Reliability Engineering

EE32XXX
Unit I
What is Reliability Engineering?

Probability of device
performing its purpose
adequately for a intended
period of time under specified
operating conditions
Failures

Failures are way of life in the modern technological Era.

•Power Distribution System

•Internet Systems

•Power plants

And many more

Things Fail

Washing Machine

Car battery

Water heater

Fan

House roof leak

Automobile engine starter

Failure case studies:

1. Challenger Space Shuttle

2. Columbia Space shuttle

3. Bhopal Gas tragedy

Many more.....
Causes for Failure of systems

1. Design , Production and use

2. System Simplification

3. Maintenance

4. Communication and Coordination

5. Human Reliability
Failures and Failures Modes
Reliability/Quality and life cycle
costs
Practical Reliability and Life cycle
Costs
Probability
Probability

Nothing in life is certain

We gauge the chances of successful outcomes in business,

medicine, weather, and other everyday situations such as the
lottery.
 History

For most of human history, probability, the formal study of the laws of
chance, has been used for only one thing: gambling
Basic Definitions
Random Experiments

in some experiments, we are not able to ascertain or control

the value of certain variables so that the results will vary from
one performance of the experiment to the next, even though
most of the conditions are the same. These experiments are
described as random.

Example 1 :

If we toss a die, the result of the experiment is that it

will come up with one of the numbers in the set {1, 2, 3, 4, 5,
6}.
Sample Spaces

A set S that consists of all possible outcomes of a random

experiment is called a sample space, and each outcome is
called a sample point

Example 2 :
If we toss a die, then one sample space is given by
{1, 2, 3, 4, 5, 6} while another is {even, odd}.
Events

An event is a subset A of the sample space S, i.e., it is a set of

possible outcomes. If the outcome of an experiment is an
element of A, we say that the event A has occurred that is
desired outcome.

Simple Event : Only one outcome

Null Event : Empty Set (Impossible Event)

Certain Event : Sure Event

Laws of Probability

The probability for any event must be between 0 and 1.

Laws of Probability

3. P(A’ ) = 1 - P(A)
For an event A, A’ is the complement of A; A’ is everything in
S that is not in A.
Mutually Exclusive Events

Mutually exclusive events-no outcomes from S in common

A∩Β=∅
S

A
B
Frequency Interpretation

An experiment is performed N times independently and let us

assume no of times favourable outcome occurs is ni for event
Ai:

P(Ai) = ni/N

True value of P(Ai) wil be obtained if N is very large(Infinity).

Example 3

Four coins are tossed simultaneously 100 times and frequency

outcomes are:

Outcome 4 Heads 3 Heads 2 Heads 1 Head 0 Head

Frequency 9 30 25 28 8

What is the probability of getting two heads?

Solution

Let P(A) be the probability of getting 2 Heads

P(A) = 25/100 = 0.25

 Probability Addition Theorem:

(a) If events are mutually exclusive

(b) If events are not mutually exclusive

Probability Addition Theorem For
Mutually Exclusive Events:
If events A1 , A2 , ....An are mutually excusive events then :

A(A1+A2....+An) = P(A1) + P(A2)....+P(An)

Example : When two coins are tossed what is the probability of

getting 2 Heads or Tails?

Solution: Sample Space S = [ HH , TT , TH , HT]

P(A1) is probability of getting 2 Heads = 1/4

P(A2) is probability of getting 2 Tails = 1/4

P(A1+A2) = P(A1) + P(A2) = ¼ + ¼ = ½ = 0.5

Probability Addition Theorem For
not Mutually Exclusive Events:
If events A1 , A2 , ....An are not mutually excusive events then :

P(A∪B) = P(A) + P(B) – P(A ∩ B)

P(A∪B ∪C) = P(A) + P(B) + P(C) – P(A ∩ B) - P(B ∩ C) - P(C ∩ A) +

P(A ∩ B ∩ C)

Example 4 : What is the probability of getting 2 or 5 when a die is

rolled?

Solution : P(2) = 1/6 ; P(5) = 1/6 ; P ( 2 ∩ 5) = 0

P(2 ∪ 5) = 1/3
 Probability Multiplication
Theorem:
(a) Independent Events

(b) Dependent Events

Probability Multiplication
Theorem for Independent Events:
Independent Events : Outcome of one event does not affect
the outcome of other events.

