Skip to main content
Log in

Time-dependent stress–strength reliability models based on phase type distribution

  • Original paper
  • Published:
Computational Statistics Aims and scope Submit manuscript

Abstract

In many of the real-life situations, the strength of a system and stress applied to it changes as time changes. In this paper, we consider time-dependent stress–strength reliability models subjected to random stresses at random cycles of time. Each run of the system causes a change in the strength of the system over time. We obtain the stress–strength reliability of the system at time t when the initial stress and initial strength of the system follow continuous phase type distribution and the time taken for completing a run, called the cycle time, is a random variable which is assumed to have exponential, gamma or Weibull distribution. Using simulated data sets we have studied the variation in stress–strength reliability at different time points corresponding to different sets of parameters of the model.

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

Access this article

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

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11

Similar content being viewed by others

References

  • Abd Elfattah AM, Mohamed MO (2008) Estimating of \(P(Y < X)\) in the exponential case based on censored samples. arXiv:0801.0922v1 [stat.ME]

  • Ahmadi R (2014) Optimal maintenance scheduling for a complex manufacturing system subject to deterioration. Ann Oper Res 217:1–29

    Article  MathSciNet  MATH  Google Scholar 

  • Asmussen S, Nerman O, Olsson M (1996) Fitting phase-type distributions via the EM algorithm. Scand J Stat 23:419–441

    MATH  Google Scholar 

  • Awad AM, Gharraf MK (1986) Estimation of \(P(Y<X)\) in the burr case: a comparative study. Commun Stat Simul Comput 15:389–403

    MathSciNet  MATH  Google Scholar 

  • Baklizi A (2008) Estimation of \(Pr(X<Y)\) using record values in the one and two parameter exponential distributions. Commun Stat Theory Methods 37:692–698

    MathSciNet  MATH  Google Scholar 

  • Barbiero A (2011) Interval estimators for reliability: the bivariate normal case. J Appl Stat 39:501–512

    Article  MathSciNet  Google Scholar 

  • Barron Y (2015) Group replacement policies for repairable systems with leadtimes. IISE Trans 47:1139–1151

    Article  Google Scholar 

  • Barron Y (2018) Group maintenance policies for an R-out-of-N system with phase-type distribution. Ann Oper Res 261:79–105

    Article  MathSciNet  MATH  Google Scholar 

  • Barron Y, Yechiali U (2017) Generalized control-limit preventive repair policies for deteriorating cold and warm standby Markovian systems. IISE Trans 49:1031–1049

    Article  Google Scholar 

  • Barron Y, Frostig E, Levikson B (2004) Analysis of R-out-of-N repairable systems: the case of phase-type distribution. Adv Appl Probab 36:116–138

    Article  MathSciNet  MATH  Google Scholar 

  • Barron Y, Frostig E, Levikson B (2006) Analysis of r out of n systems with several repairmen, exponential life times and phase type repair times: an algorithmic approach. Eur J Oper Res 169(1):202–225

    Article  MathSciNet  MATH  Google Scholar 

  • Birnbaum ZM (1956) On a use of the Mann–Whitney statistic. In: Proceedings of the third Berkeley symposium on mathematical statistics and probability, vol 1, pp 13–17

  • Chao A (1982) On comparing estimators of \(P (Y<X)\) in the exponential case. IEEE Trans Reliab 31:389–392

    MATH  Google Scholar 

  • Constantine K, Karson M, Tse S (1986) Estimation of \(P (Y<X)\) in the gamma case. Commun Stat Simul Comput 15:365–388

    MathSciNet  MATH  Google Scholar 

  • Eryilmaz S (2013) On stress–strength reliability with a time-dependent strength. J Qual Reliab Eng 2013:1–6

    Article  Google Scholar 

  • Eryilmaz S (2017) Phase type stress–strength models with reliability applications. Commun Stat Simul Comput. https://doi.org/10.1080/03610918.2017.1300266

    Article  Google Scholar 

  • Esparza LJR, Bladt M, Nielsen BF (2011) Fisher information and statistical inference for phase-type distribution. J Appl Probab 48(A):277–293

    Article  MathSciNet  MATH  Google Scholar 

  • Ghitany ME, Al-Mutairi DK, Aboukhamseen SM (2015) Estimation of the reliability of a stress–strength system from power Lindley distributions. Commun Stat Simul Comput 44:118–136

