DEFINITIONS OF SYSTEMS AND MODELS

Examples-Mathematical models-Definition-Principles of Mathematical Modelling-Deterministic

and Stochastic Models-Mechanistic and empirical modeling.

1. Discuss briefly about simulation.

Simulation is the process of imitating the operations of a real life system to understand its
behavior and then employing the results in the system. So, in a way we are generating an
artificial history of a system to draw inferences of the system. The imitation is called a model.
We can also use simulation to study systems in design stage, systems not yet operative. So,
either we are imitating an imaginary system to find the means of implementing it or to analyze
an existing system to predict the effect of changes in it.
Simulation enables study of internal interaction of a complex system. Effect of changes
made on various subsystems can be studied separately. By performing sensitivity analysis, we
can gain insight into important components of a system. We can also verify solutions using
simulation.

2. Discuss about advantages and disadvantages of simulation.

Simulation has numerous advantages. Primarily it is intuitively appealing to a client as it
mimics the system. New policies, rules, procedures can be explored. Information flows and
organization flows can be identified. New designs, layouts can be created. Hypotheses about
certain phenomenon can be identified. Exact operating nature of a system can be discovered.
Simulation has some disadvantages. Model building is a complex process, which needs
experts. The outputs may be difficult to interpret, as most of them are random variables. The
entire process can be time consuming and expensive. Due to its appeal, we tend to use it when
analytical solutions are available.

3. Discuss about systems, in simulation.

A system is a group of interrelated objects, which work towards a common goal. So, in
simulation we first need to identify the system. The objects outside the system, which could
affect activities inside the system, are called its environment. So, the line, which separates a
system from its environment, is called its boundary.
An entity of a system is any object of interest in the system. The characters of the entities
are called their attributes. Any time period of specified length is called an activity. The
collection of entities, which describe the system at any time, is called the system state. Any
instantaneous occurrence that changes the system state is called an activity. Activities and
events occurring within the system are termed as endogenous and those occurring outside it
are termed as exogenous.
Systems can be broadly classified into the following types.
(i) Discrete: Systems, whose state changes only in discrete points of time.
(ii) Continuous: Systems, whose state can change at any time.
4. Discuss about the areas of applications of simulation.
Simulation is applicable in almost all imaginable fields, of course with the restriction for
a proper requirement and a decent approach in applying it. From the history, it is applied in
manufacturing systems, public systems of many disciplines, Transportation systems,
Construction systems etc.
In the manufacturing systems, it is applied in material handling areas, inventory areas,
assembly sections, scheduling areas, product-mix decisions etc.
In the public systems, we can use it in Health care, Military, waste management, power
plant areas, oil spilling modeling etc.
In construction area, we apply it for applications in earthmoving, strip-mining, cable-
stayed bridges, strengthening the design, advanced project planning paradigm etc.
In transportation areas, we apply it for cargo transfer, personnel launch systems,
container operations, toll plaza operations etc.
We can also use it in restaurants, entertainments, Food processing, computer system
performances etc.

7. Discuss about system simulation techniques:

The term simulation is being used by different people and so sometimes its meaning is
not as clear as we would like it to be. When we try to clear that, we see it as a methodology to
solve problems for which either an analytical solution is not possible or difficult to obtain.
Secondly, it is different from numerical methods which, like simulation , solve a problem step
by step with increasing values of time and both normally yield specific solutions and not a
general one as in an analytical method. The way in which simulation is different from numerical
methods is that, it is based on observing the system under consideration and based on the
observation alone getting the solution whereas a numerical method does not normally observe
a system to get the solution. That is why, to avoid all the confusion, people started using the
term ‘system simulation ‘ as the technique of solving problems by the observation of the
performance over time of a dynamic model of a system.

8. Principles of Mathematical modeling

To gain an understanding of the processes involved in mathematical modeling, consider the
two worlds depicted in Figure below. Suppose we want to understand some behavior or
phenomenon in the real world. We may wish to make predictions about that behavior in the
future and analyze the effects that various situations have on it.

A system is an assemblage of objects joined in some regular interaction or interdependence.

The modeler is interested in understanding how a particular system works, what causes
changes in the system, and how sensitive the system is to certain changes. He or she is also
interested in predicting what changes might occur and when they occur. How might such
information be obtained? For instance, suppose the goal is to draw conclusions about an
observed phenomenon in the real world. One procedure would be to conduct some real-world
behavior trials or
experiments and observe their effect on the real-world behavior. This is depicted on the left
side of Figure below
In summary, we have the following rough modeling procedure:
1. Through observation, identify the primary factors involved in the real-world behavior,
possibly making simplifications.
2. Conjecture tentative relationships among the factors.
3. Apply mathematical analysis to the resultant model.
4. Interpret mathematical conclusions in terms of the real-world problem.

