********** ****

...Reliability
2. OBJECTIVES
The objectives of this module are to,

discuss the basic concepts relating to reliability;

introduce the measures of reliability

derive the reliability of series system, parallel system, parallel-series system and series-parallel system.

3. KEYWORDS
Reliability, probability, early failures, chance failures, wear-out failures, measures of reliability, failure density

MTBE, MTTE, Hazard rate, system reliability, series system, parallel system, series-paraliel system, parallel-series

ystem, stand-by redundancy.

4. RELIABILITY
Rcliability is defined as the probability of a device performing its intended purpose adequately for the period
of time intendd under the operating conditions encountered. The reliability is the probability
with which the devices
will not fail to perform a required operation for a certain length time. Such
of problem is known as the problem ot
Survival. Reliabilng rebers thrhe Censiotny ft mnsu
This definition brings into the focus of four important factors viz.,

The reliability of a device is expressed as a probability.

i) The device is required to give adequate performance.

emoyk
(m) The duration of adequate performance is specified.
(iv) The environmental or operating conditions are specified.
n practice, even the best design manufacturing and maintenance effots do not completely
eliminate the
occurrence of failure. During the life of a system we may experience three distinct types of failures - early failures,

random failures and wear out failures.

Farly failures:

Early failures are those which occur in the early life ofa sysem operations.
 ..*
****"'****** **
Reliability.. ***** **
' ***********

Random or Chance failures

m) actual operating of the system and occur at random, irregulary
Chance faiure are predominant during the
the system.
and unexpectedly. The phase during which only chance failures
occur is called the useful life of

Wear-out failure: C loe m Undr'u mo loy i ywd wnsah).

occur if the
wear-out failures are caused due to ageing any wearing out of components. These failures
time,
system maintain properly or not maintain at all, The frequency of such failures increases rapidiy with

Frequency of failures

Early
faiures Chance Wear-out
failures failures
fig.1 Tifne

Fig.-1: Life characteristics curve

Measures of reliability
The prediction of system reliability is based on a number of factors such as : life characteristics, operating

conditions and the failure distribution.

Ifarandom sample of items are taken from a population and are put to test (or use) under a set of fixed (or
EVen) environmental or operating conditions, some number of sample will fail successively in time. The uiata so
obtain will epresent the length ofeach item. The length of life can be measured depending on whetber the item is

Laalo, IV, Aeroplate etc.) or non-repairable (bulb, fuse, missile, rocket etc.). For repairabie items, the
 .. Reliability
******

failures whereas for non repairable items the life can be

lite can be measured by failure rate or mean time between

measured by mean time to failure,

time interval
the number of failure in a given
Failure rate The failure rate is defined as
:
A

Number of failure
Total unit of operating hour T

Failuredensity:
interval of time to the total number of items at the
This is the ratio of the number of failures during a given unit
very beginning of the test.

Exampie i
Let the total number of items at the beginning of the test i.e., the total initial population was !000 and during
components that failures 130 (say). The failure density during the first unit
the frst unit iaterval, the nunberof

interval,
a, = 130== 0.13.
0.13.
1000
Siniarly, the failure density during the second unit interval can also be calculated.

Generalization
Letn, be the number of components that fail during the first unit interval and n, be the number that fail during
the second unit intervai and so on.

etNbe the total population. 1 hen the lailure densityduringthe firstunitinterval is ja, =
N
failure density

during the second unit interval is J and so on.

is no survivor, then
Let/ be the last interval after which there
 *** '*'''********. '**********
**********
Reliability.. * ***"

4+..
NN N
n+n t+==1
N N

Time interval
Fig-2

Remarks
The sum of the areas of the vertical bars in the bar chart for failure density is ,
) The total number of failed itemstill the end of the i-th time interval say, i-th
hour
= n +n [Link]

The number of survivor at the end ofi-th time of unit interval= =

N-(n +n, t... +n,
(um) Hence the reliability for the i-th hour= probability
ofsurvivaltilli-th hour
= R() (say)
 .. Reliability

number of survivers till the i-th hour

total population

N-(n, +1, t+n,)

