Passenger Rail, Rail Supply, Signalling & Communications, Technology and IT

Prevention to prediction: The future of maintenance

Alstom’s Services Director for Australia Nicolas Thiebot says the company is working with operators and asset owners to demonstrate that it is safe to move to predictive maintenance across their networks.

For generations, rail operators and owners have relied on fixed schedules to maintain their fixed and rolling assets. General maintenance and component replacement was scheduled based on the distance travelled by a piece of rollingstock, or the amount of use endured by a section of track.

While this “preventative maintenance” gets the job done, it is far from the most efficient method. What if a replaced component is still in good condition? What if a component is faulty and degrades faster than it should? What if you’re performing general maintenance too often? Not often enough?

The basic theory behind the development of condition monitoring systems was to address these concerns. Devices installed in trains and around railways record and report data, so assets can be maintained and replaced when needed; not before or after.

Despite the rapid development of proven technologies in this space, however, Alstom’s Nicolas Thiebot says many operators are still hesitant to complete the shift from the old preventative maintenance model, to the new predictive maintenance model.

“The industry has a range of tools and initiatives around condition monitoring, but it’s often bolted onto a systematic maintenance plan,” Thiebot, Alstom’s Services Director in Australia, tells Rail Express.

“People are still a bit reluctant to move away from this systematic, preventative maintenance approach, to a more predictive model. At the moment, it’s a bit of a belt and braces approach, and we really need to move to an optimised version.

“I think the future lives in an understanding that it’s safe and more cost effective, to move on from the preventative to the predictive model.”

Since 2014, Alstom has used its platform, HealthHub, to facilitate a more wholehearted shift towards predictive methods for rail passenger rail operators. HealthHub looks to include as much of the data being collected from around the network, and incorporate it into a central hub to drive maintenance. It takes prioritisation into consideration before making recommendations for work to take place.

“Some of HealthHub’s tools are very mature such as the train to ground remote condition monitoring systems and have been a part of the maintenance industry for a long time,” he says. “The value of HealthHub is to bring them together into a shared platform, and to take the data and convert it into recommended actions to help keep assets in the best possible condition.

“When all the different initiatives around maintenance and asset management are put into one package, you can optimise your operation through dynamic maintenance planning.

“Some of those initiatives are based around asset planning, workload optimisation and work execution, some are to do with remote condition monitoring; they are all important factors in a smart maintenance program.”


Thiebot says Alstom has developed – and continues to develop – HealthHub primarily to aid its own maintenance operations.

Alstom is delivering ongoing rail maintenance for the fi rst stage of Sydney Metro, which opened in May 2019. On that project it is using HealthHub to integrate tools like Catenary Tracer, Track Tracer, Train Tracer, broken rail detection and point machine detection.

Throughout the world, the platform is also available to customers who want to do their own maintenance, however.

“We are developing HealthHub primarily for ourselves, because we believe in the technology and the benefi ts it adds to us,” Thiebot says. “But on operations where we don’t do the maintenance, we can package HealthHub as a product that a customer can use themselves, or as a service that we can perform for them: we produce the reports, do the data mining and so on, to make recommendations to their maintenance operation.”

HealthHub has included onboard rollingstock monitoring and track and catenary monitoring for some time. Thiebot says points machine monitoring is a more recent addition, and the company is looking to add signalling monitoring to the suite as it develops, as well as tools to rationalise the high volumes of data generated by power monitoring and infrastructure monitoring SCADA systems around the rail network.

“The end goal for HealthHub is to help an operator have the smallest downtime possible due to maintenance,” he concludes.