Skip to main content
SpringerLink
Log in

Two-stage classifications for improving time-to-failure estimates: a case study in prognostic of train wheels

  • Published:
Applied Intelligence Aims and scope Submit manuscript

Abstract

In order to meet the need for higher equipment availability and lower maintenance cost, much attention is being paid to the development of prognostic systems. Such systems support a proactive maintenance strategy by continuously monitoring the components of interest and predicting their failures sufficiently in advance to avoid disruptions during operation. Recent research demonstrated the potential of a comprehensive data mining methodology for building prognostic models from readily available operational and maintenance data. This approach builds a binary classifier that can determine the likelihood of a failure within a broad target window but cannot provide precise time to failure (TTF) estimations. This paper introduces a two-stage classification approach that helps improve the precision of TTF estimations. The new approach uses the initial methodology to learn a variety of base classifiers and then relies on meta-learning to integrate them. The paper details the model building process and demonstrates the usefulness of the proposed approach through a real-world prognostic application.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  1. Létourneau S, Yang C, Drummond C, Scarlett E, Valdés J, Zaluski M (2005) A domain independent data mining methodology for prognostics. In: Conference proceedings: essential technologies for successful prognostics, 59th meeting of the machinery failure prevention technology society, Virginia Beach, Virginia, USA, April 18–21, 2005

  2. Létourneau S, Famili F, Matwin S (1999) Data mining for prediction of aircraft component replacement. IEEE Intell Syst J 59–66. Special issue on data mining

  3. Yang C, Létourneau S (2005) Learning to predict train wheel failures. In: Proceedings of the 11th ACM SIGKDD international conference on knowledge discovery and data mining (KDD2005), Chicago, USA, August, 2005, pp 516–525

  4. Promellec B, Riant I, Tessier-Lescourret C (2006) Precise life-time prediction using demarcation energy approximation for distributed activation energy reaction. J Phys Condens Matter 18(7):2199–2216

    Article  Google Scholar 

  5. Sornette D, Andersen JV (2006) Optimal prediction of time-to-failure from information revealed by damage. Europhys Lett 74(5):778–784

    Article  MathSciNet  Google Scholar 

  6. Van der Duyn Schouten FA, Wartenhorst P (1980) Time to failure, time to repair and availability of a two unit standby system with Markovian degrading units. Technical Report BS-R9007, Center for Mathematics and Computer Science, Amsterdam, The Netherlands

  7. Xu L, Krzyzak A, Suen CY (1992) Methods of combining classifiers and their applications to handwriting recognition. IEEE Trans Syst Man Cybern 22(3):418–435

    Article  Google Scholar 

  8. Dzeroski S, Zenko B (2004) Is combining classifiers with stacking better than selecting the best one? Mach Learn 54:255–273

    Article  MATH  Google Scholar 

  9. Dietterich T (2000) An experimental comparison of three methods for constructing ensembles of decision trees: bagging boosting and randomization. Mach Learn 40:139–158

    Article  Google Scholar 

  10. Tsoumakas G, Katakis I, Blahavas I (2004) Effective voting of heterogeneous classifiers. In: Proceedings of the 15th European conference on machine learning (MCML2004), Pisa, Italy, pp 465–476

  11. Senator TE (2005) Multi-stage classification. In: Fifth IEEE international conference on data mining (ICDM’05), pp 386–393

  12. Provost F, Fawcett T (1997) Analysis and visualization of classifier performance: combination under imprecise class and cost distributions. In: Proceedings of international conference on knowledge discovery and data mining (KDD97)

  13. Ling CX, Huang J, Zhang H (2003) AUC: a statistically consistent and more discriminating measure than accuracy. In: Proceedings of IJCAI 2003, pp 519–524

  14. Drummond C, Holte R (2000) Explicitly representing expected cost: an alternative to ROC representation. In: Proceedings of the 6th ACM SIGKDD international conference on knowledge discovery and data mining, New York, pp 198–207

  15. Hall M (2000) Correlation-based feature selection for discrete and numeric class machine learning. In: Proceedings of the 17th international conference on machine learning, pp 359–366

  16. Smith JR, Chang SF (1997) Multi-stage classification of images from features and related text. In: 4th Europe EDLOS workshop, San Miniato, Italy, August 1997

  17. Cakmakov D, Radevski V, Bennani Y, Gorgevik D (2004) Decision fusion and reliability control in handwritten digit recognition system. J Comput Inf Technol CIT 10:283–293

    Google Scholar 

  18. Rutkowska D, Klimala JK (2004) A multi-stage classification method in application to diagnosis of larynx cancer. In: ICAISC 2004. LNAI, vol 3070, pp 1037–1042

  19. Cordella LP, Limongiello A, Sansone C (2004) Network intrusion detection by a multi-stage classification system. In: MCS 2004. LNCS, vol 3077, pp 324–333

  20. Salient Systems Inc. Preventing Train Derailment. http://www.salientsystems.com

  21. Technical report (2004) Advanced Technology Safety Initiative: Equipment Health Management System, Association of American Railroads, 24 August 2004

  22. Lechowicz S, Hunt C Monitoring and managing wheel condition and loading. http://www.ntsb.gov/events/symp_rec/proceedings/authors/lechowicz.pdf

  23. Gebraeel N, Lawley M, Liu R, Parmeshwaran V (2004) Residual life prediction from vibration-based degradation signals: a neural network approach. IEEE Trans Ind Electron 51(3):692–700

    Article  Google Scholar 

  24. Grottke M, Trivedi K (2005) On a method for mending time to failure distributions. In: Proceedings of international conference on dependable systems and networks 2005, Los Alamitos, pp 560–569

  25. Schwabacher MA (2005) A survey of data-driven prognostics. In: Proceedings of AIAA aerospace conference (AIAA 2005), Arlington, Virginia, USA, September 2005

  26. Dzeroski S, Zenko B (2002) Stacking with multi-response model trees. In: The proceeding of international workshop in multiple classifier systems (MCS2002), pp 201–211

  27. Wolpert D (1992) Stacked generalization. Neural Netw 5(2):241–260

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. National Research Council Canada, 1200 Montreal Road, Ottawa, Ontario, K1A 0R6, Canada

    Chunsheng Yang & Sylvain Létourneau

Authors
  1. Chunsheng Yang
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Sylvain Létourneau
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Chunsheng Yang.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Yang, C., Létourneau, S. Two-stage classifications for improving time-to-failure estimates: a case study in prognostic of train wheels. Appl Intell 31, 255–266 (2009). https://doi.org/10.1007/s10489-008-0123-1

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10489-008-0123-1

Keywords

Search

Navigation