Simulation is a representation of real-life situations.
It is a
method in which a replica of a real-world process or system is
developed over a period of time. The simulated model acts in the
same manner as the selected physical or abstract process or
system behaves in reality.
For example, aircraft pilots are given training through simulation
models, because training with real aircrafts can incur huge costs
as well as involve various risks.
Similarly, in the education sector, teachers are trained in the
simulated models of students (a group of individuals who imitate
as students). This avoids the risk of spoiling the future of
students if the teacher is not able to teach properly. Therefore,
simulation models have proved to be useful for training purposes.
Category of Simulation
Simulation used for training purposes is divided into three
categories, which are as follows:
Live simulation
It refers to a simulation in which equipment is used to imitate a real
system. For example, testing the battery of a car with the help of an
electrical tester.
Virtual simulation
It refers to a simulation in which real people operate on
simulated systems. For example, a pilot flying a simulated jet.
Constructive simulation
It refers to a simulation in which simulated people operate on
simulated systems. For example, in war games, the players
supervise soldiers and equipment, which move around the board.
From the discussion so far, it can be understood that simulation
models are the most appropriate for uncertain situations. By
using simulation, an analyst can introduce constants and
variables related to a problem, develop possible courses of action
and establish measures of effectiveness. Simulation helps an
analyst in determining how a system would respond under given
conditions. However, it does not generate optimal solutions.
Therefore, an analyst has to generate solutions for what he/she
wants to [Link] can introduce constants and variables
related to a problem, develop possible courses of action and
establish measures of effectiveness. Simulation helps an analyst
in determining how a system would respond under given
conditions. However, it does not generate optimal solutions.
Therefore, an analyst has to generate solutions for what he/she
wants to test.
Types of Simulation
Simulation models are used when a real system is not accessible,
can be dangerous to engage or has not been built. As discussed
earlier, these models are used in various fields, such as safety
engineering, testing, training and education.
Simulation models can be classified on the basis of the changes
produced in a system and uncertainty in the processes, events
and controlling parameters of the system. The two classifications
of simulation models are mentioned below:
On the basis of the changes produced in a system: This
classification includes the following:
Static simulation: It refers to a simulation in which a system
does not show any change with time. Static simulation
represents a system at a particular point in time.
Dynamic simulation: It refers to a simulation in which a system
changes and evolves with time. The main objective of dynamic
simulation is to estimate the future performance of a system.
On the basis of uncertainty in the processes, events and
controlling parameters of a system: This classification includes
the following:
Deterministic simulation: It refers to a simulation in which
parameters are uncertain and are expressed by using a single
value. The result of deterministic simulation is in the form of a
qualified statement.
Probabilistic simulation: It refers to a simulation in which
uncertainties are defined very clearly by specifying inputs as
probability distribution. The results of probabilistic simulation are
a quantified statement.
These aforementioned simulations are used for solving different
busi- ness problems and taking decisions accordingly.
Steps of Simulation Process
The steps involved in the process of simulation are explained as
follows:
Identifying a problem
Refers to the first step of the simulation process in which the
problem is clearly defined and the intended solutions to be
achieved are identified. If the problem is defined clearly, it will
lead to the development of an appropriate model and provide a
basis for the evaluation of simulation results.
Developing a model
Involves formulating a model by considering the problem that is
identified in the first step. A simulation model can be a physical
or mathematical model, idea or combination.
Collecting data
Involves gathering data for the development and evaluation of
the simulation model. The collected data helps in analysing the
problem and reaching at an appropriate solution.
Running the simulation model
Refers to one of the most important steps of the simulation
process. After developing the simulation model and collecting the
relevant data, the next step is to run it. If the model is
deterministic (in which all parameters are known and constant),
a single run would be sufficient.
On the contrary, if the model is stochastic (in which parameters
are subject to random variation), a number of runs would be
required to evaluate the performance of the model.
Interpreting the results
Involves analysing the results of the runs. If the model portrays
real-life situations, no rectifications will be required. However, if
the model is not closer to reality, a number of runs will be
required.
Importance and Limitations of
Simulation in Management
Simulation is a useful technique, especially in situations where
mathematical simplification is not possible. The following are
some of the advantages of simulation:
Helps an analyst in observing and studying a real-world
environment or process that cannot be studied in its actual form
due to some reasons, such as high costs and greater risks.
Enables an analyst to study processes that are long-term and
require more time to display results. For example, for studying
trends in the world population in the long run, an analyst cannot
wait for the required number of years to see results.
Helps in studying the systems that are too disrupting in nature.
However, in the case of complex business problems, a simulation
may suffer from various limitations. Some of the limitations of the
simulation are as follows:
Time consuming: It refers to one of the major drawbacks of
simulation. In certain analytical processes, simulation has proved
to be time-consuming.
Expensive: It implies that a better and more advanced
simulation model would be more expensive. Simulation often
requires significant use of computers, which increases the cost of
electricity. This incurs high costs to organisations.
No optimum results: It refers to another major drawback of
simulation. Since a simulation model generally deals with
uncertainties, optimum results cannot be produced. A simulation
model can provide a way of evaluating solutions but does not
generate any solutions.
Complicated process: Simulation involves a long and
complicated process to develop a model. Such a model is unique
and developed for obtaining the solution of a specific problem.
Monte Carlo Simulation Using
Random Numbers
Monte Carlo simulation is a problem-solving technique that is
used to represent the probability of certain outcomes by running
a number of trial runs (called simulations) and using random
variables.
This technique was introduced by John von Neumann, Stanislaw
Ulam and Nicholas Metropolis while they were working on the
atomic bomb in the 1940s. These scientists named this technique
after the city in Monaco, which is famous for casinos and games
of chance. The Monte
The steps involved in the Monte Carlo simulation technique are
explained as follows:
Making probability distribution for
variables
This step involves estimating a probability distribution for the
given vari- able based on the past experience. The distribution
can be based on Poisson, binomial or normal distribution.
Building a cumulative probability
distribution for each variable
It is one of the most important steps of the Monte Carlo
technique. The cumulative distribution of an experiment
represents all possible values and probabilities of outcomes.
Setting random number intervals
In this step, after determining the cumulative distribution for each
variable, a set of numbers to represent each possible outcome is
assigned. This set of numbers is known as random number
intervals.
Generating random numbers
A random number is a series of two-digit or three-digit numbers
(for example, 00, 01, …, 8 8). A random number can be generated
in many ways. For example, if a problem is very complex and
involves a number of simulation trials, random numbers can be
generated using computers.
Simulating the experiment through random
sampling
It involves repeating the process till the required number of
simulation runs has been generated.
Application of the Monte Carlo
Technique
The following are some areas that have an extensive application
of the Monte Carlo technique:
Physical science: It is a field in which the Monte Carlo
technique has proved to be very effective. The Monte Carlo
technique is used in computational physics, physical chemistry
and molecular modelling. Apart from this, it is also used for
designing detectors and analysing their behaviour.
Finance: It is another area where the Monte Carlo technique
has been used extensively. The technique is used for
evaluating investments in projects and analysing financial
derivatives. In such cases, stochastic or probabilistic
simulation models are used.
Telecommunication: It is an area where the Monte Carlo
technique is used to establish an effective wireless network
and evaluate its performance. If the desired performance is
not achieved, the network design is moving through the
optimisation process.
Games: Monte Carlo technique is used to design games
related to artificial intelligence.