Siemens Mobility Builds on Desiro City Commuter Rail Platform to Digitize Rail Fleets Worldwide on AWS

Rail fleets represent significant capital investments. As rolling stock ages over life cycles that span beyond 30 years, there is a risk of fleets becoming less efficient and more costly. Siemens Mobility is working to create digital solutions that move rolling stock fleets beyond benchmark levels of availability and reliability, promoting sustainability and a future of decarbonized transport solutions.

Siemens Mobility offers a suite of digital services through Railigent, a solution built on Amazon Web Services (AWS). Railigent is flexible and innovative to help asset owners and operators address risks in asset life cycles. Using the Internet of Things (IoT), the solution integrates data from trains, automated vehicle inspections, and maintenance. This integrated data supports a range of applications, from applications with low-latency observability to applications based on powerful machine learning and artificial intelligence, helping maintenance technicians respond and plan proactively. Overall, this capability significantly streamlines the maintenance process by providing virtually immediate fault identification, which in turn increases the overall reliability and life span of the fleet.

Siemens Mobility has been a provider of transport solutions since 1847. In 2013, Siemens won the contract to deliver the British Rail Class 700 Desiro City solution for the Thameslink network. Thameslink provides metro-style passenger connectivity between Brighton, Bedford, and Cambridge through digitally enabled Desiro City trains.

Siemens Mobility provides a range of solutions for digitally connected trains and has been building on AWS to deliver smart train requirements for next-generation passenger services since 2016. Over 170 Desiro City units run in the South of England across a range of network conditions, including automated train operation. The Desiro City fleet has set new standards for commuter rail across and beyond the London metro area. Overall, the UK has the world’s largest installed base of Siemens Mobility commuter rolling stock. Working closely with customers, industry partners, and its innovation environment, the company is focusing on setting new standards with its most advanced fleets while digitalizing trains that are further along in their life cycles. This creates a dynamic of continual innovation and upgrade potential. With its combination of domain and digital expertise, Siemens Mobility is positioned to handle a large volume and variety of operating data, making it valuable to optimize the maintenance and performance of rail fleets.

Working closely with industry partners and customers, Siemens Mobility is integrating new digital capability into fleets throughout the UK network, scaling industry best practices to extend the useful life of these fleets in terms of reliability and cost of maintenance. Digital upgrades, retrofits, and new data-driven services support the industry’s sustainability and innovation goals while supporting iterative improvements in cost and reliability.

“Embedding digital services into our fleets is a key pillar of our business strategy, dovetailing with the Siemens Mobility Alliance for 100% Availability, our mission toward decarbonization, and our commitment to investing in the future of the railway industry in the United Kingdom,” says Simon Rennie, head of digital at Siemens Mobility.

The goal of Siemens Mobility is centered on providing 100 percent availability of trains into service for operators and passengers. To achieve this goal, Siemens Mobility has developed product areas that integrate the digital capabilities from rolling stock, maintenance management, and automated vehicle inspection systems. For example, state-of-the-art laser monitoring and imagery systems take images and measurements of the undercarriage of trains every time they pass into the Three Bridges or Hornsey depots. Through robust Bayesian regression algorithms running on AWS, this data can be trended and combined with train and maintenance data to optimize wheel-set maintenance.

The Siemens Mobility Railigent solution uses AWS best practices to deliver a secure, ISO/IEC 27001–compliant setup for connectivity, storage, and data processing. At its foundation, the solution integrates all data sources into a data lake hosted on Amazon Simple Storage Service (Amazon S3), an object storage service that offers industry-leading scalability, data availability, security, and performance. The solution enables layers of use case–focused capability on top of the secure connectivity and storage layer. Siemens Mobility believes in a data-centric approach to innovation, where quality and deep domain knowledge of the data are as important as raw quantity and availability. Railigent uses Amazon Athena—an interactive query service that makes it easy to analyze data in Amazon S3 using standard SQL—to support a data lake that balances flexibility with structure. This balance in turn supports collaboration between domain specialists, engineers, and developers to create value-focused services. In a connected capability, Railigent uses Amazon DynamoDB—a fully managed, serverless, key-value NoSQL database—to support the aggregation of data sources into operational data stores, turning source-system-specific data into outcome-oriented, use case–focused data.

Railigent applications rely on AWS Elastic Beanstalk, an easy-to-use service for deploying and scaling web applications and services. It provides the technology Siemens Mobility needs to support scalable applications for internal and external customers. This setup is increasingly complemented by containerized workloads on Kubernetes using Amazon Elastic Kubernetes Service (Amazon EKS), a managed container service to run and scale Kubernetes applications in the cloud or on premises.

