Forecasting Risk of Service Failures Between Successive Rail Inspections: A Data-Driven Approach

Ghofrani, Faeze; Chava, Naresh Kumar; He, Qing

doi:10.1007/s42421-020-00015-y

Cited by 7 publications

(4 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Prediction of rail breaks will assist the railway companies to avoid possible derailments and to minimize the service interruptions caused by broken rails. In recent years, there has been a good attention to prediction of different types of rail defects for the purpose of railway maintenance using the recent advances in big data analytics (Ghofrani, Kumar, et al., 2020; Ghofrani, He, Mohammadi, Pathak, & Aref, 2019; Ghofrani, Pathak, Mohammadi, Aref, & He, 2020; Jamshidi et al., 2017; Mohammadi, He, Ghofrani, Pathak, & Aref, 2019). A comprehensive overview of all these studies could be found in Ghofrani, He, Goverde, and Liu (2018) and Ghofrani (2020).…”

Section: Literature Reviewmentioning

confidence: 99%

See 1 more Smart Citation

Analyzing Risk of Service Failures in Heavy Haul Rail Lines: A Hybrid Approach for Imbalanced Data

Ghofrani

Sun

2021

Risk Analysis

Self Cite

View full text Add to dashboard Cite

An incident in which a rail defect of size over a threshold value is noticed and the track is taken out of service is known as a service failure. This article aims at building accurate prediction models with binary outcome for risk of service failures on heavy haul rail segments. An analysis of the factors that influence the risk of a service failure is conducted and quantitative models are developed to predict locations where service failures are most likely to occur until the next inspection. To this end, data are collected from a Class I U.S. Railroads for six years from 2011 to 2016. Four prediction models (i.e., logistic regression, decision tree, multilayer perceptron, and gradient boosting classifier) are implemented and their results are compared. To account for the imbalanced classes between the normal operation and service failure, two treatments have been used including undersampling and oversampling. To improve the model performance, the parameters of each method are tuned using random search hyperparameter optimization. Later, bootstrap aggregation (or bagging) is incorporated into each method. The findings of the study show that the prediction performance is the highest when using bagging and oversampling as treatments with gradient boosting method. It was also identified that gross tonnage, presence of geometry defects, ambient temperature, segment length, and rail defect presence are the most important factors for predicting the risk of service failures. The results of this study are useful for railroads to develop effective strategies for rail inspections, preventive maintenance, and capital planning.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

“…Service failure is an incident when a defect of size over threshold value is noticed and the track is taken out of service (Ghofrani, Kumar, & Qing, 2020). Broken rails, as a major type of service failure, are a leading cause of derailments.…”

Section: Introductionmentioning

confidence: 99%

Analyzing Risk of Service Failures in Heavy Haul Rail Lines: A Hybrid Approach for Imbalanced Data

Ghofrani

Sun

2021

Risk Analysis

Self Cite

View full text Add to dashboard Cite

show abstract

“…This has been the subject of numerous ongoing research activities using Weibull, Bayesian, and other probability models. 10–18 It should be noted that many of the more recent probability models use a system probability approach to define risk of failure and associated derailments, while the Weibull approach is the one most commonly used to examine data on an individual segment basis.…”

Section: Introductionmentioning

confidence: 99%

Prediction of rail defect development using parametric bootstrapping modified Weibull equations

Cronin

Zarembski

Palese

2021

Proceedings of the Institution of Mechanical Engineers, Part F:

View full text Add to dashboard Cite

The railroad industry has historically used the 2-Parameter Weibull equation to determine the rate of rail fatigue defect occurrences and to forecast the fatigue life of railroad rail. However, the 2-Parameter Weibull equation has significant limitations to include inability to analyze segments of track with limited number of rail defects. These limitations are addressed through modification of the traditional 2-Parameter Weibull equation with a novel approach developed from Parametric Bootstrapping. The result is a Parametric Bootstrapping modified Weibull (PBW) forecasting approach. This methodology is applied to rail segments with insufficient numbers of defects to allow for appropriate defect forecasting analysis. Thus, the PBW method provides reasonable estimates of the rate of defects for track segments that have little or no prior defect history. This approach allows for more track to be analyzed and forecasts the probability of rail defect occurrence as a function of key parameters such as cumulative traffic over the rail. A validation of the proposed methodology was performed. Comparison of the output results of over 300,000 track segments with over 200,000 rail defects showed a major improvement in percentage of segments with reasonable Weibull parameters (alpha and beta). This percentage increased from 11% of segments using traditional Weibull analysis to 77% of segments using Parametric Bootstrap modified Weibull approach. These results show that the PBW Analysis approach introduced here offers a more accurate and effective approach to determining the probability of developing future rail defects. This provides a benefit to railroads in planning maintenance of their expensive rail assets.

show abstract

“…Bai et al [3] employed the theory of Markov stochastic processes to estimate the rail break time in a long period, such as five years. A regression model based on survival analysis was developed by Ghofrani et al [4] to predict the risk of rail breaks between two successive rail inspections. Later, Ghofrani et al [5] developed an improved version of the prediction model to predict the arrival rate of rail breaks.…”

mentioning

confidence: 99%

Rail Break Prediction and Cause Analysis Using Imbalanced In-Service Train Data

Zeng

Huang

Wang

et al. 2022

IEEE Trans. Instrum. Meas.

View full text Add to dashboard Cite

Timely detection and identification of rail breaks are crucial for safety and reliability of railway networks. This article proposes a new deep learning-based approach using the daily monitoring data from in-service trains. A time-series generative adversarial network (TimeGAN) is employed to mitigate the problem of data imbalance and preserve the temporal dynamics for generating synthetic rail breaks. A feature-level attentionbased bidirectional recurrent neural network (AM-BRNN) is proposed to enhance feature extraction and capture two-direction dependencies in sequential data for accurate prediction. The proposed approach is implemented on a three-year dataset collected from a section of railroads (up to 350 km) in Australia. A real-life validation is carried out to evaluate the prediction performance of the proposed model, where historical data are used to train the model and future "unseen" rail breaks along the whole track section are used for testing. The results show that the model can successfully predict nine out of 11 rail breaks three months ahead of time with a false prediction of nonbreak of 8.2%. Predicting rail breaks three months ahead of time will provide railroads enough time for maintenance planning. Given the prediction results, a Shapley additive explanations (SHAP) method is employed to perform a cause analysis for individual rail break. The results of cause analysis can assist railroads to plan appropriate maintenance to prevent rail breaks.

show abstract

Forecasting Risk of Service Failures Between Successive Rail Inspections: A Data-Driven Approach

Cited by 7 publications

References 20 publications

Analyzing Risk of Service Failures in Heavy Haul Rail Lines: A Hybrid Approach for Imbalanced Data

Analyzing Risk of Service Failures in Heavy Haul Rail Lines: A Hybrid Approach for Imbalanced Data

Prediction of rail defect development using parametric bootstrapping modified Weibull equations

Rail Break Prediction and Cause Analysis Using Imbalanced In-Service Train Data

Contact Info

Product

Resources

About