Modeling and simulation Past

paper (spring 2023)

Short Question:
Q no 1: describe the process of calibration and
validation in detail?
Ans: calibration:
Calibration is the iterative process of comparing the model with real
system, revising the model if necessary, comparing again, until a
model is accepted (validated).

Validation:
Validation is a process of comparing the model and its behavior to the
real system and its behavior.

Q no 2: draw and describe different types of GPSS

blocks that are used to deal with queues?
Ans: GPSS (General Purpose Simulation System) is a highly structured
and special purpose simulation language based on process interaction
approach and oriented toward queuing systems.
- The system being simulated is described by the block diagram using
various GPSS blocks.
- Each block represents events, delays or other actions that affect the
transaction flow.

The two main blocks that are used with queues are QUEUE and DEPART.

QUEUE BLOCK:
This block will instruct GPSS to start gathering queuing statistics on the
queue named in its attribute value. SYNTAX: QUEUE A

ATTRIBUTES: A = name of queue (for example: garage)

DEPART BLOCK:

This block instructs GPSS that a transaction is leaving the queue named in
it‟s attribute value. This is necessary in order to compile the statistics on
the queue.

SYNTAX: DEPART A

ATTRIBUTES: A = name of the queue (for example: checkout)

Some other blocks that are used with above blocks are:

GENERATE BLOCK:

This block will produce a flow of transactions with inter-arrival times

determined by the attribute values. The distribution of inter-arrival times
follows a uniform probability distribution.

SYNTAX: GENERATE A,B,C,D,E

ATTRIBUTES: A = average value of uniform distribution, B = half-width of

uniform distribution, C = time delay before first transaction is generated, D
= maximum number of transactions generated, E = priority allocated to
transactions

ADVANCE BLOCK:

This block represents the servicing of a transaction. The servicing times

follow a uniform probability distribution.
SYNTAX: ADVANCE A,B

ATTRIBUTES: A = average value of uniform distribution, B = half-width of

uniform distribution

TERMINATE BLOCK:

This block destroys any transaction entering it and removes it from

computer memory. Each time a transaction enters this block it decrements
a counter by an amount equal to its attribute value. The counter is set by
the user upon starting the simulation.

SYNTAX: TERMINATE A

ATTRIBUTES: A = decrements simulation counter by this amount

etc.

Q no 3: differentiate between fixed time step and event

to event model with the help of suitable example?
Ans:
Fixed time-step model:
In this the timer simulated by the computer is updated at a fixed time interval.
The system is checked to see if any event has taken place during that interval.
All the events which take place during the time interval are considered to have
occurred simultaneously at the end of the interval.

Event-to-event model:
It is also known as the next-event model. In this the computer advances the
time to the occurrence of the next event. So it shifts from one event to the next
event and the system state does not change in between. A track of the current
time is kept when something interesting happens to the system.

Q no 4: define and describe different types of elements

and components of the system?
Ans: A group of components which are interconnected and
interact to fulfill certain objective is called a system. A system
mainly consists of 4 elements: input, process, output and
feedback.

Q no 5: discuss the merits and demerits of system

simulation?
Ans:

Merits: Demerits:
Risk-Free Experimentation: Predictive Power:
 Simulations allow testing of  Simulation can be used for
scenarios and making decisions forecasting future behavior, which
without affecting the real system. is valuable in planning, resource
This is particularly important in allocation, and risk management.
industries like aviation, healthcare, For example, it can predict how
or nuclear power, where real- changes in market conditions
world experimentation could be might affect a business.
dangerous or expensive.

Cost-Efficiency: Computational Costs:

 Building and testing a physical  Complex simulations, especially
prototype can be costly. those involving large systems or
Simulations can reduce the need detailed models, can require
for physical prototypes, thereby significant computational
saving time and money. For resources and time. High-fidelity
example, in manufacturing, a simulations may be
simulation can help optimize the
 production process before actual computationally expensive and
implementation. time-consuming.

Complexity Handling: Data Requirements:

 Simulations can model complex  Simulations often require large
systems with multiple interacting amounts of accurate data to build
variables, which might be difficult and validate the model. Inaccurate
or impossible to study analytically. or incomplete data can lead to
This is useful in fields like climate unreliable results, undermining the
modeling, where numerous factors simulation's utility.
influence the outcome.

Flexibility: Overreliance on Simulation:

 A simulation model can be easily
modified to test different There's a risk that decision-makers
scenarios, configurations, or might rely too heavily on simulation
policies. This flexibility allows for results, ignoring the limitations and
exploring a wide range of
possibilities and understanding the
uncertainties inherent in the model.
impact of various factors on the This can lead to poor decisions if the
system. simulation is not interpreted carefully.