N

=1 NN-..=1-, -fd,-f
N

=1-
=1
hour)
of the failure density bars from the beginning till the i-th
=1-(sum of the areas components that fail in i hours to the total
probability offailure for i-th hour is
the ratiò of the number of
)The

opulation i.e.
2t.+n,
N
= sum of areas ofthe
tihis probability of failure is denoted
by Q() then 2(i)= Jq, + ja, + ..+ja,
lf i-th hour
ailure density bars from the beginning till the

Mean time betwsepfaifure (MTBF)

During the operating period when the failure rate Jt
fairly constant, the mean time between failures is the
a is the failure rate.
reciprocal of the constant failure rate i.e., m =4where
the
required to as the average time of satisfactory operation of the system. In this case, larger
MTBF is also
MTBE, the higher of the reliability of the system.
MTBF are made when the item is repairable.
Note: The failure rate and

Megntfmeof failures
then mean time to failures (MTTF)
we have a life test information on n items with failuretimest,,,
If
s defined as

MITTF-
MTTF =2
 **'************ '*********
**********
Reliability.
****

**********************

Note: MTTF is used when the items are non repairable.

Dorivation of reliability function:

components put under test repeatedly at the
beginning or tne test.
LetNafixed number of
components till surviving after the time
or components failed during timet. N= number of
Yner [Link] in the same rate. Thus the reliability
eliability
As the time of the test increases,,N,will increase and N,
measure from such test is a function of time t.

N,
The reliability or the probability ofsurvival is defined as R{)= N.+N,

The probability of failure at any time t can be expressed as ()

N N
N, +N
At any time 1, R(1) + Q)=1.

Remarks

Singe R() and Q0 are mutually exclusive, then

NL
R() = 1-Q)=1-
N
dR IdN,
dt d N
N. dt

Again N, +N, 2N.

ie., N, = N, -N,

dN
Hence -N R{1)
dt dt

or, N, aR()
al
 Reliability

. dN adN rate of components

Again, as i.e.,
This gives the rate at which components fail at any time dt dt
components survival during time dt.
ailure during time dt is equal to the negative rate of

We have
aN-Nodt
dR()
dt
which implies

1 dN NN dR()
N, dt N, dt
eltl2
1
dR(0) sinceR()-
R( d N
1
dR()
.)=R() dt

1
dN
where () which is the failure rateofthe components.
N, dt
Now, integrating both sides-of the above with respect to t with in the limit 0 to t, we have

Rat=1AO,
dt I=0

dt= dR(t)
R(1)

-7d--[logR"Att=0,
R
R()= =1|
- log R()

Hence R() =-|2dt

ie, R(')=exp-jadt (1)

0

for computing reliability.

This gives a general formula
 ***

Reliability. ... ****

Particular case
(Say)
If2(1) is constant overthe time A{")=A
exp-Ar=exp|-t/m), where m= MTBF
Then equation (1) reduces to R{1) =

Properties of reliability function

) Since the reliability is a probability function, 0 R(t) S1.

) R(0)=1 and lim R(?) = 0

gi) R)isa decreasing function oftime./

Remarks
1
dN N,dR(T)
We have
N. d N, dt 2 (2)

1
dN dN
If is constant, then
must be constant. Again, will remain constant ifthe number of failed

components N,increase with time from the beginning of the test.

In that case,
dN N
dt
N
dN1
N,
N AF
dt No

When failed components are substituted immediately after lailure 1.e., N,

=N, for all the time. Now, to
calculate the failure rate we need to count the number of tailtures Nand the hours
ofoperations i.e., t. We have
from (2),

1N, dN dR().
i.e.
dR 1
dN
dt N dt **dt N dt

When
When N.
N ddt is plotted against time , we obtain the distribution function of failures at time
t on per
component basis i.e., failure frequency curve per component. Ihis curve 1s
known as failure density function for
distribution curve.
 ... Reliability
****

dN, dR(
ie
ie. St)=l
No dt d
The area under this curve is unity.

t)
ft) = ae*

Fig. 4

Thercfore, the failure rate can be defined as t)=-

1
dR R(O)
R() R() d

This means that the failure rate of any time t is equal to f(1)/R{1).
This is applicable for all possible distribution.