“Per week, for a typical fleet, the scope of relevant data for a maintenance technician is estimated to be approximately nine million data points,” says Martin Klimmek, head of digital development and operations at Siemens Mobility. “Our mission is to make this scope of relevant data observable, meaningful, and actionable, working backward from the questions our customers are focused on. By building Railigent on AWS, we have the balance of flexibility and structure we need to work with customers on an agile basis.”

In 2014, Siemens Mobility would typically see train data downloaded once a day prior to maintenance. Train systems were set up to generate diagnostic codes based on expert rules encoded in the train’s operating system. The best practice was to look at these diagnostic codes prior to maintenance to calibrate maintenance priorities. Building on the rollout of the Desiro City commuter train solution, Siemens Mobility has invested in transforming this model. Train data is transmitted continually into Railigent, making it possible to generate timely analytics and gain actionable insights. The current standard is based on continuous monitoring and analysis of train data delivered through high-performance products and innovative cocreation projects. The highly reliable digital product set draws from a pipeline of innovative research and development and codevelopment initiatives with customers and industry partners.

Siemens Mobility runs 65 million passenger miles per year across the 500 units it maintains. During the COVID-19 pandemic, the highly digitized hardware and software platform around the Desiro City trains delivered a more granular understanding of passenger needs to facilitate social distancing and capacity planning. These insights continued to be useful as customers sought to rebuild and stimulate new ways of working in the world after the COVID-19 pandemic to create a better passenger experience. As a result of these innovations, Siemens Mobility was able to widen its service offerings to include better capacity planning and more passenger-focused, transparent services.

The company has developed a flexible approach to connect all Siemens Mobility trains to Railigent. Siemens Mobility uses Amazon Athena and AWS Lambda—a serverless compute service that lets users run code without provisioning or managing servers—to structure the data, create data pipelines, and provide access to the data through over 300 customer dashboard use cases spanning maintenance, fleet operations, engineering, and technical analysis. “Each train, wherever it is in the life cycle, has a digital element that runs alongside everything that happens to it physically,” says Klimmek. Siemens Mobility also uses Amazon Managed Streaming for Apache Kafka (Amazon MSK), which makes it easy to ingest and process streaming data in real time.

“In the early days, we batch-processed everything with a minimum complete latency of around 15 minutes,” says Colin McLellan, DevOps engineer in the digital team. “Now, with the inclusion of components like AWS Lambda, Amazon EKS, and Amazon MSK, we are moving toward far lower latencies, pushing our target latency into the realm of seconds rather than minutes.” Siemens Mobility has built a set of tools that lets the company share access to an abstracted view of the data, which makes it simpler for teams to share knowledge across functions in the business. This innovation increases the frequency and timeliness of results, helping teams work together more effectively and efficiently.

Trains that work well mechanically still become expensive to maintain as they age over a 30-year life cycle. The core benefits that Siemens Mobility has realized include increased availability, reliability, and long-term service optimization proposals covering the equipment life cycle. The system integrates data into the business through self-service portals. This helps users achieve efficiencies in resource allocation. Three core sources of data—diagnostic, maintenance, and AVI—are combined to support virtually 100 percent availability. Siemens Mobility’s customers gain sustainability through maintenance interval optimization, collaboration building on transparency, efficient material usage, and a healthier overall train system. All of this adds up to support a better passenger experience, which encourages more people to ride trains instead of using other travel options.

Railigent’s capabilities have grown from connecting the Desiro City fleets from commissioning phases in 2016 when Railigent was announced publicly to encompassing a broad digital retrofit of Siemens Mobility’s rail portfolio in the UK. In the next modernization phase, Siemens Mobility plans to build an even stronger connection between infrastructure and information technology systems. “We’re breaking down the barriers between the train as a data center and the processing system,” says Klimmek. Using AWS managed services is an important element of the company’s digital delivery strategy, helping teams focus on value creation. As an example, the UK digital delivery team was an early adopter of Apache Airflow to orchestrate processing and analytics tasks. It is now working to migrate to Amazon Managed Workflows for Apache Airflow (Amazon MWAA), a managed orchestration service, to gain development and run-time efficiency.

The digital capability developed for the Desiro City rolling stock solution is an ideal reference architecture on how to digitize trains and is being used as an example for digitizing legacy fleets. Klimmek says, “The collaboration between Railigent and AWS and the opportunities that arise from using AWS technology help us make trains as observable, efficient, and reliable as possible for our customers and passengers.”