Enhanced Understanding: Complexity in Model Development:

 Simulation helps in  Developing a realistic simulation

understanding how a system model can be a complex task
works by visualizing its behavior requiring specialized knowledge
and skills. This complexity can
under different conditions. This lead to increased development
can lead to insights that might time and costs.
not be apparent through
theoretical analysis alone.

Predictive Power: Interpretation Challenges:

 Simulation can be used for  The results of a simulation can be
forecasting future behavior, which difficult to interpret, especially
is valuable in planning, resource when dealing with complex
allocation, and risk management. systems. Misinterpretation of
 For example, it can predict how results can lead to incorrect
changes in market conditions conclusions and decisions.
might affect a business.

Q no 6: explain markov’s chain with a suitable

example?
Ans: A Markov chain is a stochastic model that uses
mathematics to predict the probability of a sequence of
events occurring based on the most recent event.
Example:
A common example of a Markov chain in action is the
way Google predicts the next word in your sentence
based on your previous entry within Gmail.

LONG QUESTIONS:
Q no 1: differentiate between chi-square test and KS
test for uniformity. Use KS test to check for the
uniformity for the input set of random number given
below. 0.54,0.73,0.98,0.11,0.68,0.45. Assume level of
significance to be Zalpha -0.005=>0.565?

Chi-Square Test vs. KS Test for Uniformity

Chi-Square Test:
  Purpose: The Chi-Square test is commonly used to test if a sample
data fits a particular distribution, including testing for uniformity.
 Approach: It compares the observed frequencies of data within
different intervals to the expected frequencies if the data were
uniformly distributed.
 Assumptions: Requires a sufficiently large sample size to ensure
that the expected frequency in each interval is reasonably high.
 Output: It gives a statistic that is compared against a chi-square
distribution to determine the p-value.
Kolmogorov-Smirnov (KS) Test:
 Purpose: The KS test is a non-parametric test used to determine if a
sample comes from a specific distribution, such as a uniform
distribution.
 Approach: It compares the cumulative distribution function (CDF) of
the sample data to the CDF of the uniform distribution.
 Assumptions: It does not require a large sample size or binning of
data into intervals, making it more flexible.
 Output: The KS statistic is the maximum distance between the
empirical CDF of the sample and the CDF of the uniform distribution.
This statistic is compared to a critical value based on the significance
level to decide whether to reject the null hypothesis of uniformity.
(the Q will be solve chi square method which we use in stat )

Q no 2: define congestion in a queuing system. Describe

different type of component and characteristics of a
queuing system?
A congestion system is system in which there is a
demand for resources for a system, and when the
resources become unavailable, those requesting the
resources wait for them to become available.
Characteristics/Elements of Queueing System:
1. Calling Population
The population of potential customers those require
service from system is called calling population. It may be
finite or infinite. System having large calling population is
usually considered as infinite. For e.g. customers at
banks, restaurant. And System having less and countable
population is usually considered as finite. For e.g. a
certain number of machines to be repaired by a service
man.
In finite population model, arrival rate depends on the
number of customers being served and waiting. But in
infinite population model, arrival rate is not affected by
the number of customer being served and waiting.

2. Arrival Process
The arrival process for infinite-population models is
usually characterized in terms of interarrival times of
successive customers. Arrivals may occur at scheduled
times or at random times. When at random times, the
inter arrival times are usually characterized by a
probability distribution and most important model for
random arrival is the poisson process. In schedule arrival
interarrival time of customers are constant.
3. Service Process
Service process can be measured by the number of
customers served per some unit of time or the time
taken to complete the service. Once entities have
entered to the system they must be served. The service
can be provided in single or batch. if it is batch, as in the
case of arrival the batch size can be fixed or
random. Service time may be of constant duration or of
random duration.
4. Queueing Discipline and Queueing Behaviour
Queue discipline refers to the rule that a server uses to
choose the next customer from the queue when the
server completes the service of the current
customer. Common queue disciplines include first-in-
first-out (FIFO); last-in-first-out (LIFO); service in random
order (SIRO); shortest processing time first (SPT); and
service according to priority (PR).
Queue behavior refers to the actions of customers while
in a queue waiting for service to begin.
5. Number of Servers:
Servers represent the entity that provides service to the
customer. A system may consist of single server or
multiple servers.
- A system with multiple servers is able to provide
parallel services to the customers.
Q no 3: describe differente type of mathematical
simulation models. Develop a mathematical model for
any dynamic system?
Ans: Types of Mathematical Simulation Models
Mathematical simulation models can be broadly classified into various
types based on the characteristics and behavior of the systems they
represent:

1. Deterministic Models:
o Definition: These models operate under the assumption that
the system's behavior is fully predictable if the initial
conditions and inputs are known. There is no randomness
involved.
o Examples:
 Linear Programming Models: Used for optimizing
resources in systems with constraints.
 Ordinary Differential Equations (ODEs): Used to
model systems like population growth or chemical
reactions where changes are continuous over time.
2. Stochastic Models:
 o Definition: Unlike deterministic models, stochastic models
incorporate randomness or probabilistic elements to account
for the uncertainty and variability in the system.
o Examples:
 Monte Carlo Simulations: Used in finance to model
the probability of different outcomes in processes that
cannot easily be predicted due to the intervention of
random variables.
 Markov Chains: Used in queueing theory and other
processes where the next state depends only on the
current state and not on the sequence of events that
preceded it.
3. Static Models:
o Definition: Static models represent systems at a particular
point in time or assume that time does not play a role in the
simulation.
o Examples:
 Linear Regression Models: Used for predicting
outcomes without considering time as a factor.
 Inventory Models: Used to determine optimal order
quantities at a fixed point in time.
4. Dynamic Models:
o Definition: Dynamic models consider how the system
evolves over time. These models are particularly useful in
studying processes that change over time.
o Examples:
 System Dynamics Models: Used in supply chain
management to understand the impact of delays and
feedback loops over time.
 Epidemiological Models: Used to predict the spread of
diseases over time.
5. Discrete Models:
o Definition: Discrete models simulate systems where changes
occur at specific points in time, often in steps or events.
o Examples:
  Discrete Event Simulation (DES): Used in
manufacturing to model processes where parts move
through different stages in a production line.
 Agent-Based Models: Used in social sciences to model
interactions of individual agents (people, animals, etc.)
within a system.
6. Continuous Models:
o Definition: Continuous models represent systems where
changes occur continuously over time, often described by
differential equations.
o Examples:
 Fluid Dynamics Models: Used in engineering to
simulate the flow of liquids and gases.
 Heat Transfer Models: Used to simulate how heat
spreads through materials over time.

Developing a Mathematical Model for a Dynamic System

Let's develop a mathematical model for a simple dynamic system: A

Spring-Mass-Damper System.

System Description:

 Components: A mass mmm is attached to a spring with stiffness

kkk and a damper with damping coefficient ccc. The mass is
subject to an external force F(t)F(t)F(t).
 Objective: To describe the motion of the mass over time.

Assumptions:

1. The system follows Hooke's law for the spring force: Fs=−kxF_s =
-kxFs=−kx.
2. The damping force is proportional to the velocity: Fd=−cvF_d = -
cvFd=−cv.
3. Newton's second law applies: F=ma=md2xdt2F = ma = m
\frac{d^2x}{dt^2}F=ma=mdt2d2x.
Mathematical Model:

The equation of motion for the system can be derived using Newton's
second law:

md2x(t)dt2+cdx(t)dt+kx(t)=F(t)m \frac{d^2x(t)}{dt^2} + c
\frac{dx(t)}{dt} + kx(t) = F(t)mdt2d2x(t)+cdtdx(t)+kx(t)=F(t)

Where:

 x(t)x(t)x(t) is the displacement of the mass from the equilibrium

position at time ttt.
 mmm is the mass of the object.
 ccc is the damping coefficient.
 kkk is the spring constant.
 F(t)F(t)F(t) is the external force applied to the system.

Steps to Solve the Model:

1. Identify Initial Conditions: You need to know the initial

displacement x(0)x(0)x(0) and initial velocity
dx(0)dt\frac{dx(0)}{dt}dtdx(0).
2. Solve the Differential Equation: Depending on the nature of
F(t)F(t)F(t), you can solve this second-order differential equation
analytically (if F(t)F(t)F(t) is simple) or numerically (for more
complex F(t)F(t)F(t)).
3. Analyze the System Behavior:
o Underdamped System: If c2<4mkc^2 < 4mkc2<4mk, the
system exhibits oscillatory behavior.
o Critically Damped System: If c2=4mkc^2 = 4mkc2=4mk,
the system returns to equilibrium as quickly as possible
without oscillating.
o Overdamped System: If c2>4mkc^2 > 4mkc2>4mk, the
system returns to equilibrium slowly without oscillating.

Example Application:
  In Engineering: Such a model can be used to design shock
absorbers in vehicles, where you want to minimize oscillations and
return to equilibrium quickly after a disturbance.

This example demonstrates how a dynamic mathematical model can be

constructed to describe the behavior of a physical system over time.