Skip to main content
Log in

Availability of 1-for-2 shared protection systems with general repair-time distributions

  • OriginalPaper
  • Published:
Telecommunication Systems Aims and scope Submit manuscript

Abstract

In this paper, we investigate the availability of 1-for-2 shared protection systems. We assume that there are 2 working units each serving a single user and one shared protection (spare) unit in the system. We also assume that the time to failure is subject to an exponential distribution and that the time to repair is subject to a general distribution. Under these assumptions, we derive the availability for each user by combining the state transition analysis and the supplementary variable method. We also show the effect of the repair time distribution, the failure rates and the repair rates of the units through the case study of small-sized 2 enterprises that share one spare device for backup.

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.

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

Similar content being viewed by others

References

  1. Moskowitz, F., & McLean, J. B. (1956). Some reliability aspects of systems design. IRE Transactions on Reliability and Quality Control, RQC–8, 7–35.

    Article  Google Scholar 

  2. Denson, W. (1998). The history of reliability prediction. IEEE Transactions on Reliability, 47(3–SP), SP321–SP328.

    Google Scholar 

  3. Eiger, M. I., Luss, H., & Shallcross, D. F. (2012). Network restoration under dual failures using path-protecting preconfigured cycles. Telecommunication Systems, 49, 271–286.

    Article  Google Scholar 

  4. Srivastava, S. (2011). Quality of service based protection in optical networks using diversity constraint. Telecommunication Systems., 46, 317–331.

    Article  Google Scholar 

  5. Secci, S., & Sanso, B. (2011). Survivability and reliability of a composite-star transport network with disconnected core switches. Telecommunication Systems, 46, 43–59.

    Article  Google Scholar 

  6. Oommen, B. J., & Misra, S. (2010). Fault-tolerant routing in adversarial mobile ad hoc networks: An efficient route estimation scheme for non-stationary environments. Telecommunication Systems, 44, 159–169.

    Article  Google Scholar 

  7. Linton, D. G., & Saw, J. G. (1974). Reliability analysis of the k-out-of-n: F system. IEEE Transactions on Reliability, R–23(2), 97–103.

    Article  Google Scholar 

  8. Dharmadhikari, A., & Gupta, Y. (1985). Analysis of 1-out-of-2: G warm-standby repairable system. IEEE Transactions on Reliability, R–34(5), 550–553.

    Article  Google Scholar 

  9. Linton, D. G. (1989). Generalized reliability results for 1-out-of-n: G repairable systems. IEEE Transaction of Reliability, 38(4), 468–471.

    Article  Google Scholar 

  10. Hwang, F. K. (1991). An explicit solution for the number of minimal p-cutsequences in a consecutive-k-out-of-n: F system. IEEE Transactions on Reliability, 40(5), 553–554.

    Article  Google Scholar 

  11. Hussain, I. (2005). Fault-tolerant IP and MPLS networks. Indianapolis: Cisco Press.

    Google Scholar 

  12. Lee, K.-K., Fung, L., & Ong, B.-H. (2006). Building resilient IP networks. Indianapolis: Cisco Press.

    Google Scholar 

  13. Fawaz, W., Martignon, F., Chen, K., & Pujolle, G. (2005). A novel protection scheme for quality of service aware networks. IEEE International Conference on Communications ICC2005, 3, 1720–1725.

    Google Scholar 

  14. Zhang, Jing, Zhu, Keyao, Zang, Hui, Matloff, N. S., & Mukherjee, B. (2007). Availability-aware provisioning strategies for differentiated protection service in wavelength-convertible WDM mesh networks. IEEE/ACM Transactions on Networking, 15(5), 1177–1190.

    Article  Google Scholar 

  15. Wohl, J. G. (1966). System operational readiness and equipment dependability. IEEE Transactions on Reliability, R–15(1), 1–6.

    Article  Google Scholar 

  16. Sarma, V. V. S., & Alam, M. (1975). Optimal maintenance policies for machines subject to deterioration and intermittent breakdowns. IEEE Transactions on Systems, Man, and Cybernetics, SMC–5(3), 396–398.

