Simeon D.

Poisson
(1781-1840)
Poisson distribution is a discrete probability
distribution the probability of a given number of events
occurring in a fixed interval of time and/or space if these
events occur with a known average rate
and independently of the time since the last event.
The Poisson distribution can also be used for the number
of events in other specified intervals such as distance,
area or volume.

The Poisson is nothing more than the limiting case of

the Binomial where n is large and p is small.
When there is a large number of trials, but a small
probability of success, binomial calculation becomes
impractical
For example-Consider the following conditions:
p is very small and approaches 0
Example: a 100 sided dice instead of a 6 sided dice, p
= 1/100 instead of 1/6
Example: a 1000 sided dice, p = 1/1000
N is very large and approaches
Example: throwing 100 or 1000 dice instead of 2 dice
The product Np is finite

Examples of Poisson distribution

The number of defects in a 50-yard roll of fabric.
The number of end breaks in spinning frame.
Number of dye-flecks in dyed fabric.
Radioactive decay.
The number of goals in sports involving two
competing teams.
The number of deaths per year in a given age group.
 Failure of a machine in one month.

Definition
A discrete random variable X is said to have a
Poisson distribution with parameter > 0, if,
for k = 0, 1, 2, , the probability mass
function of X is given by

Where
e is Euler's number (e = 2.71828...)
k is always an integer 0
k! is the factorial of k.

The Poisson distribution:

In many situations condition of success or failure do
not apply. One important case is when events occur
randomly in space or time and there is no theoretically
limit to number of events that can happen in trial or unit.
Provided that events (faults, end breaks, etc.) occur
independently and only at time. The appropriate
distribution is often the Poisson.
For this distribution, the variable x is the number of
events per unit of time or space and can take place the
values 0, 1, 2, 3, 4.
The probability that exactly r events will occur is given
by
Pr(X=r) = (e-m *mr)/r!
M is given constant and is known as Poisson distribution
which is equal to mean value.

Examples
1. In a certain factory, the machines tend to break down
at random, and the average number of breakdowns per
day is 2.1. The mechanic can deal with up to three
breakdowns during a normal working day. If more than
three machines breakdown, the mechanic has to work
overtime.
(a) What is the probability that the mechanic will
have to work overtime on a particular day?
(b) What is the most probable number of stoppages
in a day?
(c)If, On a certain day, six machines broke down, what
would you conclude, and why?

Solution:-
Here,
Average no. of breakdown per day m= 2.1
Mechanic can deals up to 3 break downs a day.
Therefore
Time to deal with 1 beak down = 1/3 day
The probability that exactly r events will occur is given
by
Pr(X=r) = (e-m *mr)/r! (i)

From equation (i)

Pr(X=0) = (e-2.1 *2.10)/0! =0.1225
Pr(X=1) = (e-2.1 *2.11)/1! =0.2572
Pr(X=2) = (e-2.1 *2.12)/2! =0.2701
Pr(X=3) = (e-2.1 *2.13)/3! =0.1890
Pr(X=4) = (e-2.1 *2.14)/4! =0.0992
Pr(X=5) = (e-2.1 *2.15)/5! =0.0417
Pr(X=6) = (e-2.1 *2.16)/6! =0.0146
Pr(X>0) = 1-Pr(X<=6) = 0.0058=0

Number of Probability Number of

Breakdown of this (days) of
in a day number of overtime
stoppages
0 0.1225 0
1 0.2572 0
2 0.2701 0
3 0.1890 0
4 0.0992 1/3
5 0.0417 2/3
6 0.0146 1
6 or more 0.0058 More than 1

(a) if more than 3 break down occur mechanic will

do overtime,
Probability that mechanic will do overtime = PR(X>3)
,

= 1-[PR(X=0) + PR(X=1) + PR(X=2) +

PR(X=3)]
= 1- 0.837
= 0.163
 (b) As shown in table, probability of two stoppages
is highest, therefore two number of stoppages will
occur frequently.

