Poisson distribution

Definition: in probability theory and statistics, the Poisson distribution, named after French mathematics
Simeon Denis Poisson, is a discrete probability distribution that expresses the probability of a given
number of events occurring in a fixed interval of time or space.

Examples:

➢ The number of planes landing at an airfield in 30 min.

➢ The number of emergency calls received by an ambulance control centre in 1 hour.
➢ The number of bacteria in 10 ml of pond water.

Condition for a Poisson distribution

X is said to follow a Poisson distribution if the events occur:

➢ Independently
➢ Singly
➢ Randomly
➢ At a constant average (mean) rate.

If the constant average rate is , then the distribution can be written as

𝑋~𝑃𝑜(𝜆)

Formula

−𝝀
𝝀𝒓
𝑷(𝑿 = 𝒓) = 𝒆 𝒘𝒉𝒆𝒓𝒆 𝒓 = 𝟎, 𝟏, 𝟐, 𝟑, 𝟒, … … … 𝒕𝒐 𝒊𝒏𝒇𝒊𝒏𝒊𝒕𝒚
𝒓!
i.e Poisson distribution has No upper limit.
Examples:

1. The number of amoebas in the water from a particular pond follow a Poisson distribution with
mean of 4 amoebas per 10 ml of pond water . find the probability that in 10 ml of pond water
there are ;
i. Exactly 5 amoebas.
ii. at most 2 amoebas
iii. at least 1 amoeba
Ans: 0.0156 ;0.238 ; 0.982
2. The discrete random variable X follows a Poisson distribution with parameter 5. Calculate ;
a. 𝑃(𝑋 > 3)
b. 𝑃(5 < 𝑋 ≤ 7)
𝐴𝑛𝑠: 0.735 ; 0.251
3. Weak spots in the manufacture of a certain cable , independently and randomly ,at an average
rate of 2.5 weak spots per 100 m length of cable.
i. Find the probability in a 100 m length of cable , there will be :
a. No weak spots
b. More than 2 weak spots.
Mijan cut 8 lengths of cable. Each 100 m long.
ii. Find the probability that exactly 2 of them have No weak spots.
Ans: 0.0821 ; 0.456 ; 0.113
4. A shopkeeper sells electric fans. The demand for fans follows a Poisson distribution with mean
3.2 per week.
i. Find the probability that the demand is exactly 2 fans in any one week.
ii. The shopkeeper has 4 fans in his shop at the beginning of a week. Find the probability that this
will not be enough to satisfy the demand for fans in that week.

1
 iii. Given instead that he has n fans in his shop at the beginning of a week, find, by trial and error, the
least value of n for which the probability of his not being able to satisfy the demand for fans in
that week is less than 0.05.

Ans:0.209 ; 0.219 ; 6

Home works

1. Cars arrive singly, randomly and independently at a large service station at an average rate of 18
per minute. Find the probability that in a randomly chosen one minute period the number of cars
arriving at the service station is :
a. Exactly 2 cars arrive.
b. At least 2 cars arrive.
c. Fewer than 5 cars arrive.
Ans: 0.268 ; 0.269 ;0.964
2. An insurance company receives an average of 2 claims a week from a particular company.
Assuming that the number of claims my be modelled by a Poisson distribution, find the
probability that the insurance company receives.
i. exact 3 claims in a given week
ii. more than 4 claims in a given week
iii. no claims in a given week
iv. at least one claim each week in 4 consecutive weeks.
Ans:0.180 ; 0.0527 ; 0.135 ; 0.559