Particular case
When ( is constant over time t i.e., (1) =a (say)

Then ()-- dt
")=dt le-u

Again, O(1) = N,/N, which d0/dt =

dN=
Nodt
f()
S0) [by(3)

Hence. Q()= So)a

This means that the probability
of failure Q0) at time f is equivalent to the area under the density curve taken
or
om/=0 to t. Thus Q0) is the cumulative probability Tailure function which increases with timet.
Conversely, we have R(t)=1-2(1
 *******°***Y********
. *************
Reliability. *****'******* * ********************'***
*'********'

=1-
Jsa
- fs)dt-Js)a since f)dt =

(=0

This means that the probability of survivals decreases with time t and this equal to that area under the density
curve taken from t= t to o.

f an item has a constant failure rate, say, A and the probability density function is exponential then various
function can be shown as follows:

Q0)

ft) ae*
R(t)

Fig.5

Hazarad rate
Hazarad rate or instantaneous failute rate is defined as the limit of the failure rate as the interva!
length
approaches to zero i.e.,
 ************
Reliability
'************* '*

lim)-R(1+h)]
2()= hR()
2
h0

1mR+h)-
- lim|
R(1)
R()
h
( 2y LG
1
dR
2
R(0) d

sincer - dR
() dt
=
R()
Relationskíp between Z() and R(7).

Now, Z(t)=0 dQ) 1 since Q()+ R(1) =11

Rt) dt R(t)

which implies |Z()dt = R(0)

d
1

=
-log[1-20)J--logR(0),=-log R{') [Since R(0)=1

R(i) =exp-z)a

Q)-1-R)=1-exp-z()a
(i
s)- -z()exp-[z()na
dt
articular case

When the failure rate is constant then

2(t)=/9-^e"N
e
R()
rate, the hazard rate is also constant and is equal to the failure rate
This means that for constant failure ate.
 ******'*'************'*****'***********'''***'******''**''*****I* retien ,t
******t*

Reliability. *"*****

is given by
a continuous random variablef

tn) o.
C CApected value E() ofa MTBF of

E()= MTBF- {ft)dt t (

where ) is the failure density funetion.

We have f(t)=P)_-dR (1)

dt dt

Hence, MTBF = [- dt = -
StdR
dt =0

R(+JR()dt by parts]

= 0+ R(1)dt

=
MTBF=m |R(¢)dt

Particular case
For a constant failure rate we have R(1)=
e
MTBF=m = {R(1)dt
0
=
e dt=

xgnpfe 2
V A device with 1000 hours useful life at a constarnt
iailure rate 0.0001 per
hour in a given experiment.
the reliability per 10 hours operation of Whet is
this device. Find the value of MTBF?
Solution: We knowthatthe reliability per 10 hours is
R(t)=e
Here -0.0001 and = 10
 Reliability
**********************

Hence, R(()=e" = e =e=0.999

MTBF = -10,000 hours.
0.0001

xample 3

0.999. What is the

For an equipment, the reliability per 100 hours of operation has been estimated to be

lure rate of the equipment? Calculate MTBF?

Solution

Itis given that R(F 0.999

Here e is constant, so e = 0.999 as R(t)=eM
1-2f = 0.999, neglecting square and higher powers of Al.

1-1007. = 0.999
as
= 100

.i= 0.00001
1.. Failure rate 0.00001 per hours.

=
10 per nours.

Hence, MTBF = 1/.= 10 hours

.2 System Reliability

Generally to determine the reliability factor ofa system, the system is broken down to sub systems and

lements whose individual reliability factors can be estimated or determined. Depending on the manner in which

hese subsystems and elements are connected to constituted the given system. The combinatorial rules are appolied

pobtain the system reliability. Hence the basic steps are as follows:

) First identify the elements and subsystems which constitute the given system and whose individual

reliability factors can be estimated.

(i) Next drawa block diagram or a crcuit diagram to represent the logical manner or configuration in
which these units are connected to form the system.
Reliability.
11l1dlldliitiuur=

determined i.c., it may be deciteda

e codition for the successful operation ofthe system is then
to how the units should function.

addition, multiplication and their cornbirutine

(v)
maly the combinatorial rules of probability theoryi.e.

are applied to arival at the system reliability factor.

4.2.1 Reliability of series system

Xories system
any
components are connected in series which means that if
ln this system, a large number of of the system

one of the components fails, the system fails. In other words,

ifthe system is operated, cach component connected

comprising ofn-components in series is sepresented

as
in series should successfully be operated. The system

Fig.6
Let the successful operation of these individual units be represented by X,, Xz2,.,X, and their respective

probabilities p(X,), p(X,)...p(X,). For the sucessful operation of the system,itis necessary that alila

units function satisfactorily.

Hence ifthe units are not independent one another then the system reliability is

p(S)=p(X,X,X,X,)
= p{X)p(X,/X)p{X,/X,X,).pX,/xX,..X,.)
Ifthe successful operation ofeach unit is independent of the successful operation of the remaining units the

p{S)=p(X)p(X,).-p{X.).
IfR is the reliability ofthe i-th components in series in the system and R is the reliability ofthe system having

n components in series then R, = R,R,...R, = ||Ri

-
fRR=R, .. =R, =R (say) then

R = R" = (1-0) wihere Qis the probability offailureofeach unit. When ech component has an exponential
 *'******
******* ** **********i*stet
Reliability
********* ************

ime to faihure density, then R, = RR,...R, =

e"e..e
R, =R,RR, = eMe.. where A,
2
The mean time between failures for the system having n-components in series is given by

m, JR,(0)

-yt-, dt

+ +. + ,
if , =h, =... =, = (say) then R, (¢) = eand m,
lfn=1, then R, =eand m =1/.

Example
An electronic circuit consists of 5 silicon transistor, 3 silicon diodes, 10 composite register and 2 ceramic

apacitor in series configuration. The hourly failure rate of cach component is

fortransistor:
, = 4x10
fordiode: = 3x1007

forresistor : , = 2x10

forcapacity: = 2x1C
exponential distribution.
Calculate the reliability of the circuit for 10 hours when the components follow

Solution:

Components [Link]. (N)

a, (per hour) . NA,
 Reliability.

Transistor 5
4x10 20x10
Diode 3x10 9x105
Resister 10 2x10 20x10
Capacitor 2x10 4x104
N, = 0.00269

Let , and R, be the failure rate and reliability of the system.

As each component follows exponential time to failure density, then

, 53, +3 +101, +2A, = 20x10% +9x10% +20x10 +4x10*

0.00269 per hour
R, =e=e- 0m69

The estimated reliability of the circuit for ten hours is R, (10)= eL=e u
= 0.9735
This means that the circuit is expected to operate on an average without
failure 9735 times and would fai
(10000-9735) i.e., 265 times out of 10,000 operations
of 10 hours each.
Inthis case, MTBF is m = 1/%=1/0.00269 = 371.75 hours.
This means that the circuit is expected
to operate without failure for 372 hours.

Examples

The system connected is series consist ofthree independence

parts A, B and C which have MTBF of 100,
400 and 800 hours respectively. Find MTBF ofthe system
and reliability ofthe systems for
30 hours. How much
MTBF of the parts A has to be increased to get and
improvement of MTBF of the
system by 30%.

Pig.7
 ... Reliability

Solution:

lere, MTBF of the part A=m, = 100 hours

MTBF of the part B = m, = 400 hours

MTBF of the part C= m, = 800 hours

failure rate of A = 1/m, = 1/100 failure hours

failure rate of B =1/m, = 1/400 failure hours
Afailure rate of C =1/m, = 1/800 failure hours
Hence failure rate of system A, =
a+A +Ag
= 1/100+ 1/400 + 1/800
= 11/800 failure / hour
Therefore, MTBF of the system =1/,
1/(11/800)
= 72.75 hours.
-33/80
And reliäbility of the system= R, (30)=e*30 e -11/800x30
Second part: MTBF of the system=72.75 hors =m, (say)

Let m, be the new MTBF of the system.

m, = m, +30% 800 30 8001040/11

. of m,* X

111 100 11
Again, failure rate of the system =, = l/m, =11/1040 hours.

