Joint Prediction of Remaining Useful Life and Failure Type of Train Wheelsets: Multitask Learning Approach
Publication: Journal of Transportation Engineering, Part A: Systems
Volume 144, Issue 6
Abstract
Train wheel failures account for disruptions of train operations and even a large portion of train derailments. Remaining useful life (RUL) of a wheelset measures how soon the next failure will arrive, and the failure type reveals how severe the failure will be. RUL prediction is a regression task, whereas failure type is a classification task. In this paper, the authors propose a multitask learning approach to jointly accomplish these two tasks by using a common input space to achieve more desirable results. A convex optimization formulation is developed to integrate least-squares loss and negative maximum likelihood of logistic regression as well as model the joint sparsity as the L2/L1 norm of the model parameters to couple feature selection across tasks. The experiment results show that the multitask learning method outperforms both the single-task learning method and Random Forest.
Get full access to this article
View all available purchase options and get full access to this article.
References
Askarinejad, H., Dhanasekar, M., and Cole, C. (2013). “Assessing the effects of track input on the response of insulated rail joints using field experiments.” Proc. Inst. Mech. Eng. Part F, 227(2), 176–187.
Banjevic, D. (2009). “Remaining useful life in theory and practice.” Metrika, 69(2–3), 337–349.
Bazaraa, M. S., Sherali, H. D., and Shetty, C. M. (2013). Nonlinear programming: Theory and algorithms, Wiley, Hoboken, NJ.
Boyd, S., and Vandenberghe, L. (2004). Convex optimization, Cambridge University Press, Cambridge, U.K.
Braghin, F., Lewis, R., Dwyer-Joyce, R., and Bruni, S. (2006). “A mathematical model to predict railway wheel profile evolution due to wear.” Wear, 261(11), 1253–1264.
Cummings, S. (2012). “Wheel failure is not an option.” Railway Age, 213(6), 29–30.
Czepiel, S. A. (2002). “Maximum likelihood estimation of logistic regression models: Theory and implementation.” ⟨czep.net/stat/mlelr.pdf⟩ (Nov. 10, 2017).
Ekberg, A., Kabo, E., and Andersson, H. (2002). “An engineering model for prediction of rolling contact fatigue of railway wheels.” Fatigue Fract. Eng. Mater. Struct., 25(10), 899–909.
Evgeniou, A., and Pontil, M. (2007). “Multi-task feature learning.” Adv. Neural Inf. Process. Syst., 19, 41–48.
FRA (Federal Railroad Administration). (2015). Mechanical inspections and wheel impact load detector standards for trains transportation large amounts of Class 3 flammable liquids, Department of Transportation, Washington, DC.
Hajibabai, L., et al. (2012). “Wayside defect detector data mining to predict potential wild train stops.” Annual Conf. and Exposition of the American Railway Engineering and Maintenance-of-Way Association, American Railway Engineering and Maintenance-of-Way Association, Chicago.
Huang, C.-L., and Wang, C.-J. (2006). “A GA-based feature selection and parameters optimization for support vector machines.” Expert Syst. Appl., 31(2), 231–240.
Lechowicz, S., and Hunt, C. (1999). “Monitoring and managing wheel condition and loading.” Transportation Recording: 2000 and Beyond. Int. Symp. on Transportation Recorders, Arlington, VA, 205–239.
Li, H., et al. (2014). “Improving rail network velocity: A machine learning approach to predictive maintenance.” Transp. Res. Part C Emerging Technol., 45, 17–26.
Li, Z., and He, Q. (2015). “Prediction of railcar remaining useful life by multiple data source fusion.” IEEE Trans. Intell. Transp. Syst., 16(4), 2226–2235.
Lin, J.-Y., Cheng, C.-T., and Chau, K.-W. (2006). “Using support vector machines for long-term discharge prediction.” Hydrol. Sci. J., 51(4), 599–612.
MATLAB [Computer software]. MathWorks, Natick, MA.
Melgani, F., and Bruzzone, L. (2004). “Classification of hyperspectral remote sensing images with support vector machines.” IEEE Trans. Geosci. Remote Sens., 42(8), 1778–1790.
Obozinski, G., Taskar, B., and Jordan, M. (2006). “Multi-task feature selection.”, Statistics Dept., Univ. of California, Berkeley, CA, 2.
