SIMULATION AND MODELING

WAYS TO STUDY A SYSTEM

 WAYS TO STUDY A SYSTEM

 NB:
 At some point in the lives of some systems,
there is need to study them to try to gain
some insight in to the relationships among
the various components, or to predict
performance under some new conditions
being considered.

 See the figure next slide

Sim and Mod Intro Prepared by R. Kibuku 2

 WAYS TO STUDY A SYSTEM

System

Experiment with Experiment with a

the actual system model of the
system

Physical model
Mathematical
model

Analytical solution Simulation

Sim and Mod Intro Prepared by R. Kibuku 3

 WAYS TO STUDY A SYSTEM

 The following are the possible ways in

which we can study a system.

i. Experiment with the actual system Vs

experiment with model of the system.

ii. Physical models Vs mathematical models.

iii. Analytical solution Vs simulation

Sim and Mod Intro Prepared by R. Kibuku 4
 WAYS TO STUDY A SYSTEM

1. Experiment with the actual system Vs Experiment with model

of the system.
 To study a system, sometimes one can monitor the actual system and alter it,
then allow it to operate under new conditions.

 This is only possible if it is cost-effective to change a real system with minimal

disruptions.

 However, experimenting with the real system is impossible at times because

of the following reasons:
a) The system does not exist (in the case of a proposed system)
b) The system is too expensive to alter or too dangerous to change purely
for experimental purposes.
c) The timescale for real observation may be too long.

 NB: For these reasons, it is necessary to experiment with a model of the

system and study it as a replica of the real system.

Sim and Mod Intro Prepared by R. Kibuku 5

 WAYS TO STUDY A SYSTEM

2. Physical Model Vs Mathematical Model

 Physical Models (Iconic Models)- are
ones that use miniature images of the real
object.
 E.g. 1- a miniature oil tanker in a swimming
pool. Such models are used in engineering or
management systems.
 E.g. 2- a model of a car to teach the concepts
speed, time and distance

Sim and Mod Intro Prepared by R. Kibuku 6

 WAYS TO STUDY A SYSTEM

 A Mathematical Model -on the other hand represents

a system in terms of logical and quantitative
relationships that can be changed to observe how the
model behaves.

 E.g. of mathematical model is s=d/t where s is the speed and

t is the time and d is the distance traveled.

 NB this might provide a valid model in one instance e.g.

space probe to another planet after it has attained the flight
velocity but a very poor model for other purposes e.g. rush-
hour commuting on a congested urban freeway

Sim and Mod Intro Prepared by R. Kibuku 7

 WAYS TO STUDY A SYSTEM

3. Analytical Solution Vs Simulation model

 An Analytical Solution- is one that is obtained from a
mathematical model.
 A mathematical model reveals the underlying structure and
provides insight into cause and effect relation within a system.
 This applies when the model is simple enough. The analytical
solution provides the exact value.

 Nevertheless many real system models are very complex

and analytical solutions are impossible.
 In this case a model is studied by use of simulation. A
mathematical model studied using simulation is called a
SIMULATION MODEL.
 The simulated solution is not exact, it provides estimates of the
exact solution.

Sim and Mod Intro Prepared by R. Kibuku 8

 SIMULATION MODELS

 Given that we have a mathematical model to be

studied by means of simulation (henceforth
referred to as a simulation model) then there are
tools to do this.

 Simulation models are classified as follows.

i. Static Vs Dynamic Models
ii. Deterministic Vs Stochastic Models
iii. Discrete Models Vs Continuous Models

Sim and Mod Intro Prepared by R. Kibuku 9

 SIMULATION MODELS

1. Static Models Vs Dynamic Models

 A static simulation model- is a representation of a system
at a particular time, or one that may be used to represent a
system in which time plays no role; examples of static
simulations are Monte Carlo models.

 Monte Carlo simulation, or probability simulation, is a

technique used to understand the impact of risk and
uncertainty in financial, project management, cost, and other
forecasting models.

 A dynamic simulation model- represents a system as it

evolves with time, such as a conveyor system in a factory

Sim and Mod Intro Prepared by R. Kibuku 10

 SIMULATION MODELS

2. Deterministic Vs Stochastic Models

 A Deterministic Model- is one whose output is
determined by a set of input variables and a
relationship established by the model.
 The input variables are also not random and can be
determined either by observation or otherwise.

 A Stochastic Model- is one that contains random

components. The output is therefore random and
cannot be pre-determined accurately.
 Stochastic models are estimates of the true
characteristics of the true model.

Sim and Mod Intro Prepared by R. Kibuku 11

 SIMULATION MODELS

3. Discrete Models Vs Continuous Models

 Discrete simulation model- is one whose state variables
change at separated points in time while

 Continuous simulation model- is one whose state

variables change continuously with time.

 NB. This does not mean that discrete models are used for
discrete systems only, or continuous models are used for
continuous systems only.

 Indeed, discrete models can be used on a continuous system

and vice versa.

 The choice of whether to use a discrete or continuous model

Sim and Mod Intro depends on the objective
Prepared by of the simulation study.
R. Kibuku 12
 ADVANTAGES OF SIMULATION

1. It can be used to analyze large and complex real

world systems that cannot be described accurately
by analytical solution models.

 Such systems cannot be solved analytically e.g. it may not

be possible to develop a mathematical model for operation
of a central government system.

2. It is relatively more flexible and straight forward

3. It allows to estimate the performance of an existing

system under some projected operating conditions
without interfering with the real system

Sim and Mod Intro Prepared by R. Kibuku 13

 ADVANTAGES OF SIMULATION

4. Simulation can be used to compare proposed

alternative system designs to see which are bet
meets the specified requirements.
 E.g. a manager can try out several policy decision
using simulation in a matter of minutes to identify
which one meets the organization’s needs.

5. Simulation helps to compress time.

 i.e. the effect of policies or changes on a system
which may take many years or months to observe in a
real world system can be obtained by computer
simulation in a short time.

Sim and Mod Intro Prepared by R. Kibuku 14

 ADVANTAGES OF SIMULATION

6. Simulation allows the inclusion of real world systems

where complicated mathematical models may not
permit
 E.g. in queuing models, Poisson and exponential
distributions are required but in simulation, any
probability distribution that can be used can be defined.

7. There is better control over experimental conditions in

simulation than would be possible in experimenting
with the actual system.

8. Simulation allows the study of the interactive effect of

individual variables or components to determine which
ones are important.

Sim and Mod Intro Prepared by R. Kibuku 15

THE DISADVANTAGES OF SIMULATION

1. Good simulation models are often expensive and can

consume a lot of time.

2. Simulation does not generate optimal solutions as

analytical models would do.

 Indeed, in a stochastic simulation model, only estimates of

the true parameters produce true characteristics of the
model parameters.

 Hence, if a valid analytical model is available, it is preferable

than a simulation model.

Sim and Mod Intro Prepared by R. Kibuku 16

THE DISADVANTAGES OF SIMULATION

3. Managers must generate all conditions and

constraints for the solution they want to
examine simulation does not produce
answers by itself. It is cumbersome.

4. Each simulation model is unique, and

therefore its solutions and inferences are not
usually transferable to other problems.

Sim and Mod Intro Prepared by R. Kibuku 17

THE DISADVANTAGES OF SIMULATION

5. The large volume of numbers produced by a

simulation study or the persuasive impact of
realistic animation often creates a tendency
to place greater confidence in a study’s
results than is justified.

If a model is not a valid representation of the

system under study, the simulation results no
matter how impressive they appear, will
provide little useful information about the
actual system.

Sim and Mod Intro Prepared by R. Kibuku 18

PITFALLS TO SUCCESSFUL SIMULATION

 Why do simulation processes fail?

1. Failure to have a well defined set of objectives at the
beginning of the simulation study.

2. Having incorrect level of model detail

3. Failure to communicate with the management throughout

the cause of the study

4. Lack of understanding of simulation by the management

5. Lack of adequate knowledge of simulation methodology

Use of poorly collected

6. Intro
Sim and Mod data
Prepared by R.that
Kibukuis unreliable 19
PITFALLS TO SUCCESSFUL SIMULATION

7. Use of wrong performance measures.

8. Treating simulation as if it were primarily an exercise in computer

programming

9. Failure to have people with a knowledge of simulation methodology

and statistics on the modeling team

10. Failure to collect good system data

11. Inappropriate use of simulation software

12. Misuse of animation

13. Failure to account correctly for sources of randomness in the actual

system.

Sim and Mod Intro Prepared by R. Kibuku 20

 STEPS IN A SOUND SIMULATION STUDY

1. Formulate the problem and the plan the

study
a) Problem of interest is stated by manager.

b) One or more kick off meetings for the study are

conducted, with the project manager, the
simulation analysts, and Subject Matter Experts
(SMEs) in attendance.

 The following issues are discussed:

 Overall objectives of the study
 Specific questions to be answered by the study

Sim and Mod Intro Prepared by R. Kibuku 21

STEPS IN A SOUND SIMULATION STUDY

 Performance measures that will be used to

evaluate the efficiency of different system
configurations

 Scope of the model

 System configurations to be modeled

 Software to be used

 Time frame for the study and the required

resources
Sim and Mod Intro Prepared by R. Kibuku 22
 STEPS IN A SOUND SIMULATION STUDY

2. Collect data and define a model

a) Collect information on the system layout and operating procedures
 No single person or document is sufficient
 Some people may have inaccurate information-make sure that true
SMEs are identified
 Operating procedures may not be formalized

b) Collect data (if possible) to specify model parameters and input

probability distributions

c) Delineate the above information and data in an “assumptions

document”, which is the conceptual model

d) Collect data (if possible) on the performance of the existing system

(for validation purpose

e) The level of model detail should depend on the following

Sim and Mod Intro Prepared by R. Kibuku 23
STEPS IN A SOUND SIMULATION STUDY

 Project objectives
 Performance measures
 Data availability
 Credibility concerns
 Computer constraints
 Opinions of SMEs
 Time and money constraints

f) There need not be a one-to-one correspondence between

each element of the model and the corresponding element of
the system

g) Interact with the manager (and other key project personnel)

on a regular basis

Sim and Mod Intro Prepared by R. Kibuku 24

STEPS IN A SOUND SIMULATION STUDY

3. Is the conceptual model valid?

a) Perform a structured walk through of the
conceptual model using the assumptions
document before an audience of managers,
analysts, and SMEs
 Helps ensure that the model’s assumptions are
correct and complete
 Promotes ownership of the model
 Takes place before programming begins to
avoid significant reprogramming later.

