Reliability Approximation through the Discretization of Random Variables using Reversed Hazard Rate Function

Authors: Tirthankar Ghosh, Dilip Roy, Nimai Kumar Chandra

Abstract:

Sometime it is difficult to determine the exact reliability for complex systems in analytical procedures. Approximate solution of this problem can be provided through discretization of random variables. In this paper we describe the usefulness of discretization of a random variable using the reversed hazard rate function of its continuous version. Discretization of the exponential distribution has been demonstrated. Applications of this approach have also been cited. Numerical calculations indicate that the proposed approach gives very good approximation of reliability of complex systems under stress-strength set-up. The performance of the proposed approach is better than the existing discrete concentration method of discretization. This approach is conceptually simple, handles analytic intractability and reduces computational time. The approach can be applied in manufacturing industries for producing high-reliable items.

Keywords: Discretization, Reversed Hazard Rate, Exponential distribution, reliability approximation, engineering item.

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1088310

References:

[1] D. H. Evans, ”Statistical tolerancing: the state of the art: Part II, Methods of estimating moments,” Journal of Quality Technology, 1975, Vol. 7, pp. 1-12.
[2] G. Taguchi, ”Performance analysis and design,” Int. Journal of Production Research, 1978, Vol. 16, pp. 521-530.
[3] J. R. Drrico and N. A. Zaino Jr., ”Statistical tolerancing using a modification of Taguchi’s method,” Technometrics, 1988, Vol. 30, pp. 397-405.
[4] J. R. English, T. Sargent, and T. L. Landers, ”A discretizing approach for stress/strength analysis,”IEEE Trans. on Reliability, 1996, Vol. 45, pp. 84-89.
[5] T. Ghosh and D. Roy, ”Statistical tolerancing through discretization of variables,” International Journal. of Quality and Reliability Management, 2011,Vol. 28, pp 220-232.
[6] D. Roy and R. P. Gupta ”Classification of discrete lives,” Microelectronics and Reliability, 2000, Vol.52, pp. 1459-1473.
[7] D. Roy, ”Reliability measures in the discrete bivariate set- up and related characterization results for a Bivariate Geometric distribution,” Journal of Multivariate Analysis, 1993, Vol. 46, pp. 362-373.
[8] A.W. Kemp, ”Characterization of a discrete normal distribution,” 1997, Vol. 63, pp. 223-229.
[9] S.Inusah and T. J. Kozubowski, ”A discrete analogue of the Laplace distribution,” Journal of Statistical Planning and Inference, 2006, Vol. 136, pp. 1090-1102.
[10] D. Roy and T. Dasgupta, ”A discretizing approach for evaluating reliability of complex systems under stress-strength model,” IEEE Trans. on Reliability, 2001, Vol. 50, pp. 145-150.
[11] D. Roy, ”The discrete normal distribution,” Communications in Statistics (Theory and Methods), 2003, Vol. 32, pp. 50-53.
[12] D. Roy, and T. Dasgupta, ”Evaluation of reliability of complex systems by means of a discretizing approach: Weibull set-up,” Int. J. of Quality and Reliability Management, 2002, Vol. 19, pp. 792-801.
[13] D. Roy, ”Discretization of continuous distributions with an application to stress-strength analysis,” Calcutta Statistical Association Bulletin, 2002, Vol. 52, pp. 297-313.
[14] D. Roy, ”Discrete Rayleigh distribution,” IEEE Trans. on Reliability,, 2004, Vol. 53 pp. 255-260.
[15] D. Roy, and T. Ghosh, ”A new discretization approach with application in reliability estimation,” IEEE Trans. on Reliability, 2009, Vol. 58, pp. 456-461.
[16] T. Ghosh, D. Roy and N. K. Chandra, ”Discretization of Random Variables with Applications”, Calcutta Statistical Association Bulletin, 2011, Vol. 63, pp. 323-338.
[17] Kapur K.C and Lamberson L.R., Reliability in Engineering Design, New York: John Wiley & Sons, 1977.