    Article  Google Scholar 

  17. Alam, M., & Sarma, V. V. S. (1977). An application of optimal control theory to repairman problem with machine interference. IEEE Transactions on Reliability, R–26(2), 121–124.

    Article  Google Scholar 

  18. Abrams, M., Page, E. H., & Nance, R. E. (1991). Simulation Program Development by Stepwise Refinement in Unity., Proc. 23rd Winter Simulation Conference, pp. 233–242

  19. Kouikoglou, V. S., & Phillis, Y. S. (1994). Discrete event modeling and optimization of unreliable production lines with random rates. IEEE Transactions on Reliability, 10(2), 153–159.

    Google Scholar 

  20. Chung, W. K. (1991). Reliability analysis of a series system with repair. Microelectronics Reliability, 31(2—-3), 363–365.

    Article  Google Scholar 

  21. Ozaki, H., & Kara, A. (2007). Computing the availability and MTTF of shared protection systems. Proceedings of the IEEE Compter Science and Information Technology (pp. 480–485)

  22. Ozaki, H., Kara, A., & Cheng, Z. (2009). User-perceived reliability of M-for-N (M:N) shared protection systems. IEICE Transactions on Information Systems, E92—-D3, 443–450.

    Article  Google Scholar 

  23. Ozaki, H., Kara, A., & Cheng, Z. (2009). User-perceived reliability of shared protection systems with a finite number of repairers. Proceedings of the IEEE Comptuter and Information Technology (pp. 3–8)

  24. Ozaki, H., Kara, A., & Cheng, Z. (2011). User-perceived reliability of priority shared protection systems. International Journal of Quality & Reliability Management., 28(1), 95–108.

  25. Ozaki, H., Kara, A., & Cheng, Z. (2011). User-perceived reliability of shared protection systems with the Erlang type-\(k\) repair scheme. International Journal of Information and Communication Technology., 3(3), 209–226.

  26. Ozaki, H., Kara, A., & Cheng, Z. (2011). Service-level assurance of high availability multi-unit systems using the M-for-N backup scheme. Quality and Reliability Engineering International (ISSN: 0748–8017), Wiley 2011 27(7): 895–903

  27. Birolini, A. (2004). Reliability engineering: Theory and practice (4th ed.). Berlin: Springer.

    Book  Google Scholar 

  28. Ozaki, H., & Kara, A. (2012). Reliability analysis of \(1\)-for-\(2 \)shared protection systems with general repair-time distributions. Applied Mathematical Modelling, 36(1), 333–347.

    Article  Google Scholar 

  29. Ozaki, H., & Kara, A. (2011). Reliability analysis of \(M\)-for-\(N\) shared protection systems with general repair time distributions. IEEE Transactions on Reliability, 60(3), 647–657.

    Article  Google Scholar 

  30. Kleinrock, L. (1962). Queueing systems, volume 1: Theory. New York: Wiley.

    Google Scholar 

  31. Hundsdorfer, W., & Verwer, J.-G. (2003). Numerical solution of time-dependent advection–diffusion–reaction equations. Berlin: Springer.

    Book  Google Scholar 

  32. Rausand, M., & Høyland, A. (2004). System reliability theory, models, statistical methods, and applications. Hoboken: Wiley.

    Google Scholar 

  33. Press, W. H., Flannery, B. P., Teukolsky, S. A., & Vetterling, W. T. (1988). Numerical recipes in C. Cambridge: Cambridge University Press.

    Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Hirokazu Ozaki.

Appendices

Appendix 1

Equation (3) is derived from the following expression:

$$\begin{aligned}&p_0 \left( {t+\Delta t} \right) =p_0 \left( t \right) \left\{ {1-3\lambda \Delta t} \right\} \nonumber \\&\quad +\int _{0}^{t} {p_1 \left( {t,x} \right) \left( {1-2\lambda \Delta t} \right) \frac{B\left( {x+\Delta t} \right) -B\left( x \right) }{{\overline{B}} \left( x \right) }dx}. \end{aligned}$$
(36)

In the above equation, the left hand side means the system is in state \(0\) at time \(t+\Delta t\) , while the right hand side means the sum of two events i.e. (1) the system stays in state \(0\) during the interval \((t,t+\Delta t)\), (2) the system moves from state \(1\) to state \(0\) as a broken unit gets repaired at some point in the interval \((t,t+\Delta t)\).