    Article  MathSciNet  MATH  Google Scholar 

  • Gopalan MN, Venkateswarlu P (1982) Reliability analysis of time-dependent cascade system with deterministic cycle times. Microelectron Reliab 22(4):841–872

    Article  Google Scholar 

  • Gopalan MN, Venkateswarlu P (1983) Reliability analysis of time-dependent cascade system with random cycle times. Microelectron Reliab 23(2):355–366

    Article  Google Scholar 

  • Hanagal DD (1997) Note on estimation of reliability under bivariate Pareto stress–strength model. Stat Pap 38:453–459

    Article  MathSciNet  MATH  Google Scholar 

  • Jose JK, Xavier T, Drisya M (2019) Estimation of stress–strength reliability using Kumaraswamy half-logistic distribution. J Probab Stat Sci 17(2):141–154

    Google Scholar 

  • Jose JK, Drisya M, Manoharan M (2020) Estimation of stress–strength reliability using discrete phase type distribution. Commun Stat Theory Methods. https://doi.org/10.1080/03610926.2020.1749663

    Article  Google Scholar 

  • Kelley GD, Kelley JA, Schucany WR (1976) Efficient estimation of \( P(Y<X) \) in the exponential case. Technometrics 18:359–360

    MathSciNet  MATH  Google Scholar 

  • Kotz S, Lumelskii Y, Pensky M (2003) The stress–strength model and its generalizations: theory and applications. World Scientific Publishing Co. Pvt. Ltd., Singapore

    Book  MATH  Google Scholar 

  • Kundu D, Raqab MZ (2009) Estimation of R = \(Pr(Y < X)\) for three- parameter weibull distribution. Stat Probab Lett 79:1839–1846

    MATH  Google Scholar 

  • Latouche G, Ramaswami V (1999) Introduction to matrix analytic methods in stochastic modeling. ASA-SIAM, Philadelphia

    Book  MATH  Google Scholar 

  • Lomnicki ZA (1966) A note on the weibull renewal process. Biometrica 53:375–381

    Article  MathSciNet  MATH  Google Scholar 

  • Montoya JA, Rubio FJ (2014) Nonparametric inference for \(P(X<Y)\) with paired variables. J Data Sci 12:359–375

    MathSciNet  Google Scholar 

  • Nadarajah S (2005) Reliability for some bivariate beta distributions. Math Probl Eng 2005:101–111

    Article  MathSciNet  MATH  Google Scholar 

  • Nadarajah S, Kotz S (2005) Reliability for a bivariate gamma distributions. Econ Qual Control 2005:111–119. https://doi.org/10.1155/MPE.2005.151

    Article  MathSciNet  MATH  Google Scholar 

  • Nadarajah S, Kotz S (2006) Reliability for some exponential distributions. Math Probl Eng 2006:1–14. https://doi.org/10.1155/MPE/2006/41652

    Article  MathSciNet  MATH  Google Scholar 

  • Neuts MF (1981) Matrix geometric solutions in stochastic models. An algorithmic approach, vol 2. Jhons Hopkins University Press, Baltimore, MD

    MATH  Google Scholar 

  • Owen DB, Craswell KJ, Hanson DL (1964) Non-parametric upper confidence bounds for \(P(Y<X)\) when \(X\) and \(Y\) are normal. J Am Stat Assoc 59:906–924

    MATH  Google Scholar 

  • Pak A, Khoolenjani NB, Jafari AA (2014) Inference on \(P(Y < X)\) in bivariate Rayleigh distribution. Commun Stat Theory Methods 43(22):4881–4892

    MathSciNet  MATH  Google Scholar 

  • Rezaei S, Noughabi RA, Nadarajah S (2015) Estimation of stress–strength reliability for the generalized Pareto distribution based on progressively censored samples. Ann Data Sci 2(1):83–101

    Article  Google Scholar 

  • Shoaee S, Khorram E (2015) Stress–strength reliability of a two-parameter bathtub-shaped lifetime distribution based on progressively censored samples. Commun Stat Theory Methods 44(24):5306–5328

    Article  MathSciNet  MATH  Google Scholar 

  • Siju KC, Kumar M (2016) Reliability analysis of time dependent stress–strength model with random cycle times. Perspect Sci 8:654–657

    Article  Google Scholar 

  • Siju KC, Kumar M (2017) Reliability computation of a dynamic stress-strength model with random cycle times. Int J Pure Appl Math 117:309–316

