Systematic Literature Review on Data-Driven Models for Predictive Maintenance of Railway Track: Implications in Geotechnical Engineering

Abstract

Conventional planning of maintenance and renewal work for railway track is based on heuristics and simple scheduling. The railway industry is now collecting a large amount of data with the fast-paced development of sensor technologies. These data sets carry information about the conditions of various components in railway track. Since just before the beginning of the 21st century, data-driven models have been used in the predictive maintenance of railway track. This study presents a systematic literature review of data-driven models applied in the predictive maintenance of railway track. A taxonomy to classify the existing literature based on types of models and types of applications is provided. It is found that applying the deep learning methods, unsupervised methods, and ensemble methods are the new trends for predictive maintenance of railway track. Rail geometry irregularity, rail head defect, and missing rail components detection were the top three most commonly considered issues within the application of data-driven models. Prediction of rail breaks has received increasing attention in the last four years. Among these data-driven model applications, the collected data types are the most critical factors which affect selecting suitable models. Finally, this study discusses upcoming challenges in the predictive maintenance of railway track.

Prediction of track geometry degradation using artificial neural network: a case study

“…It is worth mentioning that beside ANN, there exist other techniques such as convolutional neural network (CNN), recurrent neural networks (RNN) and long short-term memory (LSTM) which can be used for the purpose of prediction. CNNs are particularly designed for image data [21,22] and RNN and LSTM are widely used to analyse time series and sequence data [21,23,24]. In this study, we choose ANN due to its ability to easily explore the importance of the input features on the output parameter.…”

Contrastive Feature Learning for Fault Detection and Diagnostics in Railway Applications

“…For the unsupervised AD task, unsupervised methods are used for comparison. First, we compare our results to a statistical measure called the dynamic coefficient (dynCoeff ) -see (2). This coefficient describes the ratio of the maximum dynamic to the static wheel load within each sensor measurement x.…”

“…Hence, railway operators need to ensure safe and reliable operations while being economically efficient. With an increasing investment in condition monitoring (CM) devices, the implementation of data-driven solutions for fault detection and diagnostics has become a promising direction [1,2].…”

A Survey on Data-Driven Predictive Maintenance for the Railway Industry

“…Although ML/DL methods developed for the PdM practices in a wide range of applications, the literature with specific applications in the railway industry is yet scarce. A recent review regarding the data-driven PdM works in the railway tracks can be found in [ 96 ]. The works have been classified based on model types and application types.…”