Dr. K. M.

Salah Uddin
Associate Professor, Dept. of MIS
University of Dhaka
Outlines of course Plan
 1. Introduction to simulation
 2. Probability concepts in Simulation
 3. Random Number generators
 4. Generating Random Variants
 5. Simulation of Queuing systems
 6. Output analysis for a Single System
 7. Statistical Techniques for Comparing Alternative
Systems
 8. Simulation of Manufacturing and Material Handling
Systems
Basic Terminology
In most simulation studies, we are concerned with
the simulation of some system.
Thus, in order to model a system, we must
understand the concept of a system.
 Definition: A system is a collection of entities,
i.e., people or machines, that act and interact
toward the accomplishment of some logical end.
 Systems generally tend to be dynamic – their status
changes over time. To describe this status, we use
the concept of the state of a system.
Example of a System
For example, one wants to study a bank to determine the
number of tellers needed to provide adequate service for
customers who want just to cash a check or make a
savings deposit, the system can be defined to be that
portion of the bank consisting of the tellers and customers
waiting in line or being served.
Component of a System
Systems

Entity Activity
Attribute
An entity is an An activity represents a time
An attribute is a
object of period of specified length
property of an entity
interest in the
system
Simulation
 What is simulation:

The process of designing a mathematical or logical

model of a real-system and then conducting
computer-based experiments with the model to
describe, explain, and predict the behavior of the real
system.
Schematic of Simulation Study
Simulation and System Simulation
 What is simulation

Simulation is the actual running of the model system to gain

insight into its performance.

System simulation is nothing but an experiment with

the help of computers, without performing actual
experiment. It saves lot of money which is required, if
we actually perform experiments with the real system.
Simulation
 Why use simulation

Simulation is used to better understand the expected

performance of the real system and to test the
effectiveness of the system design.
When Simulation is the appropriate tool
 Informational, organizational, and environmental changes can
be simulated, and the effect of these alterations can be observed
on models behavior.
 Simulation can be used to experiment with new designs or
policies before implementation, so as to prepare what might
happen.
 Simulation can be used to verify analytic solutions.

 The modern system is so complex that its internal interactions

can be treated only through simulation.
When simulation is not appropriate
 Simulation shouldn’t be used when the problem can be
solved by common sense.
 The simulation should not be used if the problem can be
solved analytically.
 The simulation should not be used if it is easier to perform
direct experiments
 Not to use simulation if the simulation cost exceed the
savings.
 Simulation takes data, sometimes lot of data. If no data is
available, not event estimates, simulation is not advisable.
 If system behavior is too complex or can't be defined,
simulation is not appropriate.
Advantages and disadvantage of simulation
Advantage:
1. New policies, operating procedures, decision rules,
information flows, organizational procedures and so on
can be explored without disrupting ongoing operations
of the real systems
2. New hardware designs, physical layouts, transportation
systems, and so can be tested without committing
resources for their acquisitions.
3. Hypothesis about how and why certain phenomena
occur can be tested for feasibility.
4. Time can be compressed or expanded to allow
Advantages and disadvantage of simulation

Advantage:
5. Insight can be obtained about the interaction of variables.
6. Insight can be obtained about the importance of variables to
the performance of the system.
7. A simulation study can help in understanding how the system
operates than how the individual think the system operates.
8. “what if” questions can be answered. This is particularly
useful in the design of new systems.
Disadvantages of Simulation

 Model building requires special training and

experience. It is an art that is learned over time and
through experience
 Simulation results can be difficult to interpret.
 Simulation can be time consuming and expensive.
Static Simulation vs. Dynamic Simulation

 There are two types of simulation models, static and

dynamic.

 Definition: A static simulation model is a

representation of a system at a particular point in time.

Example: We usually refer to a static simulation as a

Monte Carlo simulation.
Static Simulation vs. Dynamic Simulation
 Definition: A dynamic simulation is a representation of a
system as it evolves over time.

 Within these two classifications, a simulation may be

deterministic or stochastic.

 A deterministic simulation model is one that contains no

random variables; a stochastic simulation model contains
one or more random variables.
Model of a system

 A model is defined as a representation of a system for

the purpose of the studying the system. For the most
studies, it is only necessary to consider those aspects of
the system that affect the problem under investigation.
These aspects are represented in a model of the
system; the model, by definition, is simplification of
the system.
STEPS IN A SIMULATION STUDY
STEPS IN A SIMULATION STUDY
Types of Models
Models

Physical Mathematical

Static Dynamic Static Dynamic

Numerical
Analytical
Numerical

System simulation
Distributed Lag Models
Models that have the properties of changing only at fixed
interval of time, and of basing current values of the
variables on other current values and values that
occurred in the previous intervals, are called distributed
lag models