(c) Probability of sic no. of stoppages = PR(X=6)

= (e-2.1*2.16)/6!
=0.0146
=0
Since chances if its occurrence very negligible or
impossible it would rarely occur.
2. 10% garment produced in a certain manufacturing
process turns out to be defective. Find the probability
that in sample of 10 garments chosen at random exactly
2 will be defective by using the Poisson approximation to
Binomial distribution.
Solution:-
Here,
M=np
m=mean
N- no. of samples
p- Probability of defective Garment
Given p=0.1 &n=10
Therefore,
m= 10 x 0.1
m= 1
Now,
Pr(X=r) = (e-m *mr)/r!
For=2
Pr(X=2) = (e-1 *12)/2!
=0.1839
3. If the probability that ginning mill suffers a Fire is
0.001, determine the probability that out of 2000 ginning
mill (a) exactly 3 & (b) more than 2 ginning mill will
suffers a fire.
Solution:-
The probability that mill suffers from fire p = 0.001
& no. of ginning mills n =2000
Therefore,
m=n x p
= 0.001 x2000
=2
Now,
a) Probability that exactly 3 mill catch fire is Pr(X=3)
Pr(X=3) = (e-2 *23)/3!
=0.1804
b) More than 2 mill catch fire is given by

Pr(X >2) = 1 - Pr(X= <2)

= 1 [PR(X=0) + PR(X=1) + PR(X=2)]
= 1 (0.6766)
= 0.3234

4. If 3% of the electric bulbs manufactured by a co. are

defective, find the probability that in a sample of 100
bulbs
 a). 0 , b) 5 bulbs will be defective.
Solution:-
The probability that electric bulb defective p = 0.001
& no. of bulbs n =2000
Therefore,
m=n x p
= 100 x0.003
=3
a) Pr(X=0) = (e-3 *30)/0!
=0.0498
b) Pr(X=5) = (e-3 *35)/5!
=0.1008

5. According to survey it is found that, the average

number of accident per year in Spinning mill is 3 per
100000 spindle mill.
Find the probability that in a Spinning mill of 200000
spindle there will be
 a) 6 b) between 4 & 6 accident per year.

Solution:-
In a Spinning mill of 100000 spindle
Average no. of accident per year =3
Therefore
For a Spinning mill of 200000 spindle,
Average no. of accident per year =6
m=6
a) Pr(X=6) = (e-m *mr )/r!

= (e-6 *m6)/6!
= 0.1606
b) Pr(7>X>5) = pr(X=4) + pr(X=5) + pr(X=6)
= 0.4550
6. Between 2 and 4 am the average no. of phone call per
minute coming n to switch board of a company is 2.5.
Find the probability that during one particular minute
there will be a) 0 b) 3 or fewer phone calls.

Solution:-
Phone calls / minute m= 2.5
Now,
Using formula
Pr(X=r) = (e-m *mr)/r!
a) Pr (X=0)= (e-2.5 *2.50)/0!
= 0.0821
Therefore probability that during one particular
minute there will be no call is 0.0821.

b)
Pr(X=<3) = PR(X=0) + PR(X=1) + PR(X=2) +
PR(X=3)
= 0.7577
Therefore probability that during one particular
minute there will be 3 or fewer phone calls is 0.7577.
7. A box contain 1 40s count bobbin and 7, 30s count
bobbin, A bobbin is drawn from a box from a box and its
count is tested. Then the bobbin is put back into the box
and the content are thoroughly mixed. Using Poisson
distribution, find the probability that is 8 such drawing a
40s count bobbin is selected exactly 3 times.

Solution:-
A Box contain total 8 bobbin, out of which 1 is 40s
count & 7 are of 30s count

Therefore,
Probability that drawing a 40s count bobbin is 1
out of 8
P = 1/8
P= 0.125
Total no of drawing are *
n= 8
 Therefore,
m=n x p
= 8 x 1/8
=1
Probability that bobbin is selected exactly 3 times is given
by
a) Pr (X=3)= (e-1 *13)/3!
= 0.0613

8.The number of faults that is made by the Combing and

Winding sections of the Spinning Department were noted
for fifty days and the results are shown in the table
opposite. The number of faults by Combing Sections are
given first in each pair of numbers.
Investigate the distribution of the numbers of faults made
 (a) by the combing section;
(b) by the winding section;
(c) in total each day i.e. 0+2=2, 2+2=4, etc.

0,2 2,2 6,0 3,5 1,2

2,2 1,1 2,2 1,1 2,3

7,0 1,4 3,6 2,3 3,0

4,1 5,1 4,3 5,4 6,4

1,0 2,3 3,2 3,3 6,1 Number of faults (combing,

2,3 2,2 2,1 3,5 5,3 winding)
4,3 4,2 3,4 4,3 3,1

3,1 3,3 4,4 5,4 2,1

5,6 1,2 2,2 1,2 3,3

4,2 0,5 4,4 2,2 2,1

Now Consider the result of combing two independent

Poisson variables
A PR(2) and B Po(3)
Define the new distribution C = A+B .
What can you say about C?
You know that
P(A=0) = e-2 ,
P(A=1) = e-2 x 2 , P(A=2) = e-2 22 ,
2!

