0% found this document useful (0 votes)
37 views54 pages

System Reliability 1 Student

The document discusses reliability evaluation of systems. It defines key concepts like events, probability axioms, conditional probability, and independence. It describes different types of basic system configurations like series, parallel, series-parallel, k-out-of-n, and standby systems. It introduces reliability block diagrams to represent systems and calculate reliability. Formulas are provided to calculate the reliability of series, parallel and series-parallel systems based on the reliabilities of their components. Methods to evaluate reliability for repairable and non-repairable systems are also presented.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
37 views54 pages

System Reliability 1 Student

The document discusses reliability evaluation of systems. It defines key concepts like events, probability axioms, conditional probability, and independence. It describes different types of basic system configurations like series, parallel, series-parallel, k-out-of-n, and standby systems. It introduces reliability block diagrams to represent systems and calculate reliability. Formulas are provided to calculate the reliability of series, parallel and series-parallel systems based on the reliabilities of their components. Methods to evaluate reliability for repairable and non-repairable systems are also presented.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 54

System Reliability Evaluation

Basics of Probability Theory

Events
 An event E is a collection of zero or more sample
points from S

 S and E are sets  use of set operations.


Algebra of events
 Sample space is a set and events are the subsets of
this (universal) set.
 Mutually exclusive (disjoint) events
Probability axioms


Conditional probability
 In some experiment, some prior information may
be available, e.g.,
 In general,
Mutual Independence
 A and B are said to be mutually independent, iff,

 Also, then,
Independent set of events
 Set of n events, {A1, A2,..,An} are mutually
independent iff, for each

 Complements of such events also satisfy,

 Pair wise independence (not mutually independent)


System reliability evaluation
A general system:

Input System Output


System reliability evaluation
Types of basic system configurations:

 Series system
 Parallel system
 Series-parallel system (mixed system)
 K-out-of-N system
 Standby system
Reliability Block Diagrams: RBDs

 Each component of the system is represented as a block


 System behavior is represented by connecting the blocks
 Blocks that are all required are connected in series
 Blocks among which only one is required are connected in
parallel
 When at least k of them are required are connected as k-of-n
 Failures of individual components are assumed to be
independent
Series system
 Series system: n statistically independent components.

 Let, Ri = P(Ei), then series system reliability:

 For now reliability is simply a probability, later it will be a function of time


Series system

R1 R2 Rn

This simple PRODUCT LAW OF RELIABILITIES,


is applicable to series systems of independent
components.
Series Systems

Rs  Pr ob(1  2  3          n)

RS = R1 R2 ... Rn
n
Rs   Ri
i 1

14
Parallel system

 System consisting of n independent parallel components.


 System fails to function iff all n components fail.
 Ei = "component i is functioning properly"
 Ep = "parallel system of n components is
functioning properly."
 Rp = P(Ep).
Parallel system (Continued)

E p  "The parallel system has failed "


 " All n components have failed "
__ __ __
 E1  E2 ...  En
Therefore:
__ __ __ __
P( E p )  P( E1  E 2 ...  E n )
__ __ __
 P ( E1 ) P( E 2 )... P ( E n )
(1  R1 )(1  R2 )    (1  Rn )
__ n
R p  P( E p )  1  P( E p )  1   1  Ri 
i 1
Parallel system (Continued)

R1
..
.
Input Output
..
.
Rn

• Parallel systems of independent components


follow the PRODUCT LAW OF UNRELIABILITIES
Parallel Systems

Input 2 Output

R p  Pr ob(1  2  3          n)

Rp = 1 - (1 - R1) (1 - R2)... (1 - Rn)


18
Series-Parallel Systems
RC
C
RA RB RD
Input A B D Output
C
RC
 Convert to equivalent series system
RA RB RD
A B C’ D

RC’ = 1 – (1-RC)(1-RC)
Series-Parallel System
 Series-parallel system: n-series stages, each with
ni parallel components.