Let x hours will be the MTBF of parts A to improve the MTBF of the system by 30%

Hence, 1/(100+x) + 1/400 + 1/800 = 11/1040

1.e.,X46.5

2.2 Reliabifity of parallel system

system are connected in parallel which means that the
In this system a large number of components of the
functioning of any one of the components. A system
accessful operation of the system depends on the satisfactory
parallel in represented as
mprising ofn -components connected in
283
irectorate of Distance Education
 Reliability. ****"*********'********* *****N

Fig.8

/The reliabilityof the system can be calculated considering the conditions for system failure. Let X,A2s..,X,
present the successful operation ofunits 1,2,.., respectively. Similarly, iet X,X2,.,, represent thir

unsuccessfiul operation vespectively. IfpX;)isthe probability ofsuccessful operation of unit then-pP7sthe

[Link] then p(X,) is the probability of [Link].

Hence, p(X,) =1-p(X,)

As the system will fail ifall n units ofthe system fail simultaneously. The probability offilure of he systemis

given-by

P(3)=p{KR,K,..)
p(X,)p(,/K,)p(K,/K,.)---p{X,/R,,-X,.)
As the failure of the units afe independent of oue another then piS) = p(X,)P{X,)>>(,).
Hence, the probability of success of the system is given by

p{S)=1-p(3)
=1-ER)P(F.)*P(F.)
-1-P(X)}{1-P(X,)) -P(x.)}]
Hence the reliability of the system comprising ofn components connected in parallel
is given by
 .. Reliability

R,0)=1-[1e)

is the failure of the i-th component

1

where ) = -||}1-R,()}

In n units are identical and the unit failures are independent of one another then

R,(0)=1-II1-R,()}

- 1-{1-R()

Ifthe failure rate of each unit is exponential time to failure distribution, then

R)=1-II-R0)}=1-I1-"}
where , i=l

is the failure rate of the i-th component.

For identical units R,() = 1-(1-

Mean Time between Failure
The mean time between failure ofa system having two components connected in parallel can be obtained by
ntegrating the reliability function over the range oftfrom 0 to o,

ie, m, =R,()dt = ||1-1-R,()}{1-R,(}|dt

0

e+e' -e*"dt
0

= 1/%+1/2-1/( +)
Ifthese two components are identical i.e. A = z =a (say) then m, =
1/ +1/-1/21 = 1.5/. This
neans that the mean time between failure of a
parallel system consisting of two components
of equal failure rate is
15 times the MTBF ofa single component.

The system reliability can be improved either by improving the desin or by

providing redundancy of the
wVstem.
The system with redundancy has a number of stand-by units which take over
iftheother components of the
 Reliability. ******''**''*********:
. *** *****'********

comnu
system in fail. This a rescrve stock of stand-by's improves system rcliability but at a higher cost. Thecompromise

between increase cost due to stand-by arrangement and increase given reliability will help to decide theoptimal
optim
redundancy of the system.

Exampe 6
The failure rate of an electronic subsystem is 0.0005 failure/hour. Ifa operational period of 500 hours with
th
probability of success 0.95 is desired. What label of parallel redundancy is needed?

Solution: Given that à =0.0005 failures/hour, R, =reliability of the system = 0.95.

Let there bensubsystems connected in parallel.

As the subsystems are connected in parallel, R, =1-(1-R)" whereR is the reliability ofthe system.
Again, R =e e
= -0005xS00
e -0.25

Therefore, 0.95 =1-(1-e

which implies n= log(0.05)/log(0.2214)=2
Aence the label of parallel redundancy is 2.
Exámple 7
How many identical components each of which is 90%% reliable overa period
of 50 hours be used to obtain
a 99.99% parallel redundancy system over 50 hours.
If we want to obtain the same system reliability over a period-
of 100 hours, how many components should be added?

Solution: Giventhat R reliability ofeach component=90%=0.9,R,

= reliability ofthe paralle! system=
99.99%=0.999 and r= 50 hours.
Let there be n label of parallel redundancy.