P (B=0) = e-3 , P (B=1) = e-3 x 3, P (B=2) = e-3 32 , ..

2!
This gives
P(C=0) = P(A=0) x P(B=0) = e-2 x e-3 = e-5 .
P(C=1) = P(A=0) x P(B=1) + P(A=1) x P(B=0)
= e-2 x 3e-3 + 2e-2 x e-3 = 5e-5,
and P(C=2) = P(A=0) x P(B=2) + P(A=1) x P(B=1) + P(A=2) x
P(B=0)
= e-2x 9e-3 + 2e-2 x 3e-3 + e-2 22 x e-3
2! 2!
= e-5 9 + 6 + 4
2! 2!

= 25e-5 = e-5 52
2! 2!
9. Number of breakages in Ring frame bobbin in a hour
has a poisson distribution mean of 1.2 Find the
probability that the numbers of errors
(a) On spindle number four is 2
(b) On spindle number three is less than 3
(c) On first 10 spindles is total 5
(d) On all 40 spindles adds up to at least 3
Sol:-
Let E be the number of breakage on spindle so that
E Po (1.2)
(a) P(E=2) = e-1.2 (1.2)2 0.217.
2!

(b) P(E<3) = P(E<2)

or
 P(E<3) = P(E=0) + P(E=1) + P(E=2)
= e-1.2 + 1.2 e-1.2 + (1.2)2 e-1.2
2!
= e-1.2 (1 + 1.2 + 0.72)
= 2.92 e-1.2
= 0.8795
(c) Let E10 be the number od breakages on first 10
spindles
then E10 P0 (12), as E10 = E + E + . + E .
and E10 P0 (1.2 + 1.2 + . + 1.2) = P10(12)
hence P(E10 = 5) = e-1.2 125 = 0.0127
5!
(d) Similarly E40 P0 (48).
P(E40 > 3) = 1 P(E40 < 2)
= 1 e- 48 + e- 48 x 48 + e- 48 x 482
2!
= 1 1201e- 48 1.000
 10.Over a long Period of time, the number of stoppage
per day on a Carding Machine has been
found to have approximately a Poisson distribution,
within a mean of 2 on weekdays and 3 at weekend.
(a) What is the probability that there are 3
stoppage on a day chosen at random?
(b) If there are no stoppage on a particular day,
calculate the probability that it is weekday.
Sol :- (a) The probability tree is given below:
(b)

12. The number of Machine failure per working week in

a particular textile factory is known to follow a poisson
distribution with a mean 0.5.
Find the probability that
(a) In a particular week there will be:
(i) less than 2 machine failure,
(ii) more than 2 machine failure;
(b) In a three week period there will be no machine
failure.
Let A be the number of machine failure in one week, so
A P0 (0.5).
(a) (i) P(A<2) = P(A<1)
= 0.9098
or, from the formula,
P(A<2) = P(A=0) + P(A=1)
= e-0.5 + e-0.5 x 0.5
1!
= 3 e-0.5
2
0.9098

(ii) P(A>2) = 1 - P(A<2)

or
= 1 [P(A=0) + P(A=1) + P(A=2)]
 = 1 - e-0.5 + e-0.5 0.5 + e-0.5 (0.5)2
2!

= 1 e0.5 (1 + 0.5 + 0.25)

= 1 1.625 e -0.5
0.0144
(b) P(0 in 3 weeks) = (e-0.5)3 0.223

13. A garment manufacture knows that 5%of his

product is defective. If he sells garment in batches of
100 and guarantees that not more than 10 garments
will be defective, what is the approximate probability
that a batch will fail to meet the guaranteed quality?
Solution:
Here, probability of a defective garment = 5% = 0.05
n = 10 = np = 10*0.05 = 0.5
X: Number of defective garments
P(X=x) = e- x /x!
Therefore, probability that a box will fail to meet the
guaranteed quality is
P(X>10) = 1-P (X10) = 1-e-0.5 0.5x/x! (where x =
10,9,8,7,6,5,4,3,2,1,0) = 0.195
14. A very large logging opperation has serious
problems keeping their skidders oprating properly. The
equitment fails at the rate of 3 breakdowns every 48
hours. Assume that x is the time between breakdowns
and is exponentially distributed. What is the probability
of two or less breakdowns in the next 48hour period?
Solution:

No. of breakdowns in next 48 hours= y

y follows Poisson dist with mean =lambda = 3

Putting =3 we get
P(y 2) = f(0) + f(1) + f(2)
Putting y=0,1,2 and adding we get
 y f(y)

0 0.049787068

1 0.149361205

2 0.224041808

P(y<=2) 0.423190081
Thus the probability of two or less breakdowns in the
next 48hour period is
P(y<=2) = 0.423

15. It is estimated that the average number of defects in

a 150m2 of fabric roll is 10. What is the probability of
finding no more than 2 defects and no defects in 50m2
of fabric?
Solution:
Average number of defects in 50m2 of fabric= 10/15050
i.e., = 3.3
Probability of finding no more than 2 defects,
P (x2) = P(x=0)+P(x=1)+P(x=2)
 = e-3.33.30/0! + e-3.33.31/1! + e-
3.3
3.32/2!
= .36
Probability of finding no defects,
P(x=0) = e-3.33.30/0!
= 0.037

16. A manufacturer, who produces ladies garments

finds that 0.1% of the
Garment is defective. The garments are packed in boxes
containing 500 ladies
Garments. Using Poisson distribution, find how many
boxes will contain at least
Two Defectives.
Solution:

N=100, n=500
P=prob. Of defective garments= 0.001
=np =500 x 0.001=0.5
let, X=no. of defective garments in a box of 500.
By poisson dist. Law , the prob. Of x defective
garments in a box given by
P(X=x) = e-0.5(0.5)x } /x! = 0.6065 x (0.5)2/ x!
,where , x=0,1,2,3.
Therefore , frequency(f) of boxes containing x
defective garments,
f (x)= N.P(X=x) =100x0.065x(0.5)2/x!
therefore, no. of boxes containing at least 2 defective
garments
= 100[p(X > 2)]
=100[1-p(X=0)- p(X=1)]
=100[1-0.6065-0.6065x0.5]
=100x0.09025
=9

17. When 200 samples from 5000 garments were taken

& tested and the distribution of fabric defects was
found to be as shown in the first and second columns of
the table below. Fit a Poisson distribution to the
frequency distribution of defects. Estimate the total
cost for correcting all the garments by using the
information given in the first and third columns of the
table below:
No. of defects Frequency Cost of
in a sample detection and
correction per
sample (Rs)
0 113 1.00
1 62 1.50
2 20 2.50
3 3 3.00
4 1 3.50
5 1 4.00
Solution:
FITTING OF POISSON DISTRIBUTION
 No. of defects Frequency fX
in a sample f
X
0 113 0
1 62 62
2 20 40
3 3 9
4 1 4
5 1 5
N=200 fX = 120

X = fX/N = 120/200 =
0.6
This is the mean of the Poisson distribution
i.e. m.
P3 = m = 2.7183-6 =0.5488
N ( P0 ) = 200*0.5488 = 109.76
N ( P1 ) = N ( P0 )*m = 109.76*0.6 = 65.856
 N ( P2 ) = N ( P1 )*(m/2) = 65.856*(0.6/2) =
19.756
N ( P3 ) = N ( P2 )*(m/3) = 19.757*(0.6/3) = 3-
951
N ( P4 ) = N ( P3 )*(m/4 )= -3.951*(0.6/4) =
0.5927
N ( P5 ) = N ( P4 )*(m/5) = 0.5927*(0.6/5) =
0.0711
The frequencies of the Poisson distribution are:
X 0 1 2 3
4 5
f 109.8 65.9 19.8 4.0
0.6 0.07
The total cost of correcting all the garments will
be given by:

No. of Rate per No. of fX

defects in a sample pages
sample X F
 0 1.00 109.80 109.80
1 1.50 65.90 98.85
2 2.50 19.80 49.50
3 3.00 4.00 12.00
4 3.50 0.60 2.10
5 4.00 0.07 0.28
fX =
272.53

Hence the total cost shall be Rs. 272.53.

18. A garment manufacturer knows that 5% of his

products are defective. If he sales the garment in carton
of 100, not more than 5 garments will be defective.
What is the approximate probability that the carton will
fail to meet the generated quality.
Solution
n=100
 p=0.05
= np =100*0.05 = 5
e x
P( X )
x!

P( X 5) 1 P( X 5)

e 5 x
=1- x 0
5

x!

5x
=1- e 5
x0
5

x!

=1-(6.74) 10 (91.34) 3

= 0.38
Probability that the carton will fail to meet the generated
quality is 38%