 Reliability of series parallel system


Series-Parallel system (example)

voice
control
Input Output
voice

control
voice

Example: 2 Control and 3 Voice Channels


Series-Parallel system (Continued)

 Each control channel has a reliability Rc


 Each voice channel has a reliability Rv
 System is up if at least one control channel and at
least 1 voice channel are up.
 Reliability:
R  [1  (1  Rc ) 2 ][1  (1  Rv )3 ] (3)
Exponential Distribution

 Distribution Function:F  t   1  e  t
t0
 Density Function: f t   e  t
t0
 Reliability: R t   e  t
t0
h t  

f t  
Failure Rate:
R t 

failure rate is age-independent (constant)


 MTTF: MTTF  1 / 
Reliability evaluation for repairable and non-
repairable systems

 Can be used to calculate


 Non-repairable system reliability and MTTF given
 Individual block reliabilities
Or Individual block failure rates
 Assuming mutually independent failures events
 Repairable system availability and MTBF given
 Individual block reliabilities
Or Individual block failure rates
 Assuming mutually independent failures events
Series system in RBD
 Series system of n components.
R1 R2 Rn
 Components are statistically independent
 Define event Ei = "component i functions properly.”
 For the series system:
P(" The system is functionin g properly " )
 P( E1  E2  ...  En )
 P( E1 )  P( E2 )...P( En ), by independen ce
Reliability for Series system
 Product law of reliabilities:

where Ri is the reliability of component i

 For exponential Distribution:

n n
Rs   Ri or Rs (t )   Ri (t )
 For weibull Distribution:

i 1 i 1

 i t
  it
if Ri (t )  e then Rs (t )  e i 1

 i t 
(  i ) t 
if Ri (t )  e then Rs (t )  e i 1
MTTF for Series System

 Assuming exponential failure-time


distribution with constant failure rate i
for each component, then:
1
MTTF  n


i 1
i
Parallel system in RBD
 A system consisting of n independent components in
parallel.
R1
..
.
 It will fail to function only if all n components have failed.
 Ei = “The component i is functioning” ..
.
 Ep = "the parallel system of n component is functioning
properly."
Rn
Parallel system in RBD(Continued)
E p  " The parallel system has failed "
 " All n components have failed "
__ __ __
 E1  E2 ...  En
Therefore:
__ __ __ __ __ __ __
P( E p )  P( E1  E2 ...  En )  P ( E1 ) P ( E2 )...P ( En )
P( E p )  1  P( E p )
Reliability for parallel system
 Product law of unreliabilities
n n
R p  1   (1  Ri ) , or R p (t )  1   (1  Ri (t ))
i 1 i 1

where Ri is the reliability of component i

 For exponential distribution:


n
Ri (t )  e i t , then R p (t )  1   (1  e it )
i 1
Series-Parallel system
voice
control
voice
control
voice
2 Control and 3 Voice Channels Example

•System is up as long as 1 control and 1 voice channel


are up
•The whole system can be treated as a series system
with two blocks, each block being a parallel system
Series-Parallel system (Continued)

 Each control channel has a reliability Rc(t)


 Each voice channel has a reliability Rv(t)
 System is up if at least one control channel and
at least 1 voice channel are up.
 Reliability:
R(t )  [1  (1  Rc (t )) ][1  (1  Rv (t )) ]
2 3
Homework :

 Specialize formula (3) to the case where:


 c t  v t
Rc (t )  e and Rv (t )  e

 Derive expressions for system reliability and


system mean time to failure.
An Fault Tree Example (cont.)
 Reliability of the system:
 c t  v t
Assume Rc (t )  e and Rv (t )  e ,

R(t )  [1  (1  Rc (t )) ][1  (1  Rv (t )) ]
2 3

 c t  2 c t  v t  2 v t 3v t
 (2e e )(3e  3e e )
The WFS Example

File Server

Computer Network

Workstation 1 Workstation 2
RBD for the WFS Example

Workstation 1

File Server

Workstation 2
RBD for the WFS Example
(cont.)
 Rw(t): workstation reliability
 Rf (t): file-server reliability
 System reliability R(t) is given by:

R t   1 
 1 RW  t    R  t 
2
f
 Note: applies to any time-to-failure distributions
RBD for the WFS Example
(cont.)
 Assuming exponentially distributed times to failure:
 failure rate of workstation W
 failure rate of file-server f
 t
R( t )  [1  (1  e w t 2
) ]e f

 The system mean time to failure (MTTF) is


given by:
2 1
MTTF  
0 R( t )dt  
 w   f 2 w   f
Homework :
 For the following system, write down the expression for system reliability and MTTF

C
D
A B C E
D
 Assuming for each block a failure rate i andC
 Calculate reliability for specified time period of 2500 hrs
and MTTF

Assume i  2.5 10 4 Failure/hr


Reliability evaluation of K-of-N System

 System consisting of N independent components


 System is up when K or more components are
operational.
 Identical K-of-N system: each component has the
same failure and/or repair distribution
 Non-identical K-of-N system: each component may
have different failure and/or repair distributions
Binomial Random Variable
(cont.)

