TRANSFORMING TRANSPORT RAIL UK

PILOT : PROACTIVE RAIL INFRASTRUCTURES IN THE UK

BIG DATA VALUE IN MOBILITY AND LOGISTICS

LOCATION    LONDON

COUNTRY    UNITED KINGDOM

CONTACT NAME    JOANNA THORNTON

WEB    https://transformingtransport.eu/transport-domains/predictive-rail-asset-management-pilot

EMAIL   

PILOT PRODUCT DESCRIPTION

This pilot is focused on using big data technologies and tooling to enable better maintenance scheduling of assets by predicting the health (and rate of decline), state, volatility, level of risk, and future issues likely to be incurred.

The pilot is split into a number of use cases as indicated by the following diagram.

Use Case 1 has focused on the development of an anomaly detection algorithm to detect outliers in the datasets.  This will be used to drive a decision support tool which will provide cues that an asset is “anomalous” and allows the end user to make immediate decisions on what actions need to be planned.

Use Case 2 has primarily focused on identification of health trends of a point machine prior to failure.  To date the researchers have focused on identifying the different fault types through analysis of an assets current trace.  In this respect, this use case is similar to Use Case 1 in that it allows the decision support tool to identify whether an asset is anomalous which can help the end user with making immediate decisions on what actions need to be planned.

Use Case 3 has focused on data analysis of tamping and stone blowing of the track to improve maintenance planning and thus reduce maintenance costs. The Decision Support Tool both monitors and models the degradation of all assets, based on sensor data and predictive models, and then provides the User with information to make decisions about where to tamp based on the degradation rate of the asset population as a whole, the effectiveness of tamping at a location and the amount of work that a tamping machine can achieve per shift.

OBJETIVE OF THE PILOT

The objectives of the pilot are:

  • Verify the quality, accuracy and provenance of asset data, leading to confidence to
  • Provide timely focused prioritised maintenance activities (predict and prevent), leading to
  • Improved reliability and availability of track-side assets, with
  • Higher availability of rail infrastructure for passenger and freight services; and
  • Enhancing worker safety through minimising track-side activities

EXPECTED RESULTS

Expected results:

  • It is expected that through implementation of the respective decision support tools that maintenance activities can be minimized and prioritized effectively and therefore result in a reduction in maintenance costs and in turn reduce the number of asset failures and prolong the life of all assets.
  • It is expected that via the implementation of the fault classifier algorithm the user will be able to reduce the number of false alarms raised for Point Operating Equipment and thus reducing the number of times maintenance staff are sent out to track side.

KPIS AND METRICS

The main KPIs of the pilot are:

  • RI – OE – 1 Reduction in Operating Cost
    • Demonstrate a reduction of the average circulation/journey cost for the rail track.
  • RI – OE – 8 – Reduction in Maintenance Costs
    • Demonstrate a reduction in the amount of time spent by staff performing maintenance tasks by measuring the average amount of time spent per maintenance task.
  • RI – SF- 3 – Reduction of number of track-side activities
    • Demonstrate a reduction in the number of track-side activities

PILOT PARTNERS

X