Palo, M., Schunnesson, H., Kumar, U., Larsson-Kråik, P.-O., and Galar, D. (2012). “Rolling stock condition monitoring using wheel/rail forces.” Insight-Non-Destr. Test. Condition Monit., 54(8), 451–455.
Putter, H., Fiocco, M., and Geskus, R. (2007). “Tutorial in biostatistics: Competing risks and multi-state models.” Stat. Med., 26(11), 2389–2430.
R [Computer software]. R Core Team, Auckland, New Zealand.
Railway Technical. (2017) “Train maintenance.” ⟨http://www.railway-technical.com/trains/train-maintenance/⟩ (Oct. 15, 2017).
Ruszczyński, A. P. (2006). Nonlinear optimization, Vol. 13, Princeton University Press, Princeton, NJ.
Schmidt, M. (2005). “Least squares optimization with l1-norm regularization.”, Univ. of British Columbia, Vancouver, BC, Canada, 14–18.
Scrucca, L., Santucci, A., and Aversa, F. (2007). “Competing risk analysis using R: An easy guide for clinicians.” Bone Marrow Transp., 40(4), 381–387.
Shin, K.-S., Lee, T. S., and Kim, H.-J. (2005). “An application of support vector machines in bankruptcy prediction model.” Expert Syst. Appl., 28(1), 127–135.
Si, X.-S. (2015). “An adaptive prognostic approach via nonlinear degradation modeling: Application to battery data.” IEEE Trans. Ind. Electron., 62(8), 5082–5096.
Si, X.-S., Hu, C.-H., Kong, X., and Zhou, D.-H. (2014). “A residual storage life prediction approach for systems with operation state switches.” IEEE Trans. Ind. Electron., 61(11), 6304–6315.
Si, X.-S., Wang, W., Hu, C.-H., and Zhou, D.-H. (2011). “Remaining useful life estimation—A review on the statistical data driven approaches.” Eur. J. Oper. Res., 213(1), 1–14.
Stratman, B., Liu, Y., and Mahadevan, S. (2007). “Structural health monitoring of railroad wheels using wheel impact load detectors.” J. Fail. Anal. Prev., 7(3), 218–225.
Taghavifar, H., and Mardani, A. (2014). “A comparative trend in forecasting ability of artificial neural networks and regressive support vector machine methodologies for energy dissipation modeling of off-road vehicles.” Energy, 66, 569–576.
Yang, C., and Létourneau, S. (2005). “Learning to predict train wheel failures.” Proc., 11th ACM SIGKDD Int. Conf. on Knowledge Discovery in Data Mining, Association for Computing Machinery, New York.
Yang, X., Kim, S., and Xing, E. P. (2009). “Heterogeneous multitask learning with joint sparsity constraints.” Advances in Neural Information Processing Systems, Neural Information Processing Systems Conference, Vancouver, BC, Canada, 2151–2159.
Ye, Z.-S., Chen, N., and Shen, Y. (2015). “A new class of Wiener process models for degradation analysis.” Reliab. Eng. Syst. Saf., 139, 58–67.
Ye, Z.-S., and Xie, M. (2015). “Stochastic modelling and analysis of degradation for highly reliable products.” Appl. Stochastic Models Bus. Ind., 31(1), 16–32.
Zhang, D., Shen, D., and Alzheimer’s Disease Neuroimaging Initiative. (2012). “Multi-modal multi-task learning for joint prediction of multiple regression and classification variables in Alzheimer’s disease.” Neuroimage, 59(2), 895–907.
Zong, N., and Dhanasekar, M. (2013). “Hybrid genetic algorithm for elimination of severe stress concentration in railhead ends.” J. Comput. Civ. Eng., 29(5), 04014075.
Information & Authors
Information
Published In
Journal of Transportation Engineering, Part A: Systems
Volume 144 • Issue 6 • June 2018
Copyright
©2018 American Society of Civil Engineers.
History
Received: Mar 26, 2017
Accepted: Aug 1, 2017
Published online: Mar 26, 2018
Published in print: Jun 1, 2018
Discussion open until: Aug 26, 2018
Authors
Metrics & Citations
Metrics
Citations
Download citation
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.