Any variable that appears in the form of its current value

and one or more previous intervals is said to be a lagged
variables
Cobweb Models
Distributed lag model can be constructed from the static
market model. This is related supply, S, and demand, D, to
market price, P.
The static marketing model in distributed lag form is as
follows:

D = a – bP
S = c + dP-1 1.5
D=S
Monte Carlo simulation
 Simulation can also be defined as a technique of
performing sampling experiments on the model of the
system. This is called stochastic simulation and is a part of
simulation techniques. Because sampling from a particular
probability distribution involves the use of random
numbers, stochastic simulation is sometimes called Monte
Carlo Simulation. Historically, Monte Carlo method is
considered to be a technique, using random or pseudo
random numbers.
Monte Carlo simulation Using
Empirical data
The following steps simulate an assembly activity:
1. Collect empirical data on assembly times
2. Develop probability distribution and cumulative
probability distribution
3. Assign an interval of random numbers to each class
of the distribution.
4. Using random numbers (RNs), derive simulated
assembly times
5. Interpret the results.
Problems using Monte Carlo
Q. A process planner is working on plans for producing a
new detergent. She wishes to simulate raw-material
demand in order to plan for adequate materials-handling
and storage facilities. On the basis of usage for a similar
product produced previously, she was developed a
frequency distribution of demand in tons per day for a 2-
month period. Use this data to simulate the raw material
usage requirements for 7 periods (days)
Demand, X 10 11 12 13 14 15
(tons/day)
Frequency 6 18 15 12 6 3
(no. days)
(1) Data are given in frequencies
(2) To formulate a probability distribution, divide each
frequency by the total (60). Then formulate a cumulative
probability distribution by successively summing the
probability values is shown in the table

Demand Frequency Probability Cumulative

(tons/day) (no. days) P(X) probability
10 6 0.1 0.1
11 18 0.3 0.4
12 15 0.25 0.65
13 12 0.20 0.85
14 6 0.10 0.95
15 3 0.05 1.00
(3) Next , assign random number intervals so that the
number of values available to each class corresponds
with the probability. Using 100 two-digit numbers (00 to
90) we assign 10 percent (00 to 09 ) to the first class, 30
percent (10 to 39) to the second class, and so on.

Demand Probability Corresponding

(tons/day) Random numbers
10 0.1 00-09
11 0.3 10-39
12 0.25 40-64
13 0.20 65-84
14 0.10 85-94
15 0.05 95-99
(4) We obtain random numbers (RN) from a random number
table, Assume the RNs are:

27 13 80 10 54 60 49
The first RN, 27 , falls into second class of the distribution and
corresponds to a demand of 11 tons/day.

Random number 27 13 80 10 54 60 49
Simulated demand 11 11 13 11 12 12 12
Difference between Monte Carlo
method and simulation
 In the Monte Carlo method, time does not play as substantial
role, as it does in stochastic simulation.
 2. The observations in the Monte Carlo method, as a rule, are
independent. In simulation, however, we experiment with the
model over time so, as a rule, the observations are serially
correlated.
 3. In the Monte Carlo method, it is possible to express the
response as a rather simple function of the stochastic input
variants. In simulation the response is usually a very
complicated one and can be expressed explicitly only by the
computer program itself.
Questions
1. What is the difference between static and dynamic
models?
2. Give an example of a dynamic mathematical model.
3. (a) What are distributed lagged models?
(b) If demand and supply of a product obey following
equations. D = a + bP, S = c – dP and D = S
Here a, b, c, and d are given numbers, convert this model to
distributed lagged model.
1. Generator service co. (GSCO) has ongoing contracts
with several electric utilities wherein GSCO agrees to
provide technicians whenever a customer has a
generator shutdown and needs technical assistance.
Problem No. 1: The GSCO operations manager is concerned
with maintain enough technicians to give the needed
service while staying within a limited budget for
personnel. He has collected data on the number of
service request per day over a 200-day period as shown
in Table. (a) Simulate the service requests per day for a
week period by using random numbers applied to a
cumulative distribution. (b) compare the simulated
values with the historical average.

Number of 0 1 2 3 4 5 6
request
Frequency 30 40 60 44 20 6 0
2. Empirical data collected on time required to weld a
transformer bracket were recorded to the nearest quarter
minute, as shown in Table 1

Weld time (min) Number of observations

<0.25 0
0.25<0.75 24
0.75<1.25 42
1.25<1.75 72
1.75<2.25 38
2.25<2.75 14
2.75<3.25 10
a. Formulate a cumulative distribution in percentage
terms
b. Graph the frequency and cumulative distributions.
c. A simulation is to be conducted using random
numbers. What simulated weld times (to the nearest
0.25 minute) would result from the random numbers
25, 90 and 59?
Questions (cont..)
1. Find the growth in national consumption for five years
using the model (distributed lagged model). Assume the
initial income Y-1 is 80 and take the government
expenditure in the 5 years as follows:

Year G
1 20
2 25
3 30
4 35
5 40
THE END