P(A1.A2.A3.......An) = P(A1).P(A2).P(A3)......P(An)

Example 5 : If a coin is tossed twice. What is the probability of

getting two consecutive tails?

P(1 tail in first turn) = ½

P(2 tails in two turns) = ½ * ½ = ¼
Example 6 : A pack contains four blue , 2 red and 3 black pens.
If a pen is drawn at random from the pack, replaced and
process repeated two more times , what is the probability of
drawing 2 blue pens and 1 black pen?
Solution : Total number of pens = 9

P(1 blue pen) = 4/9

P(second blue pen) = 4/9 (Replaced)

P(1 black pen) = 3/9

P(2 blue and 1 black pen) = 4/9 * 4/9 * 3/9 = 0.197

Probability Multiplication
Theorem for Dependent Events:
P(A1.A2.A3.....An) = P(An/A1A2...A(n-1)) * P(A(n-1)/A1A2....A(n-2) )
.........P(A2/A1).P(A1)

P(A2/A1) is the probability of occurrence of A2 assuming A1 has

occured

Example 7 : : A pack contains four blue , 2 red and 3 black pens. If a

pen is drawn at random from the pack, not replaced and process
repeated two more times , what is the probability of drawing 2 blue
pens and 1 black pen?
Solution:

P( 1 blue pen ) = 4/9

P(second blue pen ) = 3/8

P(1 black pen) = 3/7

P(2 blue pen and 1 black pen) = 4/9 * 3/8 * 3/7 = 0.07
Total Probability Theorem :

P(A and B) = P(A) * P(B/A)

Example 8 : A factory shop transformer feeds four motors with

input powers as 10 KVA,20 KVA ,20 KVA ,50KVA.The operation
of these motors is associated with the production process in
such a manner that when the first motor is running , the
probability of switching on of second motor is 0.7,when the
first and second motors are running,the probability of
switching on of third motor is 0.6 , when the first three motors
are running , the probability of switching on of fourth motor is
0.3.Assume the probability of switching on of first motor is
0.8.Determine the load of 100 KVA on factory shop of
transformer.
Solution :

As we know , this is a problem of Dependent events as for

100KVA load all motors should run simultaneously

P(A1. A2.A3.A4) = P(A4/A1A2A3) * P(A3/A1A2) * P(A2/A1) *

P(A1)

P(A1) = 0.8 ; P(A2/A1) = 0.7 ; P(A3/A1A2) = 0.6 and

P(A4/A1A2A3) = 0.3

Probability of 100 KVA load = 0.3*0.6*0.7*0.8 = 0.1008

 Example 9 : Consider transmission system with three lines and
three nodes , generator connected at node 1 only. Capacity of
each line is 100 MW.Loads at node 2 and 3 are :

(a) 75 and 125 MW respectively

(b) 50 MW each.

Obtain reliability of transmission system assuming reliability

of each line as 0.96.
Solution :
Case (a) :
All line should be up to supply the load

P(L1 And L2 and L3) = P(L1)*P(L2)*P(L3)

= 0.96*0.96*0.96 = 0.8847
Case(b) : To supply load of 50 MW at bus 2 and 3, any two lines
should be up atleast.

Success events :
A1 = L1L2L3
A2 =
A3 =
A4 =

P(A1) = 0.96*0.96*0.96 =0.8847

P(A2) = 0.96*0.96*0.04 = 0.03686
P(A3) = 0.96*0.04*0.96 = 0.03686
P(A4) = 0.96*0.96*0.04 = 0.03686

P(Success) = P(A1) + P(A2) + P(A3) +P(A4) = 0.9952

Random Variable :

A random variable X is a numerical

outcome of a random experiment
The distribution of a random variable is the
collection of possible outcomes along with
their probabilities:
Discrete case:

Continuous case:
Example : If a coin is tossed twice (Discrete)
Rule x : Number of tails is the outcome

Sample Space = [ HH , TT , HT, TH]

Rx = [ 0 , 1 , 2]
Probability Density Function of
Discrete Random Variable
P [X = xi ] = f(xi)

Where
f(xi ) is PDF of Discrete Random Variable
xi = 1 , 2 ....

0 ≤ f(xi) ≤1
∑ f(xi) = 1 for all i

The discrete sequence satisfying above conditions is called

Probability Density Function (PDF).
The distribution of Probability of Random variable(RV) versus
Random variable(RV) is known as Probability Density Function.

f(xi) ¼ ½ ¼
xi 0 1 2

∑ f(xi) = 1
Cumulative Density Function(CDF)

CDF of a random variable ‘x’ may be defined as the probability

that the random variable ‘x’ takes a value “less than or equal
to x”.

