Blog  /  Community

How Siemens uses IoT sensor data and Grafana to optimize train maintenance, capacity, and more

Michael Steele

Michael Steele 25 Jun 2021 3 min read


There’s something special about the interactions a train journey generates — the interesting views and perspectives that inspire insights and drive new thinking. 

Martin Klimmek, Head of Digital Development and Operations at Siemens Mobility and Haluk Tutuk, Data Platform Engineer with Periscube, are among 20 data scientists, data engineers, and DevOps engineers building the next generation of data-powered customer service for the rolling stock industry in the U.K. and beyond. With over 3,000 suppliers in the U.K. alone, Siemens prioritizes productive partnerships, an open ecosystem and observability. In their “Creating value through IoT train sensor data with Grafana” talk at GrafanaCONline 2021, Klimmek and Tutuk described how Grafana has helped along the way.  

“Observability is all about control theory,” said Klimmek. “It’s about trying to understand assets better by looking at the data and the signals those assets generate. We really believe passionately in observability and we’re so excited to be on this journey with Grafana, taking it to the next level.” 

When it comes to technology as well as partners, Siemens wants to push the boundaries in building new interactions, both from a technical point of view (in this case, thinking of trains as an extension of the platform), and also in making it easier to work with data across its network of developers, customers, and domain engineers. “We really want to develop a platform where the push is an experimental mindset that makes it easier to bring expertise together,” said Klimmek. He pointed to how Grafana lets his team generate data in real time from the trains to optimize maintenance sequencing.

Klimmek gave several other use cases, one being temperature monitoring and control. “The setpoint management in these HVAC systems is actually quite complex, and Grafana has allowed us to bring domain experts and data scientists together and allows us to visualize the data really quickly and easily,” said Klimmek. “It also allows us to set up really intuitive alerts.” 

Another example is capacity counting (very useful for social distancing during the pandemic). “What we’re really working towards is to be able to understand in fine detail the dynamics of how passengers use the services, how they spread across the trains, and how that varies across the network,” said Klimmek. “Grafana has helped us in terms of rapidly prototyping new features, exploring, calibrating, testing, and then releasing into production quickly.”

In the second part of the talk, Tutuk explained how his team approaches integrating a data center on wheels and reviewed the DevOps design choices they have made. “We have 500 trains in the UK,” he said. “We have millions of data points, every single day coming from all those trains.” Grafana helps foster a collaborative partnership environment that supports observability, drives a passenger-centric operation, and enables the team to better describe, explain, and predict with the data. 

To find out more about how Siemens is using Grafana for train observability, watch the full talk on demand now.