Abstract
Most of the reported prognostic techniques use a small number of condition indicators and/or use a thresholding strategies in order to predict the remaining useful life (RUL). In this paper, we propose a reliability-based prognostic methodology that uses condition monitoring (CM) data which can deal with any number of condition indicators, without selecting the most significant ones, as many methods propose. Moreover, it does not depend on any thresholding strategies provided by the maintenance experts to separate normal and abnormal values of condition indicators. The proposed prognostic methodology uses both the age and CM data as inputs to estimate the RUL. The key idea behind this methodology is that, it uses Kaplan–Meier as a time-driven estimation technique, and logical analysis of data as an event-driven diagnostic technique to reflect the effect of the operating conditions on the age of the monitored equipment. The performance of the estimated RUL is measured in terms of the difference between the predicted and the actual RUL of the monitored equipment. A comparison between the proposed methodology and one of the common RUL prediction technique; Cox proportional hazard model, is given in this paper. A common dataset in the field of prognostics is employed to evaluate the proposed methodology.
Similar content being viewed by others
Abbreviations
- RUL:
-
Remaining useful life
- CM:
-
Condition monitoring
- KM:
-
Kaplan–Meier
- LAD:
-
Logical analysis of data
- CBM:
-
Condition based maintenance
- ANNs:
-
Artificial neural networks
- PHM:
-
Proportional hazards model
- LR:
-
Logistic regression
- SVMs:
-
Support vector machines
- RVM:
-
Relative vector machine
- HMM:
-
Hidden Markov model
- MILP:
-
Mixed integer linear programming
- LASD:
-
Logical analysis of survival data
- SL:
-
Short life
- LL:
-
Long life
- MTTF:
-
Mean time to failure
- TTF:
-
Time to failure
- MRUL:
-
Mean remaining useful life
- RMSE:
-
Root mean squared error
- \( Cov (p)\) :
-
The set of observations covered by the pattern \(p\)
- \(\varOmega ^{SL}\) :
-
The set of SL equipment
- \(\varOmega ^{LL}\) :
-
The set of LL equipment
- \(\varOmega \) :
-
The training data set
- \(O(i,t_{F_i},Z_{i,t_{Fi}})\) :
-
The failure observation collected from the \(i\)th equipment \((i=1,2,\ldots T)\), where \(Z_{i,t_{Fi}}\) is the corresponding vector of covariates, \(t_S\) is the separation time between the two classes (SL and LL), and \(t_{Fi}\) is the failure time of the \(i\)th equipment
- \(O\left( {i,t_{Fi} \le t_S, Z_{i,t_{Fi} \le t_S}}\right) \) :
-
Set of positive (short life) observations in the training dataset
- \(O\left( {i,t_{Fi} >t_S,Z_{i,t_{Fi} >t_S}}\right) \) :
-
Set of negative (long life) observations in the training dataset
- \(O(u,t_k,Z_{u,t_k})\) :
-
The updating observation collected from the \(u\)th equipment \((u=1,2,\ldots U)\) at time \(t_k \;(t_k =1,2,\ldots t_{Fu})\), and \(Z_{u,t_k}\) is the vector of covariates at time \(t_k\)
- \(S_b(t)\) :
-
The baseline survival function estimated by KM
- \(d_{t_{Fi}}\) :
-
The number of equipment that failed at time \(t_{Fi}\)
- \(Y_{t_{Fi}}\) :
-
The number of equipment which are at risk at time \(t_{Fi}\)
- \(P\) :
-
The set of generated patterns
- \(|P|\) :
-
The cardinality of the set of generated patterns \(P\) (i.e. the number of generated patterns)
- \(p_{j}\) :
-
A pattern belongs to the set of generated patterns \(P\), where \(j=1,2,\ldots |P|\)
- \(S_{O(u,t_k,Z_{u,t_k})} (t)\) :
-
The updated survival curve of the updating observation \(O(u,t_k,Z_{u,t_k})\)
- \(S_f (t)\) :
-
The former updated survival curve obtained from the previous updating observation
- \(S(\tau )\) :
-
The survival function, where \(\tau \) is a dummy variable
- \(MRUL_u ({t_k})\) :
-
The mean remaining useful life of equipment \(u\), calculated at time \(t_k\)
- \(\varDelta t_r\) :
-
Is the monitoring interval which is the difference between two consecutive inspection intervals i.e. \(\varDelta t_r = t_{r+1} -t_r\)
- \(RMSE\left( u\right) \) :
-
The calculated RMSE for the MRUL estimation of the system \(u\) in the updating dataset
- \(N_{Fu}\) :
-
The actual number of operational cycles until the failure of the equipment \(u\)
References
Alexe, S., Blackstone, E., Hammer, P. L., Ishwaran, H., Lauer, M. S., & Pothier Snader, C. E. (2003). Coronary risk prediction by logical analysis of data. Annals of Operations Research, 119, 15–42.
