Evaluation of condition, prediction and asset management in Railway environment

Important results from the project

The purpose and goal of the project was to develop a measurement strategy for assessing railway rails, utilizing modern measurement sensors and data-driven analysis tools for tests undertaken in the Stockholm Metro. In addition to this, a model for predictive maintenance was to be developed and recommendations as to how it should be used for maintenance operations presented.

Expected long term effects

Through continuous data collection and the development of a predictive maintenance model it is expected that the characteristics of the infrastructure will be better understood and maintenance work can be optimized based on the infrastructure’s current condition. This enables better utilization of the infrastructure by balancing performance, financial resources and risks. This methodology is applicable to other track installations and can be used in an international commercial market.

Approach and implementation

The project has been implemented by studying the nature of typical rail damage. A representative test track was subsequently identified that could be used to develop and analyze a multi-sensor measurement strategy to mitigate these errors. The sensor used for detecting damage using eddy currents has been analyzed in a laboratory. A predictive model has been developed and relevant input parameters that are required for analysis and subsequent use within the maintenance business are described.