Multivariate conditional aging intensity functions and load-sharing models

Abstract

The aging intensity functions analyze the aging property quantitatively, in the sense that the larger the aging intensity, the stronger the tendency of aging. They are useful tools to describe reliability properties of distributions. In the literature, the aging intensity functions have been studied in the univariate and bivariate case but without considering the possibility of observing a dynamic history. In this paper, the concept of aging intensity function is extended to the multivariate case by the use of the multivariate conditional hazard rate functions. Some properties of those functions are studied and a focus on the bivariate case is performed. Finally, the multivariate conditional aging intensity functions are studied for the order dependent version of the time-homogeneous load-sharing model and a study on the comparison among surviving components in a system is provided. 

Keywords

• Aging intensity function
• hazard rate function
• time-homogeneous load-sharing model

References

1. [1] S. Bhattacharjee, A.K. Nanda and S.K. Misra, Reliability analysis using ageing intensity function, Statist. Probab. Lett. 83 (5), 1364–1371, 2013.
2. [2] F. Buono, M. Longobardi and M. Szymkowiak, On generalized reversed aging intensity functions, Ric. Mat. 71, 85–108, 2022.
3. [3] D.R. Cox, Regression models and life-tables, J. R. Stat. Soc. Ser. B. Stat. Methodol. 34 (2), 187–220, 1972.
4. [4] R. Foschi, G. Nappo and F.L. Spizzichino, Diagonal sections of copulas, multivariate conditional hazard rates and distributions of order statistics for minimally stable lifetimes, Depend. Model. 9, 394–423, 2021.
5. [5] R.L. Giri, A.K. Nanda, M. Dasgupta, S.K. Misra and S. Bhattacharjee, On ageing intensity function of some Weibull models, Comm. Statist. Theory Methods, Doi:10.1080/03610926.2021.1910845, 2021.
6. [6] R. Jiang, P. Ji and X. Xiao, Aging property of unimodal failure rate models, Reliab. Eng. Syst. Saf. 79 (1), 113–116, 2003.
7. [7] N.L. Johnson and S. Kotz, A vector multivariate hazard rate, J. Multivariate Anal. 5, 53–66, 1975.
8. [8] S. Khardani and A. Benkhaled, A nonparametric estimation of the conditional ageing intensity function in censored data: A local linear approach, Math. Slovaca 71 (2), 429–438, 2021.

9. [9] S. Nadarajah and S. Kotz, Reliability for some bivariate exponential distributions, Math. Probl. Eng., Doi:10.1155/MPE/2006/41652, 2006.
10. [10] A.K. Nanda, S. Bhattacharjee and S.S Alam, Properties of aging intensity function, Statist. Probab. Lett. 77 (4), 365–373, 2007.
11. [11] M. Rezaei and V.A. Khalef, On the reversed average intensity order, JSRI 11, 25–39, 2014.
12. [12] S.M. Ross, A model in which component failure rates depend on the working set, Nav. Res. Logist. Q. 31 (2), 297–300, 1984.
13. [13] Z. Schechner, A load-sharing model: the linear breakdown rule, Nav. Res. Logist. Q. 31, 137–144, 1984.
14. [14] M. Shaked, and J.G. Shanthikumar, Multivariate stochastic orderings and positive dependence in reliability theory, Math. Oper. Res. 15 (3), 545–552, 1990.
15. [15] M. Shaked and J.G. Shanthikumar, Multivariate conditional hazard rate functions - an overview, Appl. Stoch. Models Bus. Ind. 31 (3), 285–296, 2015.
16. [16] F.L. Spizzichino, Reliability, signature, and relative quality functions of systems under timehomogeneous load-sharing models, Appl. Stoch. Models Bus. Ind. 35 (2), 158–176, 2018.
17. [17] S.M. Sunoj and R.S. Rasin, A quantile-based study on ageing intensity function, Comm. Statist. Theory Methods 47 (22), 5474–5484, 2018.
18. [18] M. Szymkowiak, Characterizations of distributions through aging intensity, IEEE Trans. Rel. 67 (2), 446–458, 2018.
19. [19] M. Szymkowiak, Generalized aging intensity functions, Reliab. Eng. Syst. Saf. 178, 198–208, 2018.
20. [20] M. Szymkowiak, Lifetime Analysis by Aging Intensity Functions, Springer, 2020.