News and Events

Predictive system targets 25% of train defects

30/01/2018 30/01/2018

An innovative new project using vibration analysis will provide huge safety and efficiency savings, as well as reduced downtime for rolling stock operators in the rail industry.  In Europe, 25%-50% of defects in trains occur in the doors, while worldwide it has been recorded, that 70% of passenger injuries are caused by mechanical failures of train doors. The VA-RCM (Vibration analysis for Remote Condition Monitoring) project is working on a system which can remotely detect failures before they happen and predict the lifespan of the door.

A consortium consisting of Hitex (UK, Germany), TWI Ltd (UK), Transport Systems Catapult (UK), TMB (Spain) and Innovative Technology and Science Limited (UK) will be responsible for the release of this system. The project will be completed in early 2019.

The concept behind the VA-RCM system lies in vibration analysis algorithms. Vibration takes places when either there is a defect or an external stimulus effecting the mechanism. The system will be able to detect problems with train doors based on comparison with previous vibration data and predict their remaining lifetime. This data will be gathered remotely in real time, so interventions can be targeted rather than relying on a constant maintenance cycle.

Sofia Sampethai, Project Manager from Transport Systems Catapult

“The transport industry is increasingly seeing the value of smart solutions based on data analysis, Internet of Things and modern senor technology, to provide cost savings and increased safety in a range of scenarios. Train operators are no exception, and in this case, we have identified a clear cause of disruption where a huge impact can be made.

The VA-RCM project can reduce unexpected failures and incidents which can cause carriages to be taken out of service – reducing capacity and causing disruptions to service. We can also provide a way to reduce maintenance costs through predictive systems.”

VA-RCM is a project funded by the European Union’s H2020 research and innovation program under grant agreement No 730766.