3. The number of births announced in the personal column of local weekly newspaper may be
modelled by a Poisson distribution with mean 3.2 per week. Find the probability that, in a
particular week, the number of births announced is.
i. at most 3
ii. exactly 4
iii. more than 2
Ans:0.603 ; 0.178 ; 0.620
4. A car hire firm has 3 cars for hire. The number of cars required each day follow a Poisson
distribution with mean 2.1 .
i. find the probability that , on any given day :
a. exactly 2 cars are requested
b. all the cars are in use .
c. the firm will not be able to meet the demand.
ii. Find the probability that at least 1 car is requested each day over a period of 6 days.
Ans:0.270; 0.350 ; 0.161 ; 0.457
5. Along a stretch of motorway, vehicle breakdown occur at an average rate of 2.4 per day. Stating
any necessary assumptions.
i. Find the probability that there will be more than 4 breakdowns in a given day.
ii. Find by trial and error, the least integer 𝑛 such that the probability of more than 𝑛
breakdown in a day is less than 0.03.
Ans: breakdown occur singly; randomly and independently; 0.0959 ; 6

2
 Unit interval
Care must be taken to specify the unit interval being considered.
Example
the number of amoebas in 10 ml of pond water is 4.
so the number of amoebas in 10 ml is distributed 𝑃𝑜(4).
Now suppose you want to find a probability related to 5 ml of the pond water.
so the mean number of amoebas in 5 ml is 2.
i.e the number of amoebas in 5 ml distributed 𝑃𝑜(2).
similarly ;
the number of amoebas in 30 ml is distributed 𝑃𝑜(12).
Examples:
1. In a certain school; photocopier breakdowns occur randomly and independently
on average 8 times in a school week from Monday to Friday .Find the probability
that there will be:
i. 5 breakdowns in a week.
ii. 1 breakdown on Monday.
iii. 8 breakdowns in a fortnight.
Ans: 0.0916; 0.323; 0.0120
2. At a certain airfield planes land at random times at a constant average rate of one every
10 minutes.
i. Find the probability that exactly 5 planes will land in a period of one hour.
ii. Find the probability that at least 2 planes will land in a period of 16 minutes.
iii. Given that 5 planes landed in an hour, calculate the conditional probability that 1 plane
landed in the first half hour and 4 in the second half hour.
Ans: 0.161; 0.475; 0.156
3. The number of accidents per day at a large factory follows a Poisson distribution with
parameter 2.25.
i. Sate the mean and standard deviation of the number of accidents per day.
ii. On how many days in November would you expect the number of accidents to be
within one standard deviation of the mean?
Ans: 2.25; 1.5 ; 21 Days
4. The random variable X is distributed 𝑃𝑜(0.3𝑛).
i. Find 𝑃(𝑋 ≥ 1) when n=4.
ii. Find the largest integer value of 𝑛 so that 𝑃(𝑋 ≥ 1) is greater than 0.95.
Ans: 0.699; 10

3
Homework
1. Emails arrive on Oliver’s computer at an average rate of 1.46 per hour. stating a
necessary assumption , find the probability that between 9.00 am and 11.30 am Oliver
receives ;
i. No emails ii. At most 4 emails
Ans: independently; 0.026 ; 0.697
2. During each afternoon, helicopters land on a holiday island one at a time at a constant
average rate of 1 every 20 minutes. find the probability that , on a particular afternoon :
i. Exactly 2 helicopters landed between 2.30 pm and 3.30 pm
ii. At least 2 helicopters land between 3.40 pm and 4.20 pm.
iii. No helicopter land between 2.00 pm and 2.15 pm.
Ans: 0.224 ; 0.594 ; 0.472
3. The number of letters advertising various products that Alex receives per day may be
modelled by a Poisson distribution with mean 0.5.
Find the probability that Alex receives:
i. Exactly 2 letters on a particular day.
ii. No letters on 3 consecutive days.
iii. More than 4 letters in a 6-day period.
Ans: 0.0758 ; 0.223 ; 0.185

4. The number of emergency telephone calls to the electricity board office in a certain area
1
in t minutes is known to follow a Poisson distribution with mean 80 𝑡.
i. Find the probability that there will be at least 3 emergency telephone calls to the
office in any 20-minute period.
ii. The probability that no emergency telephone call is made to the office in a period of
k minutes is 0.9. Find k.
Ans: 0.00216 ; 8.43
5. People arrive randomly and independently at the elevator in a block of flats at an
average rate of 4 people every 5 minutes.
i. Find the probability that exactly two people arrive in a 1-minute period.
ii. Find the probability that nobody arrives in a 15-second period.
iii. The probability that at least one person arrives in the next t minutes is 0.9. Find
the Value of t.
Ans: i. 0.144 ii. 0.819

4
The Poisson distribution as an approximation to the binomial distribution

Binomial distribution 𝑛 > 50 𝑎𝑛𝑑 𝑛𝑝 < 5 Poisson approximation

𝑋~𝐵(𝑛, 𝑝) with mean (𝜆) = 𝑛𝑝

Error =|𝐴𝑐𝑡𝑢𝑎𝑙 (𝑢𝑠𝑒 𝑏𝑖𝑛𝑜𝑚𝑖𝑎𝑙) − 𝐸𝑠𝑡𝑖𝑚𝑎𝑡𝑒𝑑 𝑣𝑎𝑙𝑢𝑒 (𝑃𝑜𝑖𝑠𝑠𝑜𝑛 𝑎𝑝𝑝𝑟𝑜𝑥𝑖𝑚𝑎𝑡𝑖𝑜𝑛)|

𝐸𝑟𝑟𝑜𝑟
𝐸𝑟𝑟𝑜𝑟 % = × 100%
𝑎𝑐𝑡𝑢𝑎𝑙

Examples :

1. The random variable X follows a binomial distribution, where ~𝐵(100,0.03) . Find the following
probabilities using
i. using binomial distribution
ii. a suitable Poisson approximation ;
iii. Find error percentage of each case.
a. 𝑃(𝑋 = 0)
b. 𝑃(𝑋 = 2)
c. 𝑃(𝑋 = 4)
` Ans:
a. 0.0476 ; 0.0498 ; 4.62%
b. 0.225 ; 0.224 ; 0.444%
c. 0.171 ; 0.168 ; 1.75%
2. A newspaper reports that 8.6% of people in a particular region are left – handed. A random sample of
55 people in the region is selected. Use a suitable approximation to find the probability that fewer
than 4 people in the sample are left – handed.
Ans:0.305
3. The probability that a particular type of cell phone works properly when it is first used is 0.97. During
a busy weekend, a store sells 150 of this type of cell phone. Using a suitable approximation, find the
probability that more than 146 of the cell phones work properly when they are first used.
Ans:0.342
4. On average, 1 in 2500 people have a particular gene.
i. Use a suitable approximation to find the probability that, in a random sample of 10 000 people, more
than 3 people have this gene.
The probability that, in a random sample of n people, none of them has the gene is less than 0.01.
ii. Find the smallest possible value of n.
Ans:0.567 ; 11513

5. The probability that a new car of a certain type has faulty brakes is 0.008. A random sample of 520
new cars of this type is chosen, and the number, X, having faulty brakes are noted.
i. Describe fully the distribution of X and describe also a suitable approximating distribution. Justify this
approximating distribution.
Use your approximating distribution to find
a. 𝑃(𝑋 > 3)
b. the smallest value of n such that 𝑃(𝑋 = 𝑛) > 𝑃(𝑋 = 𝑛 + 1).
Ans: 𝐵(520 ,0.008) 𝑃𝑜(4.16) 𝑖𝑖. 0.597 𝑖𝑖𝑖. 𝑛 = 4

5
 Home works
1. The proportion of people who have a particular gene, on average, is 1 in 1000.
A random sample of 3200 people in a certain country is chosen and the number of people, X, having the
gene is found.
(i) State the distribution of X and state also an appropriate approximating distribution. Give the
values of any parameters in each case. Justify your choice of the approximating distribution.
Use the approximating distribution to find P(X ≤ 3)
Ans: 𝐵(3500,0.001) 𝑛 > 50 𝑎𝑛𝑑 𝑛𝑝 < 5 𝑖𝑖. 0.537
2. In a certain lottery, 10 500 tickets have been sold altogether and each ticket has a probability of 0.0002
of winning a prize. The random variable X denotes the number of prize-winning tickets that have been
sold.
(i) State, with a justification, an approximating distribution for X.
(ii) Use your approximating distribution to find P(X < 4).
(iii) Use your approximating distribution to find the conditional probability that X < 4, given that
X ≥ 1.
Ans: 𝑃𝑜𝑖𝑠𝑠𝑜𝑛 𝑖𝑖. 0.839 𝑖𝑖. 0.816
3. A nursery found that on average 98% of sunflower seeds germinate. Using a suitable approximation, find
the probability that in a packet containing 150 sunflower seeds.
i. More than 4 do not germinate.
ii. At least 145 germinate.
Ans:0.185 ;0.916
4. An aircraft has 240 seats. Past records show that, on average, 2% of people who have booked tickets for
a flight do not turn up .the airline sells 244 tickets for a particular flight. Using a suitable approximation
find the probability that :
i. exactly 4 people do not turn up for the flight
ii. There are empty seats on the flight.
iii. There are not enough seats for everyone who turns up for the flight.
Ans:0.180 ; 0.538 ; 0.282

6
Sum of two Poisson variables
Now suppose we know that in hospital A births occur randomly at an average rate of 2.3 births per
hour and in hospital B births occur randomly at an average rate of 3.1 births per hour. What is the
probability that we observe 7 births in total from the two hospitals in a given 1 hour period?
Solution: To answer this question we can use the following rule
So if we let X = No. of births in a given hour at hospital A
and
Y = No. of births in a given hour at hospital B
Then 𝑋 ~ 𝑃𝑜(2.3), 𝑌 ~ 𝑃𝑜(3.1) 𝑎𝑛𝑑 𝑋 + 𝑌 ~ 𝑃𝑜(5.4)
(5: 4)7
⟹ 𝑃(𝑋 + 𝑌 = 7) = 𝑒 5.4 = 0.11999
7!
Examples:
1. At a petrol station cars arrive independently and at random times at constant average rates
of 8 cars per hour travelling east and 5 cars per hour travelling west.
i. Find the probability that, in a quarter-hour period,
a. one or more cars travelling east and one or more cars travelling west will arrive,
Ans: a total of 2 or more cars will arrive. i. a. 0.617 b. 0.835

2. A hotel kitchen has two dish-washing machines. The numbers of breakdowns per year by the
two machines have independent Poisson distributions with means 0.7 and 1.0. Find the
probability that the total number of breakdowns by the two machines during the next two
years will be less than 3.
Ans: 0.34
3. A dressmaker makes dresses for GEMS school. Each dress requires 2.5𝑚2 of material. Faults
occur randomly in the material at an average rate of 4.8 𝑝𝑒𝑟 20𝑚2.
i. Find the probability that a randomly chosen dress contains at least 2 faults.
Each dress has a belt attached to it to make an outfit. Independently of faults in the
material, the probability that a belt is faulty is 0.03. Find the probability that, in an outfit,
ii. neither the dress nor its belt is faulty,
iii. The dress has at least one fault and its belt is faulty.
Ans: i. 0.122 ii. 0.532 iii.0.0135

7
Home works
1. The number of lions seen per day during a standard safari has the distribution Po(0.8). The
number of lions seen per day during an off-road safari has the distribution Po(2.7). The two
distributions are independent.
i. Asmita goes on a standard safari for one day. Find the probability that she sees at
least 2 lions.
ii. Sajina goes on a standard safari for 3 days and then on an off-road safari for 2 days.
Find the probability that she sees a total of fewer than 5 lions.
iii. Aman goes on a standard safari for n days, where n is an integer. He wants to ensure
that his chance of not seeing any lions is less than 10%. Find the smallest possible
value of n.
Ans: i. 0.191 ii.0.112
2. The independent random variables X and Y have the distributions Po(2) and Po(3)
respectively. Given that X + Y = 5, find the probability that 𝑋 = 1 𝑎𝑛𝑑 𝑌 = 4.
Ans: 0.259
3. (i) The random variable W has the distribution Po(1.5). Find the probability that the
sum of 3 independent values of W is greater than 2.
(ii) The random variable X has the distribution Po(𝜆). Given that𝑃(𝑋 = 0) = 0.523,
find the value of , correct to 3 significant figures.
(ii) The random variable Y has the distribution Po(𝜇), where μ ≠ 0. Given that
𝑃(𝑌 = 3) = 24 × 𝑃(𝑌 = 1) Find 𝜇.
Ans: i. 0.826 ii. 0.648 iii. 12

8
 Normal approximation to the Poisson distribution
𝑿~𝑷𝒐(𝝀)

Poisson 𝜆 > 15 Normal

distribution Continuity correction by 0.5 Approximatio
n

Mean (𝜇) = 𝜆

Variance (𝜎 2 ) = 𝜆

Examples:

1. The battery in Sue’s phone runs out at random moments. Over a long period, she has found that
the battery runs out, on average, 3.3 times in a 30-day period. Use an approximating distribution
to find the probability that the battery runs out more than 50 times in a year (365 days).
Ans: 0.0513
2. Computer breakdowns occur randomly on average once every 48 hours of use.
i. Calculate the probability that there will be fewer than 4 breakdowns in 60 hours of use.
ii. Find the probability that the number of breakdowns in one year (8760 hours) of use is more
than 200.
iii. Independently of the computer breaking down, the computer operator receives phone calls
randomly on average twice in every 24-hour period. Find the probability that the total number
of phone calls and computer breakdowns in a 60-hour period is exactly 4.
Ans: 0.962 ; 0.0915 ; 0.123
3. At a petrol station cars arrive independently and at random times at constant average rates of 8
cars per hour travelling east and 5 cars per hour travelling west.
i. Find the probability that, in a quarter-hour period,
a. one or more cars travelling east and one or more cars travelling west will arrive,
b. a total of 2 or more cars will arrive.
ii. Find the approximate probability that, in a 12-hour period, a total of more than 175 cars will
arrive.
Ans: 0.617; 0.835 ; 0.0593

Home works

9
 1. A publishing firm has found that errors in the first draft of a new book occur at random and
that, on average, there is 1 error in every 3 pages of a first draft. Find the probability that in a
particular first draft there are
i. exactly 2 errors in 10 pages,
ii. at least 3 errors in 6 pages,
iii. Fewer than 50 errors in 200 pages.
Ans: 0.198 ; 0.323 ; 0.0178
2. People arrive at a checkout in a store at random, and at a constant mean rate of 0.7 per
minute. Find the probability that
i. exactly 3 people arrive at the checkout during a 5-minute period,
ii. at least 30 people arrive at the checkout during a 1-hour period.
People arrive independently at another checkout in the store at random, and at a constant mean
rate of 0.5 per minute.
iii. Find the probability that a total of more than 3 people arrive at this pair of checkouts during a
2-minute period.
Ans: 0.216 ; 0.973; 0.221
3. Particles are emitted randomly from a radioactive substance at a constant average rate of
3.6 per minute. Find the probability that
i. more than 3 particles are emitted during a 20-second period,
ii. More than 240 particles are emitted during a 1-hour period.
Ans: 0.0338 ; 0.0478
4. Men arrive at a clinic independently and at random, at a constant mean rate of 0.2 per
minute. Women arrive at the same clinic independently and at random, at a constant mean
rate of 0.3 per minute.
i. Find the probability that at least 2 men and at least 3 women arrive at the clinic during a
5-minute period.
ii. Find the probability that fewer than 36 people arrive at the clinic during a 1-hour
period.
Ans: 0.0505 ; 0.842

10