1Universidad Nacional de Colombia, Facultad de Ciencias, Escuela de Estadística, Medellín, Colombia. Profesor Asistente. Email: cmlopera@unal.edu.co
2Universidad Nacional de Colombia, Facultad de Minas, Escuela de Ingeniería de la Organización, Medellín, Colombia. Profesor Asistente. Email: ecmanota@unal.edu.co
Los datos recurrentes surgen cuando una unidad (o un grupo de tales unidades) es monitoreada a través del tiempo y un evento particular (o grupo de eventos) ocurre en varios puntos del periodo de observación, por ejemplo, los tiempos de episodios recurrentes de una enfermedad en pacientes o los tiempos de reparación de un producto manufacturado. Muchos sistemas, subsistemas y componentes (que genéricamente son denominadas "unidades") tienen asociadas más de una causa o modo de falla. En algunas aplicaciones, y para ciertos propósitos, es importante distinguir entre las causas o modos de falla. Para mejorar la confiabilidad, es esencial identificar la causa de falla hasta el nivel de componente, y en muchas aplicaciones, hasta la causa física real de una falla. En este trabajo, se presenta una aplicación del análisis de datos recurrentes realizado sobre interruptores tipo FL245 (unidades reparables en Interconexión Eléctrica S.A., ISA), que incluye el uso de métodos estadísticos no paramétricos y paramétricos considerando varios modos de falla.
Palabras clave: confiabilidad, cópula, identificabilidad, riesgos competitivos.
Recurrent data arise when a unit (or group of such units) is monitored over time and a particular event (or group of events) occurs at several points of the observation period, for example, times of recurrent episodes of a disease in patients or times of repair of a manufactured product. Many systems, subsystems, and components (which we generically refer to as "units") have more than one cause of failure. In some applications and for some purposes it is important to distinguish between the different failure causes (sometimes referred to as "failure modes"). For purposes of improving reliability, it is essential to identify the cause of failure up to the component level and, in many applications, up to the actual physical cause of failure. This paper presents an application of the recurrent data analysis performed on type FL245 switches (repairable units of Interconexión Eléctrica S.A., ISA), that includes the use of non-parametric and parametric statistical methods, considering several failure modes.
Key words: Competing risks, Copula, Identifiability, Reliability.
Texto completo disponible en PDF
Referencias
]]>1. Andersen, K., Borgan, O., Gill, R. D. & Keiding, N. (1993), Statistical Models Based on Counting Processes, Springer-Verlag, New York. [ Links ]
2. Cook, R. J., Lawless, J. F. & Nadeau, C. (1996), `Robust tests for treatment comparisons based on recurrent event responses´, Biometrics 52, 557-571. [ Links ]
3. Escobar, L., Villa, E. & Yañez, S. (2003), `Confiabilidad: historia, estado del arte y desafíos futuros´, Dyna 70(140), 5-21. [ Links ]
4. Hu, X. J., Lawless, J. F. & Suzuki, K. (1998), `Nonparametric estimation of a lifetime distribution when censoring times are missing´, Technometrics 40, 3-13. [ Links ]
5. Kalbfleisch, J. D., Lawless, J. F. & Robinson, J. A. (1991), `Methods for the analysis and prediction of warranty claims´, Technometrics 33, 273-285. [ Links ]
6. Kvam, P. H., Singh, H. & Whitaker, L. R. (2002), `Estimating distributions with increasing failure rate in an imperfect repair model´, Lifetime Data Analysis 8, 53-67. [ Links ]
7. Lawless, J. F. (1995a), `The analysis of recurrent events for multiple subjects´, Journal of Applied Statistics 44, 487-498. [ Links ]
8. Lawless, J. F. (1995b), `Adjustments for reporting delays and the prediction of occurred but not reported events´, Canadian Journal of Statistics 22, 15-31. [ Links ]
9. Lawless, J. F. & Nadeau, C. (1995), `Some simple robust methods for the analysis of recurrent events´, Technometrics 37, 158-168. [ Links ]
10. Lin, D. Y., Wei, L. J., Yang, I. & Ying, Z. (2000), `Semiparametric regression for the mean and rate functions of recurrent events´, Journal of the Royal Statistical Society: Series B 62, 711-730. [ Links ]
11. Manotas, E. & Lopera, C. (2004), `Metodología para el análisis de datos de recurrencia aplicada a los interruptores FL245´, Interconexión Eléctrica S.A. E.S.P., 1-145. Documento DGM-2004-I-S-ET-023-V1. [ Links ]
12. Meeker, W. & Escobar, L. (1998), Statistical Methods for Reliability Data, John Wiley & Sons, New York. [ Links ]
13. Meeker, W. & Escobar, L. (2003), `SPLIDA: SPLUS Life Data Analysis´, . *Disponible en www.public.iastate.edu/\simsplida, accedido en 01/09/2010 [ Links ]
14. Nelson, W. (1988), `Graphical analysis of system repair data´, Journal of Quality Technology 20, 24-35. [ Links ]
15. Nelson, W. (1990), `Hazard plotting of left truncated data´, Journal of Quality Technology 22, 230-238. [ Links ]
16. Nelson, W. (1995), `Confidence limits for recurrence data: Applied to cost or number of repairs´, Technometrics 37, 147-157. [ Links ]
17. Nelson, W. (1998), `An application of graphical analysis of repair data´, Quality and Reliability Engineering International 14, 49-52. [ Links ]
18. Nelson, W. (2000), `Graphical comparison of sets of repair data´, Quality and Reliability Engineering International 16, 235-241. [ Links ]
19. Nelson, W. (2003), Recurrent Events Data Analysis for Product Repairs, Disease Recurrences, and Other Applications, ASA-SIAM, Philadelphia, ASA, Alexandria. [ Links ]
20. Ross, D. P. (1989a), `Pilot Study of Commercial Water-Loop Heat Pump Compressor Life´, Report under EPRI contract RP 2480-06, Electric Power Research Institute, Palo Alto, CA. [ Links ]
21. Ross, D. P. (1989b), Pilot Study of Commercial Water-Loop Heat Pump Compressor Life, Technical Report , Electric Power Research Institute, Palo Alto, California. [ Links ]
22. Tobias, P. A. & Trindade, D. C. (1995), Applied Reliability, 2nd edn, CRC Press, Boca Raton, FL. [ Links ]
23. Trindade, D. C. & Haugh, L. D. (1980), `Estimation of the reliability of computer components from field renewal data´, Microelectronics Reliability 20, 205-218. [ Links ]
24. Vallarino, C. R. (1988), Confidence Bands for a Mean Value Function Estimated from a Sample of Right-Censored Poisson Processes, Technical Report 02.1471 , IBM General Products Division, San Jose, California. [ Links ]
25. Wang, M. C., Qin, J. & Chiang, C. T. (2001), `Analyzing recurrent events data with informative censoring´, Journal of the American Statistical Association 96, 1057-1065. [ Links ]
Este artículo se puede citar en LaTeX utilizando la siguiente referencia bibliográfica de BibTeX:
@ARTICLE{RCEv34n2a03,
AUTHOR = {Lopera, Carlos M. and Manotas, Eva Cristina},
TITLE = {{Aplicación del análisis de datos recurrentes sobre interruptores FL245 en Interconexión Eléctrica S.A.}},
JOURNAL = {Revista Colombiana de Estadística}, ]]>
YEAR = {2011},
volume = {34},
number = {2},
pages = {249-266}
}