Alexe, G., Alexe, S., Bonates, T. O., & Kogan, A. (2007). Logical analysis of data—the vision of Peter L. Hammer. Annals of Mathematics and Artificial Intelligence, 49, 265–312.
Banjevic, D., & Jardine, A. (2007). Remaining useful life in condition based maintenance: Is it useful? In Modelling in industrial maintenance and reliability (p. 7).
Bennane, A., & Yacout, S. (2012). LAD-CBM; new data processing tool for diagnosis and prognosis in condition-based maintenance. Journal of Intelligent Manufacturing, 23, 265–275.
Bores, E., Hammer, P. L., Ibaraki, T., Kogan, A., Mayoraz, E., & Muchnik, I. (2000). An implementation of logical analysis of data. IEEE Transactions on Knowledge and Data Engineering, 12, 292–306.
Caesarendra, W., Widodo, A., & Yang, B. S. (2010). Application of relevance vector machine and logistic regression for machine degradation assessment. Mechanical Systems and Signal Processing, 24, 1161–1171.
Crama, Y., Hammer, P. L., & Ibaraki, T. (1988). Cause-effect relationships and partially defined Boolean functions. Annals of Operations Research, 16, 299–325.
Daniel, W. W. (1990). Applied nonparametric statistics (2nd ed.). Boston: PWS-KENT Pub.
Duda, R. O., Hart, P. E., & Stork, D. G. (2001). Pattern classification. A Wiley-Interscience publication. New York: Wiley.
Elsayed, E. A. (2012). Reliability engineering. London: Wiley.
Friedman, J., Hastie, T., & Tibshirani, R. (2001). The elements of statistical learning (Vol. 1). Springer Series in Statistics.
Guo, C., & Ryoo, H. S. (2012). Compact MILP models for optimal and Pareto-optimal LAD patterns. Discrete Applied Mathematics, 160, 2339–2348.
Gwet, K. L. (2011). The practical guide to statistics: Applications with excel, R, and calc. Gaithersburg, MD: Advanced Analytics, LLC.
Hamada, M. (2005). Using degradation data to assess reliability. Quality Engineering, 17, 615–620.
Hammer, P. L., Kogan, A., Simeone, B., & Szedmák, S. (2004). Pareto-optimal patterns in logical analysis of data. Discrete Applied Mathematics, 144, 79–102.
Hammer, P. L., & Bonates, T. O. (2006). Logical analysis of data—an overview: From combinatorial optimization to medical applications. Annals of Operations Research, 148, 203–225.
Heng, A., Tan, A. C. C., Mathew, J., Montgomery, N., Banjevic, D., & Jardine, A. K. S. (2009). Intelligent condition-based prediction of machinery reliability. Mechanical Systems and Signal Processing, 23, 1600–1614.
Hosmer, D. W, Jr, & Lemeshow, S. (2011). Applied survival analysis: Regression modeling of time to event data (Vol. 618). New York: Wiley.
Jardine, A., Joseph, T., & Banjevic, D. (1999). Optimizing condition-based maintenance decisions for equipment subject to vibration monitoring. Journal of Quality in Maintenance Engineering, 5, 192– 202.
