A data driven degradation-based model for the maintenance of turnouts: a case study

Abstract The goal of this paper is to share experience about the degradation assessment of railway track point machines. A real scale point machine test bench was used to perform several thousands of maneuvers. Recorded data consisted in the measurements of the shaft speed and the torque during a maneuver. Some faulty conditions were introduced to obtain degraded signals. Relevant indicators were computed to quantify the health state of the point machine and to detect faulty behaviours. The analysis also highlighted the influence of the temperature on the point machine performance. A degradation model is then built from the indicator to compute the reliability and the remaining useful life. The conclusions of this analysis will be useful to properly define an experimental design to be applied on a real turnout.