Keywords: railway tracks, in situ measurements, reliability analysis, maintenance.

This paper proposes a methodology to evaluate railway track maintenance operations in term of reliability based on numerical and experimental methods. Firstly, we propose an experimental method of diagnosing the track characteristics. This method is based on penetrometric and endoscopy in situ measurements allowing the characterization of geometrical (thickness of layers) and mechanical parameters (Young's modulus) of the different layers of the railway platform. The diagnostic devices permit an important amount of data to be obtained and thus to record statistics on the geometric and mechanical parameters and to describe their variability.

Then a numerical model of the railway track is developed using the finite element method. This model represents a portion of the track, and takes into account the variability of the model input data. Finally, using a reliability analysis method (the first order reliability method), we compare the impact of two kinds of maintenance operations on the behaviour of the track by the estimation of the failure probability: the first one is the raising of the track by the injection and consolidation of the ballast layer, the second one is of the type "renewal of track and ballast" (RTB) which consists of the treatment of ballast and sub-ballast layers. For this kind of operation two level of compacting for the sub-ballast layer are considered.

The objective of the structural reliability analysis is to compute the probability of failure of a track portion regarding a prescribed failure criterion which takes account for uncertainties arising from the thickness and the Young's modulus of the layers of the track structure. The different operations have been compared using their contribution to the reduction of the probability of failure defined as the probability of exceeding allowable threshold values of the rail deflection (criterion 1) and of the peak sleeper acceleration (criterion 2). This analysis has shown the great contribution of second types of treatment (ballast and sub-ballast treatement) in term of reducing the dispersion of the model response and especially of reducing the risk of default compared to the raising of the track.

The approach presented in this paper gives important information to railway managers enabling them to optimize maintenance operations.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £110 +P&P)