Jardine, A. K. S., Lin, D., & Banjevic, D. (2006). A review on machinery diagnostics and prognostics implementing condition-based maintenance. Mechanical systems and signal processing, 20, 1483– 1510.
Kim, H.-E., Tan, A. C., Mathew, J., & Choi, B.-K. (2012). Bearing fault prognosis based on health state probability estimation. Expert Systems with Applications, 39, 5200–5213.
Klein, J., & Moeschberger, M. (1997). Survival analysis: Techniques for censored and truncated data. New York: Spring.
Kothamasu, R., Huang, S. H., & VerDuin, W. H. (2006). System health monitoring and prognostics—a review of current paradigms and practices. The International Journal of Advanced Manufacturing Technology, 28, 1012–1024.
Kronek, L. P., & Reddy, A. (2008). Logical analysis of survival data: Prognostic survival models by detecting high-degree interactions in right-censored data. Bioinformatics, 24, i248–i253.
Le Son, K., Fouladirad, M., Barros, A., Levrat, E., & Iung, B. (2013). Remaining useful life estimation based on stochastic deterioration models: A comparative study. Reliability Engineering and System Safety, 112, 165–175.
Liao, H., Zhao, W., & Guo, H. (2006). Predicting remaining useful life of an individual unit using proportional hazards model and logistic regression model (pp. 127–132).
Mortada, M.-A., Yacout, S., & Lakis, A. (2013). Fault diagnosis in power transformers using multi-class logical analysis of data. Journal of Intelligent Manufacturing. doi:10.1007/s10845-013-0750-1.
Mortada, M.-A., Yacout, S., & Lakis, A. (2011). Diagnosis of rotor bearings using logical analysis of data. Journal of Quality in Maintenance Engineering, 17, 371–397.
Pintilie, M. (2006). Competing risks: A practical perspective (Vol. 58). New York: Wiley.
Ryoo, H. S., & Jang, I. Y. (2009). Milp approach to pattern generation in logical analysis of data. Discrete Applied Mathematics, 157, 749–761.
Saha, B., & Goebel, K. (2008). Uncertainty management for diagnostics and prognostics of batteries using Bayesian techniques (pp. 1–8).
Saxena, A., Goebel, K., Simon, D., & Eklund, N. (2008). Damage propagation modeling for aircraft engine run-to-failure simulation (pp. 1–9).
Schwabacher, M., & Goebel, K. (2007). A survey of artificial intelligence for prognostics. Paper presented at the Artificial Intelligence for Prognostics-AAAI Fall Symposium, November 9–11 (pp. 107–114). Arlington, VA.
Tan, A., Heng, A. S. Y., & Mathew, J. (2009). Condition-based prognosis of machine health. In Proceedings of the 13th Asia-Pacific Vibration Conference (pp. 1–10). University of Canterbury.
Tian, Z., Lin, D., & Wu, B. (2012). Condition based maintenance optimization considering multiple objectives. Journal of Intelligent Manufacturing, 23, 333–340.
Tian, Z., Wong, L., & Safaei, N. (2010). A neural network approach for remaining useful life prediction utilizing both failure and suspension histories. Mechanical Systems and Signal Processing, 24, 1542–1555.
Vachtsevanos, G. J., Lewis, F. L., Roemer, M., Hess, A., & Wu, B. (2006). Intelligent fault diagnosis and prognosis for engineering systems. London: Wiley.
Wang, W. (2007). A prognosis model for wear prediction based on oil-based monitoring. Journal of the Operational Research Society, 58, 887–893.
Widodo, A., & Yang, B. S. (2007). Support vector machine in machine condition monitoring and fault diagnosis. Mechanical Systems and Signal Processing, 21, 2560–2574.
Witten, I. H., Frank, E., & Hall, M. A. (2011). Data mining: Practical machine learning tools and techniques. Los Altos, CA: Morgan Kaufmann.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Ragab, A., Ouali, MS., Yacout, S. et al. Remaining useful life prediction using prognostic methodology based on logical analysis of data and Kaplan–Meier estimation. J Intell Manuf 27, 943–958 (2016). https://doi.org/10.1007/s10845-014-0926-3
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10845-014-0926-3