k-out-of-n Systems

From ReliaWiki
Revision as of 18:21, 31 July 2015 by Kate Racaza (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search
BlockSim Reference Examples Banner.png

K-out-of-n Systems

This example validates the results for k-out-of-n systems in BlockSim's analytical and simulation diagrams.

Reference Case

This data set is from example 4.6 on page 78 in the book Life Cycle Reliability Engineering by Dr. Guangbin Yang, John Wiley & Sons, 2007.


A web host has 5 independent and identical servers connected in parallel, and at least 3 of them must be operational for the web service to be operational. Each server has an exponential life distribution model with a lambda of 2.7 × 10-5 failures per hour. Mean time between failures (MTBF) and the reliability of the web host after one year of continuous operation are calculated.


It is assumed that the server is repaired immediately to a good-as-new condition if failed; therefore, the MTBF is the same as the MTTF and is calculated by using Equation 4.24 on page 78:

[math]MTTF = \frac{1}{\lambda}\sum^{n}_{i=k}\frac{1}{i} = \frac{1}{2.7 \times 10^{-5}}\sum^{5}_{i=3}\frac{1}{i} = 2.9 \times 10^{4}\,\![/math]   hours

Since the time to failure is exponential, system reliability is calculated via Equation 4.23 on page 78 by substituting the given data as:

[math]R(t) = \sum^{n}_{i=k}C^{i}_{n} e^{-\lambda i t} (1-e^{-\lambda t})^{n-i}\,\![/math]
[math]R(8760) = \sum^{5}_{i=3}C^{i}_{5} e^{-2.7 \times 10^{-5} \times 8760i} (1-e^{-2.7 \times 10^{-5} \times 8760})^{5-i} = 0.9336\,\![/math]

Results in BlockSim

In BlockSim, the server system RBD is configured as shown below.

Kn rbd.png

Each server is modeled with an exponential distribution with a lambda of 2.7 × 10-5 failures per hour.

Analytical Proof

The reliability of the web host after one year of continuous operation (8760 hours) is calculated in the QCP as 93.36%, which is the same as the result calculated in the reference book.

Kn qcprel.png

The mean time between failures (MTBF) is estimated to be 29,004 hours.

Kn qcpmtbf.png

Simulation Proof

We can also estimate the results by using the simulation tool in BlockSim. The simulation settings are shown below.

Kn sim.png

The point reliability after 8760 hours of use is estimated to be 93.45%.

Kn simqcprel.png

And the mean time to failure (MTTF) is estimated to be 29,048 hours.

Kn simqcpmttf.png