=
R =1-(1-R 0.9999 1-(1-0.9(0.1)' =0001n=4
Hence 4 identical components are required to be connected in parallel.

Second part:

As the reliability of the system per 100 hours is given, we have

to evauate the reliability ofeach components
per 100 hours.

As the reliability of each component per 50 hours is 0.9,

 ***
. Reliability

0.9 s or 0.9-1-501 [neglectinghigherpowers of] .A=0.002.

Let R be the reliability ofeach component per 100
hours

R=eM =e"02NI0=e=0.81.
R, =1-1-R'" where n is the label ofredundancy
=1-(1-0.81)
0.9999 =1-(0.19)>n-6
Hence for the 2nd case, 6-4 i.c., 2 identical components are to be added.

xamplo8

Show that MTBF of the system of n identical units is parallel connected is 2; where à is the failure rate

f each conponent.

Sotion: Lct the parallel redundant system is having n units with failure rate ,Ag , A, (say). Hene the
liability of the system is given by

R, =1-(1-R)(1-R,).(1-R,).
1-1 -e).(1-e)
=1-T](-)
Now, MTBF of the system - R)d-[-1-"}1-e").-(1-e)]d
- eM+e +...+e)-{e +e4*,k}d
+.
(1/2+1/2, *..+1/%,)-{1/(%+h)+1/(4, +2,)++{/(A,+2,,)+1/( +A, +A,)

.. (-1)"
S
Foridentical units, A, A, =..=,
 ''***'***"* **i
***'
*****
****'****'*
Reliability..

case, MTBF = 1/A -1/2 +1/na =

=y-
in that

2.3 Mixed Configuration

Parallel series system

Ra
B
Ra

ADBC
Fig.9

As the unitB and C are connected in parallel, these units may be replaced by a unit
BC (say) with reliabili-

Rac Where

Rge=1-(1-R,)(1-R,)
and the changed configuration of the system will be as

A RC

Fig. 10

Now the units A and BC are connected in series. The reliability of the system can be calculated as=

R, = R,Rge=R,1-(1-R,)(1-Re)}

Another nixed configuration

RIL Ra
Ra
Rya

Ru, Ra

Fig. 11
 .. Reliability

The above system can be represented

by an equivalent system in the following
figure.

Ret Rs Rs
Fig. 12
ln this case

g1-(1-R,(0-R,)01-R,)
1-(1.)01-A,)(1-R,)
8g1-(-R, )3(1-,)
Hence the reliabsility ofthe system will be R, - RgRgR
Series paralel system

Ru Ri2 Ri R4

R K Ra3 R4

R Ry2 Ra3 H
[Link]
R34

Fig. 13

an equivalent in Fig. 14.

be representedby
The above system can
RsL

KSZ
sspkuanonav

Rss

Fig. 14
 Reliability..
Rss R31 R32 K33 34
1Cre R, Rz R3.R4. Rs2 = Rai R22 R2 Rz and
Rs
Rs2)(1- Rs3)
So, Rreliability of the 'system- = 1-(1- Rs(1-
Kample 9
consists of 3 identiç:
parailel. Each subsysteil
A System consisting of4 identical subsystems connected in is 0.95 obtain th
for unit the probability of each over a certain period oftime
uits connected in series. If each
vstem of reliability.
Fig. 16.
The given system can be represented by an equivalent system in

R R R

R R

RSB
R R

R
RsB

Fig.15

RsB

RsB
reR 0.95 (given) Fig. 16
nce the reliability of each sub system is Rs = R° = (0.95) = 0.8557
ncethesystemreliability (R5) =1 -(1-Rsu)=1-(1-0.857)' = 0.999582
.2.4 Stand-byrBdandaney
. Reliability
In paraliei configuration,
all the components
uring the operation operate
the system. In the stand-by-system, simultaneously andtheexperiences wear and
ot
one or more components there is a primary active
are standing-by element which is operatmg anu
to take over the
operation ot stand-by opcration one after another when
components is sequential the first one fails. lhe
i.e., each ofthe
onoya fter the duplicate clements becomes active and is
failure of he previous energzed
active elem en [Link] system
Following two condition w
llsurvive un til the requ ired tm el i enuner uc
holds.

Active

Stand-by system
Fig. 17

operating upto timet.

Primary elcment is successful
takes over and survives fromt, toi.
the active stand-by element
Primary element fails at l, and

Assuming the failure density function of the primal active and the
stand-by unit as f () and f.(t)

spectively we have

(a) probability ofprimary

element surviving upto the timetis 1-S0)dt = dde
element failing at time t, is /,(G)
(b) probabilityofprimary

working successfully from, to =l- JS0)dt = L0da

(c)
probabilityofstand-by element
 Reliability..
. ***'*i
****

Hence the reliabilityofthe system is given by R,()= [/,(oxirfs,(,) J)djat,

L-

ft)

R)

0
Fiig- 18

For theexponential case, , (t)=A, e*",St)=»,e

where and are the failure rates of the primarily active and stand-by elements respectively.

Hence the reliability of'the system R,(0)= Ja,e"*dt+ ,e a,edt d

=e + jeMe d, =
=e+iee dt,

For , = Ad =7,R(1) =e" +1 le"d

=e+he" jdt =eH +Ate

Now the failure density function is exponential. Number of failures follows Poisson distribution i.e., prob

(Numberof failures = i) = p{i) =€" (24)/i

From this, p(0) = e",p()=te" and so or

So, R, ()= e + te = p(0)+ p(1)

 *******
Reliability

Similarly, it can be show that

=
R,()= P(0)+ p(1)+ p(3)=e+ue4M () Ý-e*(a)
2 As
&,0)-
As R,

Hence, for a stand-by redundant system with n components of equal failure rate, the reliability of the system

sgivenby R(t)
=- i=0

MTBF
Let us consider the MTBF ofa stand-by redundant system with one active and one stand-by. Then

m, = R,(1) dt = j(1+21)e"dt= fe*dt+ jaue= =2

Hence the mean time betwecn the failures in the case of a stand-by redundant system consisting of two
componets is twice that of the single component.

Stand-by sytem with imperfect sensing over device

Let sensing over device is not perfectly reliable and hence its
probability of the failure must be considered.

Let the primary and secondary components be not similar and higher
different failure rate and the system consist of

rates of both components follow the exponential

two dissimilar components placed in stand-by redundancy. The failure

failure time distribution. The prinary components have a

failure rae A (say) and the stand-by component
,
(say).
switching over device.
Letr be the reliability of the sensing and

The
reliability of such a system is given by R(1) =e% +r. -a, -]

When A, =A =2, then R{(?) =e +(ue") =e"|1+(A)]

is operating
stand-by system ofn components which have equal failure rate and when one unit
In general for a
MTBF ofasystem is
rest (-1) units are standing-by to take over the operation in succession then
activity and the

n, = n/..
 Reliubiliny.. **'**' ****
'**** ****

Adple 10

dustrial process in controlled by a computer and twO similar components are operated in stand-by
ney a
Such that if computer fails another is instantaneously brought into use in its plae. The tailure rate of

uter isgiven by = 0.01 failure/hour. Compare the improvement in reliability over a single compiter
Cnd then two computers are used in a stand-by. The operating period is 100 hours and the switch is
cousidered to be perfect.
i) n the first case, only one computer is in the system.
Hence the reliability ofthe system having single
computer is given by

R(t)=e = e009 =1/e =

0.37

Fig. 19-
For the system having one computer along with another one as stand-by, the reliability of the system is given
by

A,(0)-
i-0

=eM+(t)e" =0.37 +0.01x100x0.37 0.37+0.37 = 0.74

As R, (/)= 2R, (), hence,thereis 100%improvementinreliabilitybyusing one stand-by computer.

