis, NS is also able to automatically detect any inappropriate behaviour and intervene more quickly in peak situations. 

Real-time data processing is essential to obtain up-to-date insight into a current situation and facilitate instant intervention. After all, effective decisions can only be taken with the latest data to hand. AMIS Conclusion stands out from the crowd with 20 years of experience in linking systems and processing real-time data. In doing so, we're adding instant value to primary business processes, and we're laying the foundations for automated decision making. Using an agile and iterative approach, we're able to respond flexibly to the daily dynamics of an organization, delivering tangible results in just a matter of months.