Scribd
Upload
63 views

CST-501 Simulation (Period 8) : Presented by Daw Ank Phyu Win Associate Professor

The document discusses various methods for generating random observations from different probability distributions using inverse transformation. These include: 1) The inverse transformation method for uniform distributions by generating a uniform random number r and setting it equal to the cumulative distribution function F(x). 2) Examples of using this method to generate random observations from uniform, exponential, Erlang, and normal distributions. 3) An example problem where service times for customers are simulated using a uniform distribution between 3 to 8 minutes and the average service time is calculated.

Uploaded by

Win Myo
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
63 views

CST-501 Simulation (Period 8) : Presented by Daw Ank Phyu Win Associate Professor

The document discusses various methods for generating random observations from different probability distributions using inverse transformation. These include: 1) The inverse transformation method for uniform distributions by generating a uniform random number r and setting it equal to the cumulative distribution function F(x). 2) Examples of using this method to generate random observations from uniform, exponential, Erlang, and normal distributions. 3) An example problem where service times for customers are simulated using a uniform distribution between 3 to 8 minutes and the average service time is calculated.

Uploaded by

Win Myo
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 12

CST-501

Simulation ( Period 8 )

Presented by

Daw Ank Phyu Win

Associate Professor
Generation of Random Observations from a Probability Distribution

The Inverse Transformation Method

• The cumulative distribution function

• Summary of inverse transformation method

• Generate a uniform random number r between 0 and 1.

• Set F(x) = r and solve for x, which then is the desired random observation from the
probability distribution.
Problem 22.4-3(a)

• to generate three random observations

• uniform random numbers: 0.096, 0.569, 0.665

(a) the uniform distribution from 25 to 75

= = [x] 0.096 29.80

0.569 53.45
=
0.665 5.25
Problem 22.4-3(b)

0.096 0.1133
= (
0.569 0.737
= 0.665 0.06

-1
Problem 22.4-3(c)

• r x=

• = ( 0.096 46.197

0.569 55.086
• =
0.665 56.310
• +40
Problem 22.4-5

,
b. to generate five random observations c. to identify the three uniform random
numbers

Uniform random Random


number, r observations, x Random
observations, , r
0.0956 0.3092
x
0.5629 0.7503 0.09 0.0081
0.6659 0.8160
0.64 0.4096
0.7634 0.8737
0.49 0.2401
0.8426 0.9179
Problem 22.4-7

• to serve a customer on each visit has a uniform distribution between 3 minutes and 8 minutes

• to apply the inverse transformation method

• to simulate the service times for five customers,

• to calculate the average of the five service times

• to compare it to the mean of the service-time distribution

• to apply this method graphically,

• to apply this method algebraically


Problem 22.4-7 (continued)

a. uniform random numbers: 0.6505, 0.0740, 0.8443, 0.4975, 0.8178

1
0.8443
0.8178
Uniform random number

0.6505
0.4975

0.0740

0 3 8

7.22
5.49

6.25
3.37

7.09
Service time (minutes)
Problem 22.4-7

b. = = [x] = (x-3)

x=5r+3

r x = 5r + 3
0.6505 6.25
0.0740 3.37
0.8443 7.22
0.4975 5.49
0.8178 7.09

c. average service time = 5.88


mean service time distribution =
it is higher than mean of 5.5.
Exponential and Erlang Distributions, Normal Distributions

• the cumulative distribution function for the exponential distribution

, for

• the desired random observation from the Erlang distribution

• a random observation from a normal distribution


Problem 22.4-13

to generate two random observations

a. by the exponential distribution b. by the Erlang distribution

mean = mean = , k=2

= = -4 ln (1-r) = -2 ln [(1-1-

r x
0.096 29.80 0.096 0.569 1.8851
0.569 53.45 0.665 0.764 5.075
Problem 22.4-13 (continued)

by the normal distribution

= 4, ,
=4(

x=4(
0.096 0.224 3.428 5.712
0.569 0.950 2.965 3.860
0.665 0.610
0.764 0.145
0.842 0.484
0.492 0.552

You might also like