    Google Scholar 

  • Tong H (1974) A note on the estimation of \(Pr(Y<X)\) in the exponential case. Technometrics 16:54–56

    MathSciNet  MATH  Google Scholar 

  • Van Noortwijk JM (2009) A survey of the application of gamma processes in maintenance. Reliab Eng Syst Saf 94:2–21

    Article  Google Scholar 

  • Wong A (2012) Interval estimation of \(P(Y<X)\) for generalized Pareto distribution. J Stat Plan Inference 142:601–607

    MathSciNet  MATH  Google Scholar 

  • Xavier T, Jose JK (2020) A study of stress-strength reliability using a generalization of power transformed half-logistic distribution. Commun Stat Theory Methods. https://doi.org/10.1080/03610926.2020.1716250

    Article  Google Scholar 

  • Yadav RPS (1973) A reliability model for stress strength problem. Microelectron Reliab 12(2):119–123

    Article  Google Scholar 

  • Yoo YK (2011) Operating characteristics of a failure counting group replacement policy. Int J Syst Sci 42:499–506

    Article  MathSciNet  MATH  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Joby K. Jose.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendix

Appendix

Proof of Property 2.3

Let \(\lambda _1, \lambda _2,\ldots , \lambda _n\) are the distinct eigenvalues of A. Then we have \(A=SDS^{-1}\), where \(D = diag(\lambda _1, \lambda _2,\ldots , \lambda _n)\) and S is a nonsingular matrix.

Hence

$$\begin{aligned} e^{At}= & {} e^{SDS^{-1}t} = \sum _{0}^{\infty } \frac{(SDS^{-1}t)^{n}}{n!} \nonumber \\= & {} S \sum _{0}^{\infty } \frac{(Dt)^{n}}{n!}S^{-1}= S e^{Dt}S^{-1} \nonumber \\= & {} S \, diag(e^{\lambda _1 t},e^{\lambda _2 t},\ldots ,e^{\lambda _n t})S^{-1} \end{aligned}$$
(43)

Hence

$$\begin{aligned} \int _{0}^{\infty } e^{At} dt= & {} S \, diag\left( \int _{0}^{\infty }e^{\lambda _1 t} dt,\int _{0}^{\infty }e^{\lambda _2 t}dt,\ldots ,\int _{0}^{\infty }e^{\lambda _n t}dt\right) S^{-1}\nonumber \\= & {} S \, diag\left( -\frac{1}{\lambda _1},-\frac{1}{\lambda _2},\ldots ,-\frac{1}{\lambda _n}\right) S^{-1} \nonumber \\= & {} S (-D)^{-1} S^{-1} = (-(S D S^{-1}))^{-1} \nonumber \\= & {} (-A)^{-1}. \end{aligned}$$
(44)

\(\square \)

Proof of Property 2.4

Proceeding as in the above case, we can write

$$\begin{aligned} e^{At} = S \, diag(e^{\lambda _1 t},e^{\lambda _2 t},\ldots ,e^{\lambda _n t})S^{-1} \end{aligned}$$
(45)

and similarly,

$$\begin{aligned} \int _{0}^{x} e^{At} dt= & {} S \, diag\left( \int _{0}^{x}e^{\lambda _1 t} dt,\int _{0}^{x}e^{\lambda _2 t}dt,\ldots ,\int _{0}^{x}e^{\lambda _n t}dt\right) S^{-1}\nonumber \\= & {} S \, diag\left( \frac{(e^{\lambda _1 x}-1)}{\lambda _1},\frac{(e^{\lambda _2 x}-1)}{\lambda _2},\ldots ,\frac{(e^{\lambda _n x}-1)}{\lambda _n}\right) S^{-1} \nonumber \\= & {} S D^{-1}\left( e^{Dx}-I\right) S^{-1}\nonumber \\= & {} S D^{-1} S^{-1} S e^{Dx} S^{-1}-S D^{-1} I S^{-1} \nonumber \\= & {} (S D S^{-1})^{-1} (S e^{Dx} S^{-1})-(S D S^{-1})^{-1}\nonumber \\= & {} A^{-1} (e^{Ax}-I) \end{aligned}$$
(46)

where I is an identity matrix of order n. \(\square \)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Jose, J.K., Drisya, M. Time-dependent stress–strength reliability models based on phase type distribution. Comput Stat 35, 1345–1371 (2020). https://doi.org/10.1007/s00180-020-00991-3

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00180-020-00991-3

Keywords

Navigation