 Y(t) is binomial with parameters n,p


pk  P (Y ( t )  k )  C(n, k) p (1 - p)
k n- k

 x
F ( x )  P (Y ( t )  x )   C(n, k) p
k
(1 - p) n- k

k 0

E[Y ( t )]  np
Reliability evaluation of K-of-N System

For Binomial distribution,

P ( X  x)  C (n, x)  R x  (1  R) n  x

Rs  P( X  K )  P( X  K  1)        P( X  N )
N
Rs   C ( N , i )  R i  (1  R) N i
iK
Reliability for identical K-of-N
Reliability of identical k out of n system

n
Rkofn (t )  FYnk 1 (t )   ( nj )[ R (t )] j [1  R (t )]n  j ,
j k

R (t )  1  F (t ) is the reliability for each component



k=n, series system
Rs (t )  [ R (t )]n

k=1, parallel system
R p (t )  1  [1  R (t )]n
Reliability Evaluation of Standby System
Standby system:

Operating unit
A
Switching device

Output
Input

B
Standby unit
As = Reliability of the switching device

As = 1.0, for perfect switching device

As < 1.0, for imperfect switching device


Classification of Standby System
Standby system

Standby system with Standby system with


perfect switching device imperfect switching device

( As  1.0) (0  As  1.0)

1-Operating unit (OU) n-Operating unit 1-Operating unit n-Operating unit


and and and and
N-standby unit (SU) N-standby unit N-standby unit N-standby unit
(CASE-1) (CASE-2) (CASE-3) (case-4)

General notation of standby system: (n/N) standby system


n = Number of operating units (OU)
N = Number of standby units (SU)
Reliability Evaluation of Standby System
(Perfect switching device)
Case-1: (1/1) standby system (1-OU and 1-SU)
Operating unit
A
Switching device
( As  1)
Output
Input

B
Standby unit

(1/1) Standby system


Reliability Evaluation of Standby System

Let,

 = Failure rate of component A and component B


t = Specified time period
m  t = Parameter of Poisson distribution
em  m x
Px (t )  = Probability that exactly x number of components will
x! fail upto time period t

System reliability is given by,

Rs (t )  P ( x  0)  P ( x  1)  e  m  me  m  (1  t )e  t

2
MTTF   Rs (t )dt 
0

Case-1: (1/2) standby system (1-OU and 2-SU)
OU-1
Switching device
SU-1
Input Output

SU-2
System reliability is given by,
2 m
m m m e
Rs (t )  P( x  0)  P( x  1)  P( x  2)  e  me 
2!

 1  t 
  t    t
2

e
 2! 

3
MTTF   Rs (t )dt 
0

Case-1: (1/N) standby system (1-OU and N-SU)

OU-1
Switching device

Input SU-1

SU-2

Output


SU-N
Rs (t )  P( x  0)  P( x  1)      P( x  N )

1 N
MTTF   Rs (t )dt 
0

Case-2: (n/N) standby system n-OU and N-SU

OU-1
OU-2

Switching devices

Input OU-n Output

Here, m  n t SU-1
Rs (t )  P( x  0)  P( x  1)      P( x  N ) SU-2


1 N

MTTF   Rs (t )dt  SU-N
0
n
Reliability Evaluation of Standby System
(Imperfect switching device)
Case-1: (1/1) standby system (1-OU and 1-SU)
Operating unit
A
Switching device
( As  1)
Output
Input

B
Standby unit

Standby system
Reliability Evaluation of Standby System
(Imperfect switching device)
Case-3: 1-OU and 1-SU

Rs (t )  P ( x  0)  P ( x  1)  As  e  m  As me  m  (1  tAs )e  t


1  As
MTTF   Rs (t )dt 
0

Case-3: 1-OU and 2-SU
Case-3: (n/N) standby unit (n-OU and N-SU)

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy