Notes on

Engineering Data Analysis

Notes on
Engineering Data Analysis

by

Dennis Noel G. de Lara

2023-Mar-02 , 0653H

Silliman University

Philippines

1 Discrete Probability Distributions 2

1.1The Binomial Distribution . . . . . . . 2
1.2The Poisson Distribution . . . . . . . . 15

Dennis Noel G. de Lara 2023-Mar-02, 0653H Page 1 of 22

Notes on
Engineering Data Analysis

Chapter 1

Discrete Probability Distributions

1.1 The Binomial Distribution

Consider an experiment that has the following
characteristics:

1. The experiment consists of a sequence of n

smaller experiments called trials, or
sub-experiments, where n is fixed in advance of
the experiment. A big experiment subdivided into
smaller experiments.
2. Each trial or sub-experiment can result in only
one of two possible outcomes, which we denote by
Success (S) or desirable outcomes, and Failure (F)
or undesirable outcomes.
3. The trials or sub-experiments are independent of
each other. This means that the outcome on a
particular trial does not influence the outcome of
any other trial.
4. The probability of success, p, is constant from one
trial to the next trial. Therefore, the probability
of failure q = (1 − p), is also constant from one trial
to the next trial.

An experiment for which conditions 1 to 4 are

satisfied is called a binomial experiment.

The binomial random variable X associated with a

binomial experiment consisting of n trials is defined
as:
X =number of successes for each experimental
outcome corresponding to the sequence of n trials
(1.1)

Dennis Noel G. de Lara 2023-Mar-02, 0653H Page 2 of 22

Notes on
Engineering Data Analysis
Experiment 1: “Break” or “Not Break”. n = 1

Apply the same constant load to a single concrete

specimen. Find out if the specimen will “break” or
“not break”. Suppose, statistical studies have shown
that the probability of this specimen not breaking is
0.70.

Number of specimens/trials: n = 1

Possible Trial Outcomes:

Trial Outcome 1: Specimen will “break (F)”.
failure. undesirable outcome
Trial Outcome 2: Specimen will “not break (S)”.
success. desirable outcome

It is also possible that the desired trial outcome is to

break the specimen.

Trial Outcome Probability of F: q = 1 − p = 0.30

Trial Outcome Probability of S: p = 0.70

Possible Experimental Outcomes:

Trial 1 Experimental No. of S,
Probability
Outcome Outcome x
F F X (F) = 0 1 − p = 0.3
S S X (S) = 1 p = 0.7

Sample Space for the Experiment:

{F, S}
(1.2)
2n = 21 = 2 experimental outcomes in all

What is the probability that the experimental

outcome is F?
P (F) = 1 − p = (1 − p)1 = 0.3 (1.3)

What is the probability that the experimental

outcome is S?
P (S) = p = p1 = 0.7 (1.4)

Dennis Noel G. de Lara 2023-Mar-02, 0653H Page 3 of 22

Notes on
Engineering Data Analysis
Possible Outcomes for X:
X = No. of S Probability
X=1 p1 = 0.7
X=0 (1 − p)1 = 0.3
There are n + 1 = 2 possible outcomes for X.

What is the probability that 0 specimens will not

break, that is, that X = 0?
P (X = 0) = 1 − p = (1 − p)1 = 0.3 (1.5)

What is the probability that 1 specimen will not

break, that is, that X = 1?
P (X = 1) = p = p1 = 0.7 (1.6)

Find the expected value or the mean:

n
X
µ= (pi xi )
i=0
= p 0 x0 + p 1 x1
(1.7)
= (0.3) (0) + (0.7) (1)

µ = 0.7 = np = (1) (0.7)

Find the variance, with the assumption of a

population:
Xn h i
2 2
σ = pi (xi − x̄)
i=0
= p0 (x0 − µ)2 + p1 (x1 − µ)2
(1.8)
= (0.3) (0 − 0.7)2 + (0.7) (1 − 0.7)2

σ 2 = 0.21 = np (1 − p) = (1) (0.7) (0.3)

What is the equation for the probability

distribution/probability mass function (the
probability that exactly x specimens will not break,
that is, that X = x)?



 0.3 x = 0




P (X = x) = 0.7 x = 1 (1.9)






0 elsewhere

Dennis Noel G. de Lara 2023-Mar-02, 0653H Page 4 of 22

Notes on
Engineering Data Analysis

Graph the probability distribution/probability mass

function.

What is the equation for the cumulative distribution

function (the probability that at most x specimens
will not break, that is, that X ≤ x)?



 0 x<0




CDF (X ≤ x) = 0.3 0 ≤ x < 1 (1.10)






1.0 1 ≤ x < ∞

Graph the cumulative distribution function.

What is the probability that at most 0 specimens will

not break, that is, that X ≤ 0?
P (X ≤ 0) = 0.3 (1.11)

What is the probability that at most 1 specimen will

not break, that is, that X ≤ 1?
P (X ≤ 1) = 0.3 + 0.7 = 1.0 (1.12)

Dennis Noel G. de Lara 2023-Mar-02, 0653H Page 5 of 22

Notes on
Engineering Data Analysis
Experiment 2: “Break” or “Not Break”. n = 2

Apply the same constant load to 2 concrete

specimens, in sequence. Find out if the specimens
will “break” or “not break”. Suppose, statistical
studies have shown that the probability of a specimen
not breaking is 0.70.

Number of specimens/trials: n = 2

Possible Trial Outcomes:

Trial Outcome 1: Specimen will “break (F)”.
failure. undesirable outcome
Trial Outcome 2: Specimen will “not break (S)”.
success. desirable outcome

Trial Outcome Probability of F: q = 1 − p = 0.30

Trial Outcome Probability of S: p = 0.70

Possible Experimental Outcomes:

Trial 1 Trial 2 Experimental No. of S,
Probability
Outcome Outcome Outcome x
F F FF X (FF) = 0 (1 − p)2 = 0.09
F S FS X (FS) = 1 (1 − p) p = 0.21
S F SF X (SF) = 1 p (1 − p) = 0.21
S S SS X (SS) = 2 p · p = p2 = 0.49

Sample Space for the Experiment:

{FF, FS, SF, SS}
(1.13)
2n = 22 = 4 experimental outcomes in all

What is the probability that the experimental

outcome is FF?
P (FF) = (1 − p)2 = 0.09 (1.14)

What is the probability that the experimental

outcome is FS?
P (FS) = (1 − p) p = 0.21 (1.15)

What is the probability that the experimental

outcome is SF?
P (SF) = (1 − p) p = 0.21 (1.16)
Dennis Noel G. de Lara 2023-Mar-02, 0653H Page 6 of 22
Notes on
Engineering Data Analysis
What is the probability that the experimental
outcome is SS?
P (SS) = p2 = 0.49 (1.17)

Possible Outcomes for X:

X = No. of S Probability
X=2 p2 = 0.49
X=1 2p (1 − p) = 0.42
X=0 (1 − p)2 = 0.09
There are n + 1 = 3 possible outcomes for X.

What is the probability that 0 specimens will not

break, that is, that X = 0?
P (X = 0) = (1 − p)2 = 0.09 (1.18)

What is the probability that 1 specimen will not

break, that is, that X = 1?
P (X = 1) = (1 − p) p + p (1 − p) = 2p (1 − p) = 0.42 (1.19)

What is the probability that 2 specimens will not

break, that is, that X = 2?
P (X = 2) = p2 = 0.49 (1.20)

Find the expected value or the mean:

n
X
µ= (pi xi )
i=0
= p 0 x 0 + p 1 x1 + p 2 x2
(1.21)
= (0.09) (0) + (0.42) (1) + (0.49) (2)

µ = 1.4 = np = (2) (0.7)

Find the variance, with the assumption of a

population:
Xn h i
2 2
σ = pi (xi − x̄)
i=0
= p0 (x0 − µ)2 + p1 (x1 − µ)2 + p2 (x2 − µ)2
= (0.09) (0 − 1.4)2 + (0.42) (1 − 1.4)2 + (0.49) (2 − 1.4)2

σ 2 = 0.42 = np (1 − p) = (2) (0.7) (0.3)

(1.22)
Dennis Noel G. de Lara 2023-Mar-02, 0653H Page 7 of 22
Notes on
Engineering Data Analysis
What is the equation for the probability
distribution/probability mass function (the
probability that exactly x specimens will not break,
that is, that X = x)?



0.09 x = 0





0.42 x = 1



P (X = x) = (1.23)

0.49 x = 2










0 elsewhere

Graph the probability distribution/probability mass

function.

What is the equation for the cumulative distribution

function (the probability that at most x specimens
will not break, that is, that X ≤ x)?



 0 x<0





0.09 0 ≤ x < 1



CDF (X ≤ x) = (1.24)

0.51 1 ≤ x < 2









1.00 2 ≤ x < ∞


Graph the cumulative distribution function.

What is the probability that at most 0 specimens will

not break, that is, that X ≤ 0?
P (X ≤ 0) = 0.09 (1.25)

What is the probability that at most 1 specimen will

not break, that is, that X ≤ 1?
P (X ≤ 1) = 0.09 + 0.42 = 0.51 (1.26)

What is the probability that at most 2 specimens will

not break, that is, that X ≤ 2?
P (X ≤ 2) = 0.09 + 0.42 + 0.49 = 1.0 (1.27)

Dennis Noel G. de Lara 2023-Mar-02, 0653H Page 8 of 22

Notes on
Engineering Data Analysis
Experiment 3: “Break” or “Not Break”. n = 3

Apply the same constant load to 3 concrete

specimens, in sequence. Find out if the specimens
will “break” or “not break”. Suppose, statistical
studies have shown that the probability of a specimen
not breaking is 0.70.

Number of specimens/trials: n = 3

Possible Trial Outcomes:

Trial Outcome 1: Specimen will “break (F)”.
failure. undesirable outcome
Trial Outcome 2: Specimen will “not break (S)”.
success. desirable outcome

Trial Outcome Probability of F: q = 1 − p = 0.30

Trial Outcome Probability of S: p = 0.70
Possible Experimental Outcomes:
Trial 1 Trial 2 Trial 3 Exper No. of S,
Probability
Outc Outc Outc Outcm x
F F F FFF X (FFF) = 0 (1 − p)3 = 0.027
F F S FFS X (FFS) = 1 (1 − p)2 p = 0.063
F S F FSF X (FSF) = 1 (1 − p) p (1 − p) = 0.063
F S S FSS X (FSS) = 2 (1 − p) p2 = 0.147
S F F SFF X (SFF) = 1 p(1 − p)2 = 0.063
S F S SFS X (SFS) = 2 p (1 − p) p = 0.147
S S F SSF X (SSF) = 2 p · p (1 − p) = 0.147
S S S SSS X (SSS) = 3 p3 = 0.343

Sample Space for the Experiment:

{FFF, FFS, FSF, FSS, SFF, SFS, SSF, SSS}
(1.28)
2n = 23 = 8 experimental outcomes in all

What is the probability that the experimental

outcome is FFF?
P (FFF) = (1 − p)3 = 0.027 (1.29)

What is the probability that the experimental

outcome is FFS?
P (X = 1) = (1 − p)2 p = 0.063 (1.30)

What is the probability that the experimental

Dennis Noel G. de Lara 2023-Mar-02, 0653H Page 9 of 22
Notes on
Engineering Data Analysis
outcome is FSF?
P (X = 1) = (1 − p) p (1 − p) = 0.063 (1.31)

What is the probability that the experimental

outcome is FSS?
P (X = 2) = (1 − p) p2 = 0.147 (1.32)

What is the probability that the experimental

outcome is SFF?
P (X = 1) = p(1 − p)2 = 0.063 (1.33)

What is the probability that the experimental

outcome is SFS?
P (X = 2) = p (1 − p) p = 0.147 (1.34)

What is the probability that the experimental

outcome is SSF?
P (X = 2) = p2 (1 − p) = 0.147 (1.35)

What is the probability that the experimental

outcome is SSS?
P (X = 3) = p3 = 0.343 (1.36)

Possible Outcomes for X:

X = No. of S Probability
X=3 p3 = 0.343
X=2 3p2 (1 − p) = 0.441
X=1 3p(1 − p)2 = 0.189
X=0 (1 − p)3 = 0.027
There are n + 1 = 4 possible outcomes for X.

What is the probability that 0 specimens will not

break, that is, that X = 0?
P (X = 0) = (1 − p)3 = 0.027 (1.37)

What is the probability that 1 specimen will not

break, that is, that X = 1?
P (X = 1) = (1 − p)2 p+(1 − p) p (1 − p)+p(1 − p)2 = 3p(1 − p)2 = 0.189
(1.38)
What is the probability that 2 specimens will not

Dennis Noel G. de Lara 2023-Mar-02, 0653H Page 10 of 22

Notes on
Engineering Data Analysis
break, that is, that X = 2?
P (X = 2) = 3p2 (1 − p) = 0.441 (1.39)

What is the probability that 3 specimens will not

break, that is, that X = 3?
P (X = 3) = p3 = 0.343 (1.40)

Find the expected value or the mean:

n
X
µ= (pi xi )
i=0
= p 0 x0 + p 1 x1 + p 2 x2 + p 3 x3
= (0.027) (0) + (0.189) (1) + (0.441) (2) + (0.343) (3)

µ = 2.1 = np = (3) (0.7)

(1.41)

Find the variance, with the assumption of a

population:
Xn h i
2 2
σ = pi (xi − x̄)
i=0
= p0 (x0 − µ)2 + p1 (x1 − µ)2 + p2 (x2 − µ)2 + p3 (x3 − µ)2
= (0.027) (0 − 2.1)2 + (0.189) (1 − 2.1)2 + (0.441) (2 − 2.1)2
+ (0.343) (3 − 2.1)2

σ 2 = 0.63 = np (1 − p) = (3) (0.7) (0.3)

(1.42)

What is the equation for the probability

distribution/probability mass function (the
probability that exactly x specimens will not break,
that is, that X = x)?


 0.027 x = 0






0.189 x = 1









P (X = x) = 0.441 x = 2 (1.43)






0.343 x = 3










0 elsewhere

Dennis Noel G. de Lara 2023-Mar-02, 0653H Page 11 of 22
Notes on
Engineering Data Analysis

Graph the probability distribution/probability mass

function.

What is the equation for the cumulative distribution

function (the probability that at most x specimens
will not break, that is, that X ≤ x)?


 0 x<0






0.027 0 ≤ x < 1









CDF (X ≤ x) = 0.216 1 ≤ x < 2 (1.44)






0.657 2 ≤ x < 3










1.000 3 ≤ x < ∞


Graph the cumulative distribution function.

What is the probability that at most 0 specimens will

not break, that is, that X ≤ 0?
P (X ≤ 0) = 0.027 (1.45)

What is the probability that at most 1 specimen will

not break, that is, that X ≤ 1?
P (X ≤ 1) = 0.027 + 0.189 = 0.216 (1.46)

What is the probability that at most 2 specimens will

not break, that is, that X ≤ 2?
P (X ≤ 2) = 0.027 + 0.189 + 0.441 = 0.657 (1.47)

What is the probability that at most 3 specimens will

not break, that is, that X ≤ 2?
P (X ≤ 3) = 0.027 + 0.189 + 0.441 + 0.343 = 1.0 (1.48)

Dennis Noel G. de Lara 2023-Mar-02, 0653H Page 12 of 22

Notes on
Engineering Data Analysis
What is the probability that at least 0 specimens will
not break, that is, that X ≥ 0?
P (X ≥ 0) = 1.000 − 0 = 1.000 (1.49)

What is the probability that at least 1 specimen will

not break, that is, that X ≥ 1?
P (X ≥ 1) = 1.000 − 0.027 = 0.973 (1.50)

What is the probability that at least 2 specimens will

not break, that is, that X ≥ 2?
P (X ≥ 2) = 1.000 − 0.216 = 0.784 (1.51)

What is the probability that at least 3 specimens will

not break, that is, that X ≥ 3?
P (X ≥ 3) = 1.000 − 0.657 = 0.343 (1.52)

What is the probability that the number of

specimens that will not break is between 0 and 1
inclusive, that is, that 0 ≤ X ≤ 1?
P (0 ≤ X ≤ 1) = 0.027 + 0.189 = 0.216 (1.53)

What is the probability that the number of

specimens that will not break is between 0 and 3
exclusive, that is, that 0 < X < 3?
P (0 < X < 3) = 0.189 + 0.441 = 0.63 (1.54)

What is the probability that the number of

specimens that will not break is between 0 and 3 but
not including 3, that is, that 1 ≤ X < 3?
P (0 < X < 3) = 0.027 + 0.189 + 0.441 = 0.657 (1.55)

Dennis Noel G. de Lara 2023-Mar-02, 0653H Page 13 of 22

Notes on
Engineering Data Analysis

Given n trials/specimens.
What is the probability that x specimens will not
break?

(n)! x n−x
 (x)! (n − x)! p (1 − p) x = 0, 1, 2, 3, · · · , n



P (X = x; n, p) =



0 elsewhere
(1.56)

This is the PMF of the Binomial Distribution.

Expected value, µ, and variance, σ 2 , of X of the

binomial distribution, assuming a population:
µ = np
(1.57)
σ 2 = np (1 − p)
This is the CDF of the Binomial Distribution.


 0 x<0






 ⌊x⌋  
(n)!
X
n−i
CDF (X ≤ x; n, p) = pi (1 − p) x≥0


 i=0
(i)! (n − i)!






1 x→∞



 0 x<0





0 
 
(n)!

 X
 i n−i

 p (1 − p) 0≤x<1
(i)! (n − i)!


i=0








 1  

 X (n)! i n−i
p (1 − p) 1≤x<2


(i)! (n − i)!



 i=0

=
 2  
 X (n)!
pi (1 − p)n−i

2≤x<3


−



 i=0
(i)! (n i)!
.. ..


. .




⌊x⌋ 

 

 X (n)! i n−i
p (1 − p) ⌊x⌋ < ⌊x⌋ + 1


−



 i=0
(i)! (n i)!
.. ..


. .





1 x→∞

(1.58)

Dennis Noel G. de Lara 2023-Mar-02, 0653H Page 14 of 22

Notes on
Engineering Data Analysis

1.2 The Poisson Distribution

Experiment 4: “Weak” or “Not Weak”. n = 300

For a set of concrete specimens, a company exerts

every effort to ensure that the probability of finding
any single specimen being weak (low compressive
strength) is 0.01. Finding a weak specimen is
independent from one specimen to another. There
are 300 specimens in a given set of specimens.

Use the binomial distribution.

Find the expected value or the mean:

µ = np
= (300) (0.01) (1.59)
µ=3

On the average, you would find 3 specimens that are

weak in the 300-specimen set.

Find the variance, with the assumption of a

population:
σ 2 = np (1 − p)
= (300) (0.01) (1 − 0.01) (1.60)
σ 2 = 2.97

If the given set would be inspected in search of weak

specimens, what is the probability of finding exactly
0 weak specimens?

(n)!
P (X = x; n, p) = px (1 − p)n−x (1.61)
(x)! (n − x)!

(300)!
P (X = 0; 300, 0.01) = (0.01)0 (1 − 0.01)300−0
(0)! (300 − 0)!

P (X = 0; 300, 0.01) = 0.04904089 · · ·

(1.62)

Dennis Noel G. de Lara 2023-Mar-02, 0653H Page 15 of 22

Notes on
Engineering Data Analysis
If the given set would be inspected in search of weak
specimens, what is the probability of finding exactly
1 weak specimen?

(n)!
P (X = x; n, p) = px (1 − p)n−x (1.63)
(x)! (n − x)!

(300)!
P (X = 1; 300, 0.01) = (0.01)1 (1 − 0.01)300−1
(1)! (300 − 1)!

P (X = 1; 300, 0.01) = 0.14860876 · · ·

(1.64)

If the given set would be inspected in search of weak

specimens, what is the probability of finding exactly
2 weak specimen?

(n)!
P (X = x; n, p) = px (1 − p)n−x (1.65)
(x)! (n − x)!

(300)!
P (X = 2; 300, 0.01) = (0.01)2 (1 − 0.01)300−2
(2)! (300 − 2)!

P (X = 2; 300, 0.01) = 0.22441425 · · ·

(1.66)

If the given set would be inspected in search of weak

specimens, what is the probability of finding exactly
3 weak specimen?

(n)!
P (X = x; n, p) = px (1 − p)n−x (1.67)
(x)! (n − x)!

(300)!
P (X = 3; 300, 0.01) = (0.01)3 (1 − 0.01)300−3
(3)! (300 − 3)!

P (X = 3; 300, 0.01) = 0.22516985 · · ·

(1.68)

Dennis Noel G. de Lara 2023-Mar-02, 0653H Page 16 of 22

Notes on
Engineering Data Analysis
If the given set would be inspected in search of weak
specimens, what is the probability of finding exactly
4 weak specimen?

(n)!
P (X = x; n, p) = px (1 − p)n−x (1.69)
(x)! (n − x)!

(300)!
P (X = 4; 300, 0.01) = (0.01)4 (1 − 0.01)300−4
(4)! (300 − 4)!

P (X = 4; 300, 0.01) = 0.16887739 · · ·

(1.70)

If the given set would be inspected in search of weak

specimens, what is the probability of finding exactly
5 weak specimens?

(n)!
P (X = x; n, p) = px (1 − p)n−x (1.71)
(x)! (n − x)!

(300)!
P (X = 5; 300, 0.01) = (0.01)5 (1 − 0.01)300−5
(5)! (300 − 5)!

P (X = 5; 300, 0.01) = 0.100985 · · ·

(1.72)

If the given set would be inspected in search of weak

specimens, what is the probability of finding exactly
6 weak specimen?

(n)!
P (X = x; n, p) = px (1 − p)n−x (1.73)
(x)! (n − x)!

(300)!
P (X = 6; 300, 0.01) = (0.01)6 (1 − 0.01)300−6
(6)! (300 − 6)!

P (X = 6; 300, 0.01) = 0.05015261 · · ·

(1.74)

Dennis Noel G. de Lara 2023-Mar-02, 0653H Page 17 of 22

Notes on
Engineering Data Analysis
Let’s try to calculate the probabilities using the
Poisson Distribution, using µ = np = 3.

If the given set would be inspected in search of weak

specimens, what is the probability of finding exactly
0 weak specimens?

−µ µx
P (X = x; µ) = e (1.75)
(x)!

30
−3
P (X = 0; 3) = e
(0)!
(1.76)

P (X = 0; 3) = 0.04978706 · · ·

If the given set would be inspected in search of weak

specimens, what is the probability of finding exactly
1 weak specimens?

−µ µx
P (X = x; µ) = e (1.77)
(x)!

