Unplanned-maintenance needs related to rail track geometry

Abstract

This paper explores unplanned-maintenance needs regarding rail track geometry degradation using a logistic regression approach. The authors conducted an exploratory statistical analysis using the inspection records from the Portuguese Infrastructure Manager in order to predict the probability that a given track section needs unplanned maintenance depending on the planned-maintenance indicators proposed by the European Standard EN 13848, and on other explaining variables such as the presence of bridges and switches. The main conclusions are that the typical planned-maintenance criteria (the standard deviations of longitudinal-levelling and of horizontal-alignment defects) are statistically significant predictors of unplanned-maintenance needs. Moreover, for some speed groups and quality levels, the presence of bridges and switches is also statistically significant. These findings contribute to a quantified understanding of the underlying trade-off between planned-maintenance and unplanned-maintenance needs in the current suggested limits contained in the standard EN 13848.