CDF : F(x) = P( X ≤ x)= ∑f(xi) for all i from -∞ to n

Example : If a coin is tossed twice (Discrete)

Rule x : Number of tails is less than two.
Properties of CDF:

1. F(x) = P( X ≤ x) , so its value is always between 0 and 1.

0 ≤ F(x) ≤ 1

2. F(-∞) = 0 (min)
F(∞) = 1 (max)

3. F(x1) ≤ F(x2)
If x1 ≤ x2

So according to this property CDF is a non decreasing function

of x.
CDF of Discrete variable :

CDF of a discrete variable at any certain event is equal to the

summation of the probability of random variable upto the
certain event.

Example : CDF for number of tails less than 2 as outcome for

tossing of two coins:

CDF : P(0) + P(1) = ¼ + ½ = ¾

P(X ≤ 2) = ¾
Mean Value

The mean value or expected value of a random variable is

defined as follows:

µ = E[x] = ∑ xi * f(xi) for all i from -∞ to ∞

Variance

Variance of a random variable is defined as dispersion around

mean value

σ2 = E[(x - µ)2] = ∑ [(x - µ)2 f(xi) ] for all i from -∞ to ∞

Large value of variance signifies large variation around mean

value and signifies large uncertainty.

σ2 = E[(x - µ)2] = E[x2 ] + E[µ2] – E[2xµ] = E[x2 ] + µ2 - 2 µ2

E[x2 ]= µ2 + σ2
Binomial Distribution

The pdf of binomial Distribution:

E[X] = np, Var(X) = np(1-p) = npq

CDF:
Example : A biased coin is tossed thrice. Probability of getting
head is ¾ and that of tail is ¼ . Obtain probability of getting
two heads in three trials.

Solution : P = ¾ q=¼

f(r) = 3C2 P2 q1 = 27/64

Example : A power plant has 10 generators each having
probability of success as 0.94.Assume independent failures.
Determine the probability of operating at least four
generators.
Solution

N = 10
P = 0.94
Q = 0.06
10
10
P(X≥ 4) = ∑ Cr (0.94)r (0.06)10-r
r=4
Poisson Distribution

When we expect some independent events to occur µ times

over a specified time interval.The probability of exactly
‘x’instances is :

f(x ,µ) = e-µ µx/ x !

Example

The failure of power transformers is assumed to follow a

Poisson probability distribution. Suppose on average, a
transformer fails once every 5 years. What is the probability
that it will not fail in the next 12 months? That it will fail once
in the next 24 months?
Solution
Practice Question

Assume X has distribution as Poisson if P[x = 0] = 0.1. Calculate

P [ x≥ 3].
Distribution Function of
Continuous Random Variable
b
P(a ≤ X ≤ b) = ∫ f(x) dx
a
It will quantify as pdf if it satisfies:

f(x) ≥0

∞
∫ f(x) dx = 1
-∞

This means PDF will always have non negative values and total
area under the curve is unity.

Point Probability is zero.

Cumulative Distribution Function

x
F(x) = P[X≤x] = ∫ f(x) dx
-∞
It is a non decreasing function

F(- ∞) = 0 and F(∞) = 1

Also f(x) = dF(x) /dt

Example
Solution
Example
Solution
Mean Value

It is the expected value of the distribution.

Median

xmedian
P[x ≥ xmedian] = 0.5 = ∫ f(x) dx
-∞
Mode

Point where f(x) is maximum

Variance

It shows dispersion around the mean value. Larger variance

shows larger uncertainity in the random variable.
∞
Var[x] = σ2 = E[(x - µ)2] = ∫ [(x - µ)2 f(x) dx ]
-∞

E[x2 ]= µ2 + σ2

σ is Standard devition
Uniform Distribution
Mean Value = (a+b)/2

Variance[x] = (b-a)2 / 12
Exponential Distribution

PDF : f(x) = λ exp(- λx) for x ≥ 0

CDF = F(x) = 1 - exp(- λx)

Mean = 1/ λ

Variance = 1/ λ2

Median = -ln(0.5) / λ

Mode = 0
Normal Distribution
 Example

Load at substation is normal distributed with mean 1000 MW

and Standard deviation as 100 MW. Obtain:

(a) Probability of load equal to or more than 1100 MW

(b) Probability of load between 900 to 1100 MW

(c) Probability of load is lesser than or equal to 800 MW

 Solution
∞
(a) P[L≥1100] = ∫ N(1000 , 1002) dx
1100
We have to normalize it and

Z = (1100 - 1000 )/ 100 = 1

∞
P[L≥1100] = ∫ N(0 , 1) dz = ϕ (∞) - ϕ (1)
1

ϕ (∞) = 1
ϕ (1) = ϕ (0) + ϕ (1) = 0.5 + 0.34 = 0.84 (from z table)
P[L≥1100] = 1 - 0.84 = 0.16
 1100

(b) P[900 ≤L≤1100] = ∫ N(1000 , 1002) dx

900

1
= ∫ N(0 , 1) dz = ϕ (1) - ϕ (-1) = 0.683
-1
 800
(c) P[L≤800] = ∫ N(1000 , 1002) dx
∞
-2
= ∫ N(0 , 1) dz = ϕ (-2) - ϕ (-∞) = 0.025
-∞
Rayleigh Distribution
Weibull Distribution
Gamma Distribution
Function of one random variable
and its distribution function
If length of a square is random variable x and we want to find
area or perimeter using this random variable so we map this
RV to new RV y and we want to find the PDF of this new RV
y.We do this using CDF technique. Steps are:
Example

Let x be RV with PDF

f(x) = 2e-2x , x>0

Determine the PDF of y = √x.

Solution

F(y) = P(Y≤ y) = P(√x ≤ y) = P(x ≤ y2 )

For integrating limits : lower limit will be x = 0 and upper limit

will be x = y2

y2
F(y) = ∫ 2e-2x dx = 1 – exp(-2 y2)
0
This is CDF of RV y. To find PDF:

f(y) = dF(y)/dt = 4y(exp(-2y2))

For limits : x> 0 we see y = √x , hence y>0

f(y) = 4y(exp(-2y2)) , y>0

Note

We can check the solution by :

∞
∫ f(y) dy = 1
0
Example

Let x be a uniform RV in the interval (0,1). Find PDF of y = 1/x.

It is seen that it is continuous RV.

Solution
F(y) = P(Y≤ y) = P(1/x ≤ y) = P(x ≥ 1/y )

Lower limit will be 1/y and upper limit will be 1

F(y) = 1 – 1/y

PDF :

f(y) = dF(y)/dt = 1/ y2

For y limit: x is defined in (0,1) this gives

x = 0 we have y = ∞
x = 1 we have y = 1

f(y) = 1/ y2 , y>1
Example

A RV x has distribution

f(x) = 3x2 0<x<1

Find PDF of y = 3x + 1
Solution

F(y) = P(Y≤ y) = P(3x + 1 ≤ y) = P(x ≤ (y-1)/3 )

(y-1)/3
F(y) = ∫ 3x2 dx = (y-1)3 /3
0

PDF : f(y) = (y-1)2 /3 , 1<y<4

Example

Current in resistance of 1 ohm is uniformly distributed

between 8 to 10 A. Obtain pdf of Power P.
Solution

Current is uniformly distributed (8 , 10)

Pdf of i = f(i) = 1/(b-a) = 1/(10-8) = 1/2

f(i) = ½

Also P = i2 R
As R = 1 ohm
P = i2

we have x as i and P as y
Solution

F(y) = P(Y≤ y) = P(x2 ≤ y) = P(x ≤ √y )

√y
F(y) = ∫ 1/2 dx = -4 + √y/2
8

CDF of Power (P) = F(P) = -4 + √i/2

PDF of Power(P) = 1/4√P 64<P<100

Example

A RV is uniformly distributed [-1 , 1]. Obtain PDF of y = x2

Solution

f(x) = 1/b-a = 1/2

F(y) = P(Y ≤ y) = P (x2 ≤ y) = P(-√y ≤ x ≤ √y)

√y

F(y) = ∫ 1/2 dx = √y
-√y

f(y) = 1/2 √y , 0 ≤ y ≤ 1
References

1.Practical Reliability Engineering, Fifth Edition, Patrick

D.T.O’Connor and Andre Klyener. 2012 John Wiley and Sons ,
Ltd.

2. Reliability Engineering ,Volume 3 , K.K. Agrawal , 1993

Springer Science+Business Media Dordrecht.