Expression (6) is derived from the following expression and (4):

$$\begin{aligned} p_1 \left( {t+\Delta t,x+\Delta t} \right) \!=\!p_1 \left( {t,x} \right) \left( {1\!-\!2\lambda \Delta t} \right) \frac{1-B\left( {x+\Delta t} \right) }{{\overline{B}} \left( x \right) }.\nonumber \\ \end{aligned}$$
(37)

In the above equation, the left hand side means the user stays in state \(1\) during the interval \((t,t+\Delta t)\), while the right hand side means there is neither failure nor repair during the same interval. By substituting (4) to (37), we obtain the following (38).

$$\begin{aligned} \begin{array}{l} \exp \left\{ {-2\lambda \left( {x+\Delta t} \right) } \right\} g_1 \left( {t+\Delta t,x+\Delta t} \right) {\overline{B}} \left( x \right) \\ \quad =\exp \left\{ {-2\lambda x} \right\} g_1 \left( {t,x} \right) {\overline{B}} \left( x \right) \left( {1-2\lambda \Delta t} \right) \frac{1-B\left( {x+\Delta t} \right) }{{\overline{B}} \left( x \right) } \\ \end{array} \end{aligned}$$
(38)

(38) is equivalent to the following (39) and (6):

$$\begin{aligned} g_1 \left( {t+\Delta t,x+\Delta t} \right)&= g_1 \left( {t,x} \right) +\frac{\partial g_1 }{\partial t}\Delta t\nonumber \\&+\frac{\partial g_1 }{\partial x}\Delta t=g_1 \left( {t,x} \right) \end{aligned}$$
(39)

Equation (10) is derived from the following expression and (4), (8):

$$\begin{aligned} p_2 \left( {t+\Delta t,x+\Delta t} \right)&= p_2 \left( {t,x} \right) \left( {1-\lambda \Delta t} \right) \frac{1-B\left( {x+\Delta t} \right) }{{\overline{B}} \left( x \right) }\nonumber \\&+p_1 \left( {t,x} \right) \frac{1-B\left( {x+\Delta t} \right) }{{\overline{B}} \left( x \right) }2\lambda \Delta t.\nonumber \\ \end{aligned}$$
(40)

In the above equation, the left hand side means the user stays in state \(2\) during the interval \((t,t+\Delta t)\), while the right hand side means the sum of two events i.e. (1) there is neither failure nor repair during the same interval, (2) the system moves from state \(1\) to state \(2\) as a unit was broken at some point in the interval \((t,t+\Delta t)\).By substituting (4) and (8) to (40) and letting \(\Delta t\rightarrow 0\), we obtain (10).

Equation (11) is derived from the following expression and (8), (9):

$$\begin{aligned} p_3 \left( {t+\Delta t,x+\Delta t} \right)&= p_3 \left( {t,x} \right) \frac{1-B\left( {x+\Delta t} \right) }{{\overline{B}} \left( x \right) }\nonumber \\&+p_2 \left( {t,x} \right) \frac{1-B\left( {x+\Delta t} \right) }{{\overline{B}} \left( x \right) }\lambda \Delta t.\nonumber \\ \end{aligned}$$
(41)

In the above equation, the left hand side means the system stays in state \(3\) during the interval \((t,t+\Delta t)\), while the right hand side means the sum of two events i.e. (1) there is no repair during the same interval, (2) the system moves from state \(2\) to state \(3\) as a unit is broken at some point in the interval \((t,t+\Delta t)\).By substituting (8) and (9) to (41) and letting \(\Delta t\rightarrow 0\), we obtain (11).