(i) Again, for system having one computer along with two computer as stand-by, the reliability of the system is

gvcnby

R,)-()., +M(")+ e =0.74+ .37x1 = 0.74 +0.185 = 0.925

2 2

Now, R,(1)/R,(1)=0.925/0.37 =2.5

stand-by computers, reliability is improved by 2.5 times ie, there is a 150% improvement
Hence by using two
computer.
i
1Ciability over the single
**
*********************
.. Reliability

Example'11

An clecironic device has a failure rate of S00 1ailkures per 10 hours. One identical stand-by unit is added to
increase the reliability of the basic device. The operating time is 1000 hours. The failure rate of the sensing and
SWiiching clement is 0.97, What is the system reliability? What will be the system reliability if the sensing and
switching element is 100% reliable?

Solution:
Here, the failure rate A=500 failures/10hours
500/105 failure per hour

0.0005 failure per hours,

and the opera ting tim eí= 10O00 hours. Fig. 20
Hence t =0.0005 x 1000 = 0.5
Given that the failure rate of the sensing and switching element is 0.97.
Hence the reliability of sensing and switching element is rs1-0.97=0.03.
perfect, the system reliability is given by
So, when the sensing and switching element is not
(1+0.015) =0.6156
R() = e1+()=e
is perfect, the system reliability is iven by
When the sensing and switching elcnent

R()= e"(1+1) = e(1+0.05) = 0.909

/ exercise
Self Assessment questions series and parallel system.
system. Compare the reliability ofa
5.1 Define the reliabilityofa systemreliability
function for random failures is an exponential distribution. How
5.2 Show that the reliability

can be improved?
fwo components of which one is a
Reliability. Find the reliability of a system with
5.3 Define system
are connected in parallel.
stand-by. The components
tailurs/hour. Ifan operational
failure rate of an electronic sub-system is 0.0005
What is MTBF? The
5.4
probability of success p =0.95 is desired, what level of parallelredundancy
with
Deriod of 500 hours

is needed?
Reliability...***********************
*****'************'*'****"****"*****

5.5 Showthat R[) = exp-Ja()dtwhere R() is the reliability function and () representsthe

failure rate.

5.6 system, there are n number of components connected in parallel with reliability
In a

R().i = 1,2,,. Find thereliability ofthesystem. If R, (1) = R, =...R, =e" then () (t)
what will be the expression of system reliability?

How many identical components each ofwhich is 90% reliable over a period of 50 hrs. be used to
obtain a 99.99% parallel redundancy system over 100 hrs.

5,8 Showthatthe reliability ofan item can beexpressedas an exponential functioni.e., exp-(1)dt

An industrial process is controlled by a computer and two similar components are operated in stand-
by redundancy, such that, ifa computer fails, another is instantaneousily brought into use in its place

The failure rate of each computer is given by A=0.01 failure/hr. Compare the improvement in
reliability over a single computer when one and then two computers are stand-by. The operating
period is 100 hrs. and the switch system is 100% perfect.

0 Asystem consists of 5 identical and independent units with one unit operating and 4 units stand-by.
One of the stand-by units takes over when any operating unit fails. Assuming that the switching
devie
is perfect. Obtain the system reliability for a period of 100 hours each of
if the 5 units has a failure rae
of 100 failures/10 hours.

5.11 Obtain reliability over 100 hours period of a system consisting of two subsystermAand B connectert
in parallel where A consist of 4 identical components in series and B consist identical components
of
in parallel. Each component has a reliability 0.90 over a period
of 100 hours.
5.12 The system connected in series of 500 transistors, 10500 resistors and 500 capacitors. Failure rate
of these components are as follows

transistor n, = 0.7x10 perhour

 *****

'********''****
*******
Reliability

diode: =0.3x10° per hour

resistor: A, = 0.1x10 perhour

capacity: 2x0.2x10 per hour

What is the failure rate of the system? What is the reliability system of 100 hours.

6. Suggested Further Readings

Aggarwal, K.K., Reliability Engineering, 1993.

andApplications, 1993
Ramakumar Ramachandra, Enginering Relialiability Fundamentals
India.
Mustafi, C.K., Statistical Methods in Managerial Decisions, Macmillan,
Wiley Eastern, India.
Sinba, S.K. and Kaie B.K., Life Testing and Reliability Estimation,
Sultan Chand & Sons.
Swarup, K., Gupta, P.K. and Man Mohan, Operations Research,