Scribd
Upload
5 views12 pages

Lecture 35 - Simulation Modeling I

Uploaded by

nehasidar152
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
5 views12 pages

Lecture 35 - Simulation Modeling I

Uploaded by

nehasidar152
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 12

LECTURE 35

SIMULATION MODELING I

DR. ANUP KUMAR TRIPATHI

DEPARTMENT OF MINING ENGINEERING


NATIONAL INSTITUTE OF TECHNOLOGY KARNATAKA (NITK)
SURATHKAL − 575025, INDIA

1
SIMULATION MODELING I

INTRODUCTION
 Simulation is the next best thing to observing a real system in operation.

It allows collection of pertinent information about the behavior of the system by running a
computerized model.

The task of using simulation starts with the development of the logic of the computer model in a
manner that will allow collecting needed data.

The collected data is used to design the system.

Simulation is not an optimization technique.

Rather it is a technique for estimating the measures of performance of the modeled system.

2
 Simulation has applications and uses in all aspects of science and technology as the following
partial list of examples demonstrates.

Examples
1. Basic science
 Estimation of area under a curve
 Evaluation of multiple integrals
 Matrix inversion
 Study of particle diffusion

2. Practical situations
(a). Industrial problems
 Design of queueing systems
 Communication networks
 Inventory control
 Chemical processes

3
(b). Business and economic problems
 Consumer behavior
 Price determination
 Economic forecasting
 Total firm operation

(c). Behavioral and social problems


 Population dynamics
 Environmental health effects
 Epidemiological studies
 Group behavior

4
(d). Biomedical systems
 Fluid balance
 Electrolyte distribution in body
 Blood cell proliferation
 Brain activities

(e). War strategies and tactics

 Estimation of simulation output is based on random sampling, much the same way we do when
observing real situation.

This means that output of simulation is subject to random variations, and thus, as in any statistical
experiment, must be examined using formal statistical inference tests.

5
MONTE CARLO SIMULATION
 A forerunner to present-day simulation is the Monte Carlo technique, a modeling scheme that
estimates stochastic or deterministic parameters based on random sampling.

Examples of Monte Carlo applications include evaluation of multiple integrals, estimation of the
constant π (= 3.14159), and matrix inversion.

This section uses an example to demonstrate the Monte Carlo technique.

The objective of the example is to emphasize the statistical nature of the simulation experiment.

6
Example 1
 We will use Monte Carlo sampling to estimate the area of the following circle:

The radius of the circle is r = 5 cm, and its center is (x, y) = (1, 2).

Solution
 The procedure for estimating the area requires enclosing the circle snugly in a square whose side
equals the diameter of the circle, as shown in Figure 1.

The corner points are determined from the geometry of the square.

7
Figure 1: Monte Carlo estimation of the area of a circle.

8
 The estimation of the area of the circle is based on assumption that all the points in the square are
equally likely to occur.

If m out of n sampled points fall within the circle, then:

 To ensure that all the points in the square are equally probable, the coordinates x and y of a point
in the square are represented by the following uniform distributions:

9
 The determination of a sample (x, y) is based on the use of independent 0 − 1 random numbers.

Table 1 lists a sample of such numbers which we will use in the examples in this chapter.

For the purpose of general simulation, special arithmetic operations are used to generate (pseudo)
0 − 1 random numbers as described in another section later.

Table 1: A short list of 0 – 1 random numbers.

10
 A pair of 0 − 1 random numbers, R1 and R2, can be used to generate a random point (x, y) in the
square by using the following formulas:

 To demonstrate the application of the procedure, consider R1 = 0.0589 and R2 = 0.6733.

This point falls inside the circle because:

 The procedure is repeated n times, keeping track of the number of points m that fall within the
circle.

The estimate of the area is then computed as 100 .

11
 The estimate of area of the circle improves as the sample size increases.

Average of 10 replications for each sample size n provides an improved estimate of the area than
any one replication.

Accuracy of the mean value of all 10 replications increases with the increase in sample size n,
which is evident by decline in the value of the standard deviation.

Higher values of n (sample size) and N (number of replications) provide more reliable simulation
results.

The drawback however is that higher values of n and N imply higher costs for conducting the
simulation experiment.

12

You might also like