Equations (14) through (16) are derived from the following expressions and (4), (8), (9), (12), (13):

$$\begin{aligned}&p_1 \left( {t+\Delta t,0} \right) \Delta t=p_0 \left( t \right) 3\lambda \Delta t\nonumber \\&\quad +\int _{0}^{t} {p_2 } \left( {t,x} \right) \left( {1-\lambda \Delta t} \right) \frac{B\left( {x+\Delta t} \right) -B\left( x \right) }{{\overline{B}} \left( x \right) }dx\end{aligned}$$
(42)
$$\begin{aligned}&p_2 \left( {t+\Delta t,0} \right) \Delta t=\int _{0}^{t} {p_3 } \left( {t,x} \right) \frac{B\left( {x+\Delta t} \right) -B\left( x \right) }{{\overline{B}} \left( x \right) }dx,\nonumber \\ \end{aligned}$$
(43)

and

$$\begin{aligned} p_3 \left( {t+\Delta t,0} \right) \Delta t=0. \end{aligned}$$
(44)

In (42), the left hand side means the system arrived in state \(1\) just at the time \(t+\Delta t\) , while the right hand side means the sum of two events i.e. (1) the system moves from state \(0\) to state \(1\) by a failure of a unit within \((t,t+\Delta t)\), (2) the system moves from state \(2\) to state \(1\) as a broken unit gets repaired at some point in the interval \((t,t+\Delta t)\).By substituting (8) to (42) and letting \(\Delta t\rightarrow 0\), we obtain (14).

In (43), the left hand side means the system arrived in state \(2\) just at the time \(t+\Delta t\) , while the right hand side means the system moves from state \(3\) to state \(2\) as a broken unit gets repaired at some point in the interval \((t,t+\Delta t)\). By substituting (9) to (43) and letting \(\Delta t\rightarrow 0\), we obtain (15).

(44) means the system is never in state \(3\) without the elapsed repair time. By letting \(\Delta t\rightarrow 0\), we obtain (16).

Appendix 2

\(A_{1:2} \) (Expression (24) for the case of the Erlang type-\(k\) repair-time distribution (25)) is expressed as follows:

$$\begin{aligned} A_{1:2}&=\frac{\left( {k\mu } \right) ^{k}}{\Delta _{Ek} }\left[ \left( {k\mu } \right) ^{2k}+\frac{3}{2}\left( {\lambda \!+\!k\mu } \right) ^{k}\left\{ {\left( {2\lambda \!+\!k\mu } \right) ^{k}\!+\!\left( {k\mu } \right) ^{k}} \right\} \right. \nonumber \\&\quad \ \left. \times \left\{ {1-\left( {\frac{k\mu }{\lambda +k\mu }} \right) ^{k}} \right\} \right] , \end{aligned}$$
(45a)

where

$$\begin{aligned} \Delta _{Ek}&= 3k\lambda \left\{ {\left( {2\lambda +k\mu } \right) ^{k}+\left( {k\mu } \right) ^{k}} \right\} \\&\left[ \left( {k\mu } \right) ^{k-1}\left( {\lambda +2k\mu } \right) \left( {\lambda +k\mu } \right) ^{k-1}\right. \nonumber \\&\left. +\left( {k\mu } \right) ^{k}\left\{ {\left( {k\mu } \right) ^{k-1}-\left( {\lambda +k\mu } \right) ^{k-1}} \right\} \right] \\&+\left( {k\mu } \right) ^{2k}\!\left\{ \! {\left( {2\lambda +k\mu } \right) ^{k}\!+\!3\lambda k\left( {2\lambda \!+\!k\mu } \right) ^{k-1}\!+\!2\left( {k\mu } \right) ^{k}} \!\right\} \\&-\left( {k\mu } \right) ^{2k}\left\{ {\left( {2\lambda +k\mu } \right) ^{k}+\left( {k\mu } \right) ^{k}} \right\} \\&-3k\lambda \left( {k\mu } \right) ^{2k-1}\left\{ k\mu \left( {2\lambda +k\mu } \right) ^{k-1}\right. \\&\left. +2\left( {2\lambda +k\mu } \right) ^{k}+\left( {k\mu } \right) ^{k} \right\} . \end{aligned}$$

If \(k\) is equal to one, we obtain the following availability in the case of the exponential distribution of repair time.

$$\begin{aligned} A_{1:2} =\frac{\mu ^{3}+3\lambda \mu ^{2}+3\lambda \mu ^{2}}{6\lambda ^{3}+6\lambda ^{2}\mu +3\lambda \mu ^{2}+\mu ^{3}} \end{aligned}$$
(45b)

\(A_{1:2} \) (Expression (24) for the case of the fixed repair-time (27))is expressed as follows:

$$\begin{aligned} A_{1:2}&= \frac{1}{\Delta _F }\left[ \hbox {exp}\left( {-\frac{3\lambda }{\mu }} \right) +\frac{3}{2}\left\{ {1+\hbox {exp}\left( {-\frac{2\lambda }{\mu }} \right) } \right\} \right. \nonumber \\&\times \left. \left\{ {1-\hbox {exp}\left( {-\frac{\lambda }{\mu }} \right) } \right\} \right] , \end{aligned}$$
(45)

where

$$\begin{aligned} \Delta _F \!=\!\hbox {exp}\left( \! {-\frac{3\lambda }{\mu }} \right) +\frac{3\lambda }{\mu }\left\{ {1+\hbox {exp}\left( \! {-\frac{2\lambda }{\mu }} \right) -\hbox {exp}\left( {-\frac{\lambda }{\mu }} \right) } \right\} . \end{aligned}$$

\(A_{1:2} \) (Expression (24) for the case of the sharp-peaked and long-tailed repair-time distribution (29) (\(k=1))\) is expressed as follows:

$$\begin{aligned} A_{1:2}\! =\!\frac{1}{\Delta _{F+E1} }\!\left[ \!\! {\begin{array}{l} \left( {a\mu } \right) ^{3}\exp \!\left( \! {-\frac{3\lambda }{a\mu }} \!\right) \!+\! \frac{3}{2}\left( {a\mu } \right) \left\{ {\left( {a\!-\!1} \right) \lambda \!+\!a\mu } \right\} \\ \quad \times \left\{ {2\left( {a-1} \right) \lambda +a\mu +a\mu \exp \left( {-\frac{2\lambda }{a\mu }} \right) } \right\} \\ \quad \times \left\{ {1-\exp \left( {-\frac{\lambda }{a\mu }} \right) \frac{a\mu }{\left( {a-1} \right) \lambda +a\mu }} \right\} \\ \end{array}} \!\!\right] \!,\nonumber \\ \end{aligned}$$
(46)

where

$$\begin{aligned}&\Delta _{F+E1} =3\lambda \left\{ {2\left( {a-1} \right) \lambda +a\mu +a\mu \exp \left( {-\frac{2\lambda }{a\mu }} \right) } \right\} \\&\quad \times \left\{ {a\left( {a-1} \right) \lambda +a^{2}\mu +a\left( {a-2} \right) \mu \exp \left( {-\frac{\lambda }{a\mu }} \right) } \right\} \\&\quad +\left( {a\mu } \right) ^{2}\exp \left( {-\frac{\lambda }{a\mu }} \right) \left\{ {3\left( {a-1} \right) \lambda -\left( {3\lambda -a\mu } \right) \exp \left( {-\frac{2\lambda }{a\mu }} \right) } \right\} \\&\quad +3a\lambda \mu \exp \left( {-\frac{\lambda }{a\mu }} \right) \left\{ {2\left( {a-1} \right) \lambda +a\mu +a\mu \exp \left( {-\frac{2\lambda }{a\mu }} \right) } \right\} \\&\quad -3a\lambda \mu \exp \left( {-\frac{\lambda }{a\mu }} \right) \left[ 4\left( {a-1} \right) \left\{ {\left( {a-1} \right) \lambda +a\mu } \right\} \right. \\&\quad \left. +a\left( {a-2} \right) \mu \exp \left( {-\frac{2\lambda }{a\mu }} \right) +2\left( {a-1} \right) \lambda -a\left( {a-2} \right) \mu \right] . \end{aligned}$$

\(A_{1:2} \) (Expression (24) for the case of the sharp-peaked and long-tailed repair-time distribution (29) (\(k=2))\) is expressed as follows:

$$\begin{aligned} A_{1:2} \!=\!\frac{1}{\Delta _{F+E2} }\!\left[ \!\! {\begin{array}{l} 4\left( {a\mu } \right) ^{6}\exp \!\left( \! {-\frac{3\lambda }{a\mu }} \!\right) \!+\!\frac{3}{2}\left( {a\mu } \right) ^{2}\left\{ {\left( {a\!-\!1} \right) \lambda \!+\!2a\mu } \right\} ^{2} \\ \quad \times \left[ {\left\{ {\left( {a\!-\!1} \right) \lambda \!+\!a\mu } \right\} ^{2}\!+\!\left( {a\mu } \right) ^{2}\exp \left( {-\frac{2\lambda }{a\mu }} \right) } \right] \\ \quad \times \left[ {1-\exp \left( {-\frac{\lambda }{a\mu }} \right) \left\{ {\frac{2a\mu }{\left( {a-1} \right) \lambda +2a\mu }} \right\} ^{2}} \right] \\ \end{array}} \!\!\right] \!,\nonumber \\ \end{aligned}$$
(47)

where

$$\begin{aligned}&\Delta _{F+E2} =3\lambda \left[ a^{2}\mu \left\{ {\left( {a-1} \right) \lambda +2a\mu } \right\} ^{2}+4\left( {a-2} \right) \left( {a\mu } \right) ^{3}\right. \\&\left. \quad \times \exp \left( {-\frac{\lambda }{a\mu }} \right) \right] \times \left[ {\left\{ {\left( {a-1} \right) \lambda +a\mu } \right\} ^{2}\!+\!\left( {a\mu } \right) ^{2}\exp \left( {-\frac{2\lambda }{a\mu }} \right) } \right] \\&\quad +4\left( {a\mu } \right) ^{4}\exp \left( {-\frac{\lambda }{a\mu }} \right) \Big [ \left\{ {\left( {a-1} \right) \lambda +a\mu } \right\} \left\{ {4\left( {a-1} \right) \lambda +a\mu } \right\} \\&\quad +a\mu \exp \left( {-\frac{2\lambda }{a\mu }} \right) \left( {2a\mu -3\lambda } \right) \Big ] \\&\quad +4\left( {a\mu } \right) ^{3}\left( {3\lambda -a\mu } \right) \exp \left( {-\frac{\lambda }{a\mu }} \right) \left[ \left\{ {\left( {a-1} \right) \lambda +a\mu } \right\} ^{2}\right. \\&\quad \left. +\left( {a\mu } \right) ^{2}\exp \left( {-\frac{2\lambda }{a\mu }} \right) \right] -12\lambda \left( {a\mu } \right) ^{3}\exp \left( {-\frac{\lambda }{a\mu }} \right) \\&\qquad \times \left[ {\begin{array}{l} 2\left( {a-1} \right) \left\{ {\left( {a-1} \right) \lambda +a\mu } \right\} \left\{ {\left( {a-1} \right) \lambda +2a\mu } \right\} \\ \quad +\left( {a-2} \right) \left( {a\mu } \right) ^{2}\exp \left( {-\frac{2\lambda }{a\mu }} \right) +\left\{ {\left( {a-1} \right) \lambda +a\mu } \right\} \\ \quad \left\{ {\left( {a-1} \right) \lambda -a\left( {a-2} \right) \mu } \right\} \\ \end{array}} \right] . \end{aligned}$$

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Ozaki, H. Availability of 1-for-2 shared protection systems with general repair-time distributions. Telecommun Syst 58, 3–16 (2015). https://doi.org/10.1007/s11235-014-9863-x

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11235-014-9863-x

Keywords

Navigation