31
−3
P (X = 1; 3) = e
(1)!
(1.78)

P (X = 1; 3) = 0.14936120 · · ·

If the given set would be inspected in search of weak

specimens, what is the probability of finding exactly
2 weak specimens?

−µ µx
P (X = x; µ) = e (1.79)
(x)!

32
−3
P (X = 2; 3) = e
(2)!
(1.80)

P (X = 2; 3) = 0.22404180 · · ·

Dennis Noel G. de Lara 2023-Mar-02, 0653H Page 18 of 22

Notes on
Engineering Data Analysis
If the given set would be inspected in search of weak
specimens, what is the probability of finding exactly
3 weak specimens?

−µ µx
P (X = x; µ) = e (1.81)
(x)!

33
−3
P (X = 3; 3) = e
(3)!
(1.82)

P (X = 3; 3) = 0.22404180 · · ·

If the given set would be inspected in search of weak

specimens, what is the probability of finding exactly
4 weak specimens?

−µ µx
P (X = x; µ) = e (1.83)
(x)!

34
−3
P (X = 4; 3) = e
(4)!
(1.84)

P (X = 4; 3) = 0.16803135 · · ·

If the given set would be inspected in search of weak

specimens, what is the probability of finding exactly
5 weak specimens?

−µ µx
P (X = x; µ) = e (1.85)
(x)!

35
−3
P (X = 5; 3) = e
(5)!
(1.86)

P (X = 5; 3) = 0.10081881 · · ·

Dennis Noel G. de Lara 2023-Mar-02, 0653H Page 19 of 22

Notes on
Engineering Data Analysis
If the given set would be inspected in search of weak
specimens, what is the probability of finding exactly
6 weak specimens?

−µ µx
P (X = x; µ) = e (1.87)
(x)!

36
−3
P (X = 6; 3) = e
(6)!
(1.88)

P (X = 6; 3) = 0.05040940 · · ·

Compare the results for the Binomial Distribution

and the Poisson Distribution.

When n → ∞ and p → 0, and np → µ, the Binomial

distribution approaches the Poisson distribution.

Rule of Thumb: If n > 50 and np < 5, we can use the

Poisson Distribution to approximate the Binomial
Distribution.

Find the expected value or the mean for the Poisson

Distribution:

Expected value = µ (1.89)

Find the variance for the Poisson Distribution:

Variance = µ (1.90)

Dennis Noel G. de Lara 2023-Mar-02, 0653H Page 20 of 22

Notes on
Engineering Data Analysis
Poisson Process
number of earthquakes for a given period of time

number of vehicles passing over a bridge per minute

number of people in a building per hour

number of reinforced concrete specimens

manufactured per day

In the Poisson process, we are interested in the

number of occurrence of an event per unit time. Or,
the probability that a certain number of occurrence
of an event will happen for a given period of time.

Binomial: probability of the number of occurrence of

an event

Poisson: probability of the number of occurrence of

an event for a given fixed period of time
k
−αt (αt)
P (t; α, k) = e (1.91)
(k)!

This is the probability that k events will occur for a

period of time t, where α is a constant of
proportionality which signifies the probability of
exactly one event as time becomes very small (an
“average” value).

Dennis Noel G. de Lara 2023-Mar-02, 0653H Page 21 of 22

Notes on
Engineering Data Analysis
Example:

The number of vehicles passing over a bridge per

second is, on the average, 4. The number of vehicles
is a random variable with a Poisson Distribution. We
would like to find the probability of vehicles passing
through the bridge.

vehicles
α=4 (1.92)
second

What is the probability that exactly 6 vehicles will

pass through the bridge in 5 seconds?

k = 6 vehicle
(1.93)
t = 5 seconds
k
−αt (αt)
P (t; α, k) = e (1.94)
(k)!

· 5)6
−(4·5) (4
P (5; 4, 6) = e
(6)! (1.95)
P (5; 4, 6) = 0.0001832

Dennis Noel G. de Lara 2023-Mar-02, 0653H Page 22 of 22