Sim and Mod Intro Prepared by R. Kibuku 25

STEPS IN A SOUND SIMULATION STUDY

4. Construct a computer program and

Verify(walkthrough and testing)
a) Program the model in a programming language (e.g C or
FORTRAN) or in simulation software (e.g Arena, AutoMod,
Extend, Promodel, WITNESS).

 Benefits of using programming language are that one is often

known, they have a low purchase cost, and they may result in
a smaller model execution time.

 The use of simulation software on the other hand, reduce

programming time and results in a lower project cost.

b) Verify (debug) the simulation computer program

Sim and Mod Intro Prepared by R. Kibuku 26

 STEPS IN A SOUND SIMULATION STUDY

5. Make pilot study or a dry runs

a) Make pilot runs for validation purpose

6. Is the programmed model valid?

a) If there is an existing system, then compare model and
system performance measures for the existing system

b) Regardless of whether there is an existing system, the

simulation analysts and SMEs should review the model
results for correctness.

c) Use sensitivity analyses to determine what model factors

have a significant impact on performance measures and
thus have to be modeled carefully.

Sim and Mod Intro Prepared by R. Kibuku 27

 STEPS IN A SOUND SIMULATION STUDY

7. Design experiments
 Specify the following for each system
configuration of interest
 Length of each run
 Length of the warm-up period, if one is
appropriate
 Number of independent simulation runs using
different random numbers facilitates
construction of confidence intervals

Sim and Mod Intro Prepared by R. Kibuku 28

STEPS IN A SOUND SIMULATION STUDY

8. Make production runs

 Production runs are made for use in step 9
9. Analyze output data
a) Two major objectives in analysing output
data are;
 Determining the absolute performance of
certain system configuration
 Comparing alternative system configurations in
a relative sense

Sim and Mod Intro Prepared by R. Kibuku 29

 STEPS IN A SOUND SIMULATION STUDY

10. Document, present, and use results

a) Document assumptions, computer program and
study’s results for use in the current and future
projects
b) Presents study’s results
 Use animation to communicate model to managers and
other people who are not familiar with all the model
details
 Discuss model building and validation process to
promote credibility.
c) Results are used in decision making process if
they are both valid and credible

Sim and Mod Intro Prepared by R. Kibuku 30

 PRINCIPLES USED IN MODELING

 These principles represent the different viewpoints from which

to judge the information included in a model.

a) Block-building – the description of the system should be

organized in a series of blocks.

The main aim of constructing blocks is to simplify the

specification of the interactions within the system.

Each block describes a part of the system that depends upon

a few, preferable one, input variable and results in a few output
variables.

The system as a whole can then be described in terms of

interconnections between the blocks.

Sim and Mod Intro Prepared by R. Kibuku 31

PRINCIPLES USED IN MODELING

b) Relevance- the model should only include those

aspects of the system that are relevant to the study
objectives.

E.g. if the factory system study aims to compare the

effects of different operating rules on efficiency, it is
not relevant to consider the hiring of employees as
an activity.

While irrelevant info in the model may not do no

harm, it should be excluded because it increases the
complexity of the model and causes more work in
solving the model

Sim and Mod Intro Prepared by R. Kibuku 32

PRINCIPLES USED IN MODELING

c) Accuracy- accuracy of info gathered for the model should be

considered. E.g. in the aircraft system the accuracy with which
the movement of the aircraft is described depends upon the
representation of the airframe.

 It may suffice to regard the airframe as a rigid body and derive a

very simple relationship between control surface movement and
aircraft heading, or it may be satisfied with the simple
representation. Another engineer responsible for estimating the
fuel consumption may be satisfied with the simple
representation.

 Another engineer, responsible for considering the comfort of the

passengers, needs to consider vibrations and will want the
detailed descriptions of the airframe.

Sim and Mod Intro Prepared by R. Kibuku 33

PRINCIPLES USED IN MODELING

 Aggregation- a further factor to be considered is the extent to which the

number of individual entities can be grouped together into larger entities.

 The general manager of the factory may be satisfied with the description
that has been given. The production control manger, however, will want
to consider the shops of the departments as individual entities.

 In some studies it may be necessary to construct artificial entities through

the process of aggregation. E.g. an economic or socail study will usually
treat a population as a number of social classes and conduct a study as
though each social class were a distinct entity.

 Similar consideration should be given to the representation of activities.

E.g. in studying a missile defense system, it may not be necessary to
include the details of computing a missile trajectory for each firing. It may
be sufficient to represent the outcome of many firings by a probability
function.

Sim and Mod Intro Prepared by R. Kibuku 34

 SIMULATION AND MODELING Intro

THE END

Sim and Mod Intro Prepared by R. Kibuku 35