Given some real-world system, we gather sufficient data to formulate a model. Next we analyze
the model and reach mathematical conclusions about it. Then we interpret the model and
make predictions or offer explanations. Finally, we test our conclusions about the real-world
system against new observations and data. We may then _nd we need to go back and refine
the model to improve its predictive or descriptive capabilities. Or perhaps we will discover that
the model really does not _t the real world accurately, so we must formulate a new model.

we define a mathematical model as a mathematical construct designed to study a particular

real-world system or phenomenon. We include graphical, symbolic, simulation, and
experimental constructs. Mathematical models can be differentiated further.
The distinction between the various model types as depicted in Figure above is made solely for
ease of discussion. For example, the distinction between experiments and simulations is based
on whether the observations are obtained directly (experiments) or indirectly (simulations). In
practical models this distinction is not nearly so sharp; one master model may employ several
models as sub-models, including selections from existing models, simulations, and experiments.
Nevertheless, it is informative to contrast these types of models and compare their various
capabilities for portraying the real world. To understand that , consider the following
properties of a model.
Fidelity: The preciseness of a model's representation of reality
Costs: The total cost of the modeling process
Flexibility: The ability to change and control conditions affecting the model as required data are
gathered

So, to construct a model, we follow the steps below

 5. Discuss about models of a system.
Model is a representation of a system for the purpose of studying about the system. In a
model, we inscribe those characters of a system, which interests us. We model a system, partly
because it is not easy to study the system directly or we can’t make any changes in an active
system or partly due to the non-existence of the system in the real life. We study the models
thoroughly to know about the system and implement those facts in the system to improve its
efficiency.
Models can be broadly classified as mathematical or physical. Mathematical models use
symbolic notations and mathematical equations to represent a system. Physical models try to
represent the physical structure of a system.
(i) Static: Models which represent systems only in discrete points of time.
(ii) Dynamic: Models which represent systems continuously.
(iii) Deterministic: models, which have all the input variables as deterministic.
(iv) Stochastic: Models, which have at least one input variable as random.
(v) Discrete: Models, which represent discrete systems.
(vi) Continuous: Models, which represent continuous systems.

6. Discuss about steps in a simulation study

1. Problem definition ;
2. Setting of objectives and project plan
3. Model building
4. Data collection
5. Model translation
6. Verification
7. Validation
8. Experimental design
9. Production runs & analysis
10. Decision o no. of runs
11. Documentation & reporting
12. Implementation
Termination Conditions for a Simulation:
How the simulation run is terminated can also have an effect on the accuracy of the data
collected. There are two generally used options for choosing when to stop the simulation of a
performance model:
Option 1
• begin the simulation at time 0
• begin data collection at specified time w _ 0
• complete data collection at specified time w + t
• terminate execution of the simulation at time w + t
• calculate summary statistics based on sample path data collected in the time interval
(w,w + t).
Option 2
• begin the simulation at time 0
• begin data collection when the Mth job completes all service
• complete data collection when the (M + N)th job completes all service
• terminate execution of the simulation when the (M+N)th job completes all service
• calculate summary statistics based on sample path data collected in the time interval
(tM, tM+N), where tj is the time at which the jth job completes all service.

Deterministic and Stochastic models

Classified based on the model output.
Deterministic models ignore random variation and have no components that are inherently
uncertain.
For fixed starting values, a deterministic model will always produce the same result.
A stochastic model will produce different results with same starting values.

Simulations in which one or more input variables are random are referred to as stochastic or
probabilistic simulations. A stochastic simulation produces output that is itself random and
therefore gives only one data point of how the system might behave.
Simulations having no input components that are random are said to be deterministic .
Deterministic simulation models are built the same way as stochastic models except that they
contain no randomness. In a deterministic simulation, all future states are determined once the
input data and initial state have been defined.
As shown in Figure below, deterministic simulations have constant inputs and produce constant
outputs. Stochastic simulations have random inputs and produce random outputs. Inputs might
include activity times, arrival intervals, and routing sequences. Outputs include metrics such as
average flow time, flow rate, and resource utilization. Any output impacted by a random input
variable is going to also be a random variable. That is why the random inputs and random
outputs of Figure below are shown as statistical distributions.
A deterministic simulation will always produce the exact same outcome no matter how many
times it is run. In stochastic simulation, several randomized runs or replications must be made
to get an accurate performance estimate because each run varies statistically. Performance
estimates for stochastic simulations are obtained by calculating the average value of the
performance metric across all of the replications. In contrast, deterministic simulations need to
be run only once to get precise results because the results are always the same.

Static Versus Dynamic Simulation Models

Another dimension along which simulation models can be classified is that of time. If the model
is used to simulate the operation of a system over a period of time, it is dynamic. The baseball
example above uses dynamic simulation. On the other hand, if no time is involved in a model, it
is static. Many gambling situations (e.g., dice, roulette) can be simulated to determine the odds
of winning or losing. Since only the number of bets made, rather than the duration of gambling,
matters, static simulation models are appropriate for them
Monte Carlo Simulation (named after a famous casino town1 in Europe) refers to the type of
simulation in which a static, approximate, and stochastic model is used for a deterministic
system.
Let us now look at an example of Monte Carlo simulation. Consider estimating the value of p by
finding the approximate area of a circle with a unit radius. The first quadrant of the circle is
enclosed by a unit square.
If pairs of uniformly distributed pseudo-random values for the x and y coordinates in [0, 1] are
generated, the probability of the corresponding points falling in the quarter circle is simply the
ratio of the area of the quarter circle to that of the unit square.

Mechanistic and Empirical models

Classified based on the level of understanding on which the model is based.
Mechanistic models uses large amount of theoretical information generally describes what
happens at one level in the hierarchy by considering processes at lower levels.
In empirical models, no account is taken of the mechanism by which changes to the system
occur.
Instead, it is merely noted that they do occur, and the model tries to account quantitatively for
changes associated with different conditions.

Empirical Models:

• A model where the structure is determined by the observed relationship among experimental
data.
• These models can be used to develop relationships for forecasting and describing trends.
• These relationships and trends are not necessarily mechanistically relevant.
Example:
Investigating the relationship of inflowing nutrients in a lake to algal biomass production
(eutrophication).
• Most early (circa 1970) lake eutrophication models based on statistical relationships between
mass
loading of nutrients and average algal biomass (e.g., Vollenweider models with numerous
adaptations by others)
• Applied to PL-566 reservoirs in North Bosque River Watershed

Mechanistic model:
A model that has a structure that explicitly represents an understanding of biological, chemical,
and/or physical
processes.
• These models attempt to quantify phenomena by their underlying casual mechanisms.

Types of Mechanistic models

Steady state:
A mathematical model of fate and transport of waterborne constituents using constant input
parameters to
predict constant values of receiving water quality concentrations (typically under low-flow
conditions)
Dynamic Models:
A mathematical model of fate and transport of waterborne constituents formulated to describe
the physical behavior of a system and its temporal variability

Model Validation
The validation process consists of model calibration andverification.
Calibration - model parameters are adjusted within allowable limits until model output for a
given time period matches measured output within some predetermined measure of model
performance.
Verification - refers to running the calibrated model (i.e., holding adjustment parameters
constant) during a different time period and comparing model output to measured values.
Calibration and verification increases confidence that the model will accurately simulate
watershed conditions for different management scenarios.
This process gives confidence that the model output during management scenarios will
reasonably reflect what the true measured values would be.

1.Discuss about Modeling sound propogation in a medium.

Modeling the above is basically about expressing the wave field at a random point in terms of
the same at some other points. The difficulty is in modeling the scatter of the sound waves in a
medium, usually three dimensional. Sound waves move from a SOURCE to a RECIEVER through
a medium and hence are subjected to refractions, reflections , diffractions at surfaces in the
medium which definitely has an effect on the original signal. So, when the receiver gets the
signal, it usually has reverberations in addition to the original sent signal and it complicates the
modeling aspect a lot.

Modeling sound is similar in some sense to that of light modeling but not without their
significant differences. We need to consider factors like wavelength, speed, coherence, Latency.

These kind of models are frequented in communication domain.

To form a mechanistic model to study this, we have to consider the above mentioned factors
along with the basic laws and rules of physics, transmission which govern the process, try and
get some integral or differential equations , solve them and come up with some solutions for
questions in the domain.
 We can also employ empirical models by collecting data during the transmissions, employ
statistical, machine learning tools to come with some usable and simple solutions .
Linear model:
This is a one-way model to communicate with others. It consists of the sender encoding a
message and channeling it to the receiver in the presence of noise. In this model there is no
feedback or response which may allow for a continuous exchange of information (F.N.S. Palma,
1993). The linear model was first introduced by Shannon & Weaver in 1949. In the linear
communication model, the message travels one direction from the start point to the endpoint.
To predict the acoustic behavior of bio-based materials and structures, models with the power
of predicting the noise absorption coefficient have been investigated. Hamilton's principle was
used to derive the differential equations governing the bending of rectangular sandwich panels.
Many conventional models are available using methods like Transfer matrix methods, Finite
element method.
Due to structural and geometrical complexities, it is extremely hard to define the acoustical
behavior of most sound absorbers based on theoretical models.
Thus, a number of empirical models have been developed for sound absorption behavior. One
of the most used empirical models for absorbent materials has been proposed by Delany and
Bazley (1970). They obtained simple power-law relations by best fitting a large amount of
experimental data for a range of fibrous porous absorbers. The empirical model is a good and
fast approximation to the theoretical calculations because the model needs only one input
parameter, the airflow resistivity.