SNCF goes Fully Digital: IoT on Track integrated with Railway Lines

IBM Watson IoT helps keep SNCF French National Railway running smoothly, deliver superior customer experiences, heightened operational excellence and enhanced rail safety.

IBM announced that French national train operator SNCF is leveraging IBM Watson Internet of Things (IoT) technologies to deliver greater customer experiences and heightened operational excellence.

The high-speed rail network in France is something to behold. While city-goers spend an average of 45 minutes traveling to the office, that same span of time is all it takes to travel from Paris to Reims, cities over 125 km apart.

The National Society of French Railways (SNCF) maintains 32,000 km of railways that span every region of the country. And it has announced a plan to integrate smart technology into its railway lines, which could mean enormous cost savings for one of the most internationally renowned transit systems. SNCF serves the needs of more than two billion passengers annually. Underlying the company’s success has been its ongoing commitment to delivering exceptional services to its customers, including minimizing delays, ensuring passenger safety and delivering a superior on-train experience for commuters.

By utilizing the Internet of Things (IoT), the SNCF envisions a railway network that is more energy efficient and less reliant on operator maintenance to function. The organization has dedicated 600 million euros to the project, of which 300 million will be spent over the next three years.

Among the companies partnered with the SNCF for this project include IoT networking company Sigfox, 2G technology and 4G/LTE narrow-band provider Ericsson, and IBM for its PaaS Bluemix cloud platform. The recent start-up Intensens will also participate in the project by providing new IoT sensors. DOWNLOAD HERE the SNCF IOT PresentationSNCF Industrial IOT Presentation.

IoT keeps things on track with superior service

SNCF turned to IBM Watson IoT to help keep things on track, deploying thousands of sensors on trains and tracks which are securely sending tens of thousands of data points in the cloud, all in real-time. By analyzing the data from these sensors, engineers and other personnel can connect to running trains in real time, allowing SNCF to anticipate when a specific item, such as a faulty signal component, is in need of repair. By predicting when maintenance is needed, SNCF can prevent trains from being taken out of service while avoiding more costly repairs. SNCF estimates that this train and track maintenance approach could reduce costs while significantly improving the reliability of its signals and trains. In addition, with this up-the-second insight, maintenance teams will also constantly be updated on the state of the rail infrastructure and when needed, can provide early warning to expert teams in charge, when there is risk of dysfunction.

Critical to this effort is SNCF’s use of IBM Cloud-based Watson IoT Platform capabilities, which connect its entire rail network made up of thousands of components, the rails and stations. With Watson IoT, the railway company is successfully minimizing delays, ensuring passenger safety and delivering a superior on-train experience for commuters.

Remote monitoring with industrial IoT

At the heart of this effort are thousands of sensors which SNCF is deploying on its trains, covering more 30,000 kilometres of track, 15,000 trains and 3,000 stations. Each of these sensors immediately and securely will send tens of thousands of data points to the IBM Watson IoT Platform on IBM Cloud where the data is analyzed in real-time.

Increasing safety, reliability and savings with predictive maintenance

Using IoT technology, engineers can connect to running trains in real time, enabling SNCF to monitor components and remotely manage work carried out on each individual potential train or rail issue. For example, SNCF can remotely monitor train doors for potential failures, air conditioning, windshield water levels and oil temperatures. By connecting to running trains, engineers and other personnel can anticipate when specific issues need to be addressed.

Using predictive analytics, SNCF has been able to successfully prevent trains from being taken out of service while avoiding more costly repairs. One key area of focus is the temperature of rails. Excessive heat can cause long term maintenance problems that affect the safety of trains, especially those running at high speed.

IoT used to make trains run more safely and efficiently. One example is temperature monitoring. When railway lines become too hot it can impede the train’s speed and functionality. This is not only a nuisance, but it may also increase the risk of an accident. Despite the advanced safety features we see on modern trains today, derailings do occasionally happen, and they are more likely to occur if the tracks are at a higher temperature. This is therefore one of the aspects of train operations that the SCNF wants to address.

The design team plans to do this by implementing temperature sensors at every kilometer of the track. When high temperatures are detected in certain locations, a signal is sent to oncoming trains to slow down—improving train operation and decreasing risk to passengers.

It is worth noting that monitoring equipment does already exist on the SCNF railways. It is estimated that around 30 percent of the railway network is equipped with signaling and telecommunications technology. What is different about the IoT is that engineers predict it will allow data infrastructure to be built faster and cheaper, covering far more tracks in the years to come.

Railway tracks are also just one part of the plan. A single train has millions of moving components operating cohesively, with IoT analytics monitoring certain parts and sending data to a central cloud. A model predictive system can help train operators perform proactive maintenance—something that has enormous potential for cutting costs.

It is for this reason that the SNCF believes it is possible to reduce maintenance costs by between 10 and 30 percent—costs that may translate to employee wages or passenger tickets. If you weren’t already on board to go for a train trip through the French countryside, perhaps this project will convince you.