How IoT can revolutionise running railways

Mobile computing, mobile sensor networks and deep learning are behind French rail operator SNCF’s efficiency drive

This article can also be found in the Premium Editorial Download: CW Europe: European law enforcement seeking smart ways to fight cyber crime

Railways defined the first industrial revolution – now technology is at the heart of train networks’ modernisation in the 21st century.

In a recent blog post, Forrester principal analyst Dan Bieler wrote: “Businesses can obtain major benefits – including better customer experiences and operational excellence – from the internet of things (IoT) by extracting insights from connected objects and delivering feature-rich connected products.”

A rail network comprises thousands, if not millions of components, from rolling stock to signals, rails, stations and the staff who run it all.

All the constituent parts need to work cohesively to avoid delays and dissatisfaction among commuters.

The industrial internet, where IoT devices are combined with analytics, is set to revolutionise train operations.

French national train operator SNCF is using the industrial internet powered by IBM Watson’s deep learning analytics platform and the SigFox IoT network to boost efficiency.

The strategy is part of the company’s 2020 Vision, to become an industrial champion striving for operational excellence and optimum efficiency by using the IoT.

Using the IoT to support the operations of a rail service is now possible thanks to advances in the underlying technology, according to Yves Tyrode, digital director at SNCF.

Telecommunications networks are becoming dedicated to industrial internet applications and broadband is getting cheaper. The train company runs fibre along its tracks and has relationships with mobile operators to make use of this network to maintain continuous mobile broadband connectivity.

The mobile network supports both commuters and the SNCF’s remote data acquisition requirements.

Sensors for data acquisition are getting smaller and now consume less energy. In some cases, battery life can extend up to five years. SNCF said this is important because it is not always possible to be close to an electrical supply.

The third area is maturity of the cloud, which SNCF said would be used to store sensor data and provide the elastic computing required for big data analytics.

Working with Ericsson, IBM, SigFox and sensor specialist Intesens, the rail operator will deploy sensors on trains and tracks to reduce failures and improve the reliability of trains, signals and tracks.

Analysing sensor data

Making sense of sensor data collected remotely over the internet is computationally challenging.

IBM France president Nicolas Sekkaki said: “We have entered an industrial era.” The company has established an IoT centre for its Watson deep learning technology centre in Munich. “We have developed an infrastructure to measure IoT data,” said Sekkaki.

According to SNCF, IoT will enable the company both to improve customer service and the competitiveness of its trains. SNCF estimated that the maintenance of tracks and trains could reduce costs by a factor of 10.

Examples of how SNCF is using IoT

Engineers can connect to running trains in real time, enabling the company to figure out whether a component is likely to fail, which could lead to the train being taken out of service.

The cloud enables SNCF to run distributed calculations, the results of which can be reinjected into its train and rail maintenance processes.

For example, said Vincent Mazarguil, director of asset management at SNCF, the company used big data systems to capture telematics data. “We can anticipate breakdown and model predictive maintenance,” he said. For instance, he said the train operator could identify a faulty signal component, which could be fixed before it failed.

Reduce maintenance

Remote monitoring is also helping the rail operator reduce the maintenance time in the train depot. Train windshield water tanks are being equipped with a level sensor, which uses Sigfox to communicate information back to SNCF. A technician is then able to access this information via a web application on a tablet to see whether the water needs topping up.

Other data acquisition devices are fitted to the transmission system on TGV trains run by SNCF. The company has developed a prototype where sensors take gearbox oil temperature and oil level measurements. The data is transmitted over GSM and can be accessed remotely at the train depot, enabling technicians to see how well the gearbox is performing.

There are also Sigfox communication devices to measure the water level tank in the TGV toilets. This is also used speed up the turnaround time when the train arrives at a depot.

Read more on Big data analytics

Data Center
Data Management