SIMULATION AND MODELING -- IV INFORMATION TECHNOLOGY

Chapter 1 Introduction Simulation and Modeling

System-A system is defined as an aggregation or assemblage of objects joined in

some regular interaction or interdependence toward the accomplishment of some purpose.

E ample ! Production Control System

"roduction Control System

Components of a System
Entity An entity is an object of interest in a system. E ! In the factory system# departments# orders# parts and products are the entities. Attribute An attribute denotes the property of an entity. E ! $uantities for each order# type of part# or number of machines in a %epartment are attributes of factory system. Activity Any process causing changes in a system is called as an acti&ity. E ! Manufacturing process of the department

System Models
A model is defined as a representation of a system for the purpose of studying the system. It is necessary to consider only those aspects of the system that affect the problem under in&estigation. Page 1

HARAMAYA UNIVERSITYDEPARTMENT OF IT

SIMULATION AND MODELING -- IV INFORMATION TECHNOLOGY

'hese aspects are represented in a model# and by definition it is a simplification of the system.

Types of System Models

Mathematical or "hysical Model Static Model %ynamic Model %eterministic Model Stochastic Model %iscrete Model Continuous Model Mathematical Model (ses symbolic notation and the mathematical e)uations to represent a system. Static Model *epresents a system at a particular point of time and also +nown as Monte-Carlo simulation. Dynamic Model *epresents systems as they change o&er time. E ! Simulation of a ban+ Deterministic Model Contains no random &ariables. 'hey ha&e a +nown set of inputs which will result in a uni)ue set of outputs. E ! Arri&al of patients to the %entist at the scheduled appointment time. Stochastic Model ,as one or more random &ariable as inputs. *andom inputs leads to random outputs. E ! Simulation of a ban+ in&ol&es random inter arri&al and ser&ice times. Discrete and Continuous Model (sed in an analogous manner. Simulation models may be mi ed both with discrete and continuous. 'he choice is based on the characteristics of the system and the objecti&e of the study. A Simulation is the imitation of the operation of a real-world process or system o&er time. Brief Explanation 'he beha&ior of a system as it e&ol&es o&er time is studied by de&eloping a simulation model. 'his model ta+es the form of a set of assumptions concerning the operation of the system. 'he assumptions are e pressed in o Mathematical relationships o -ogical relationships Page .

Simulation

HARAMAYA UNIVERSITYDEPARTMENT OF IT

SIMULATION AND MODELING -- IV INFORMATION TECHNOLOGY

o Symbolic relationships /etween the entities of the system.

Needs of Simulation
Simulation enables the study of and e perimentation with the internal interactions of a comple system# or of a subsystem within a comple system. Informational# organi0ational and en&ironmental changes can be simulated and the effect of those alternations on the model1s beha&ior can be obser&er. 'he +nowledge gained in designing a simulation model can be of great &alue toward suggesting impro&ement in the system under in&estigation. /y changing simulation inputs and obser&ing the resulting output1s# &aluable insight may be obtained into which &ariables are most important and how &ariables interact. Simulation models designed for training# allow learning without the cost and disruption of on-the-job learning.

Application of Simulation
Manufacturin Applications Semiconductor Manufacturin Military Applications !o istics" Transportation and Distribution Applications Business Process Simulation

Continuous System Simulation

Continuous Systems Systems in which the changes are predominantly smooth are called continuous system. E ! ,ead of a water behind a dam.

HARAMAYA UNIVERSITYDEPARTMENT OF IT

Page 2

SIMULATION AND MODELING -- IV INFORMATION TECHNOLOGY

System Dynamics
In scientific literature studies connected with industrial problems are called Industrial Dynamics 3hereas study of urban problems is called Urban Dynamics. Similarly control of en&ironmental problems is called World Dynamics. In all these systems there is no difference in the techni)ues to be used to study the system# therefore it is appropriate to call this field as System Dynamics.

"robability (sed in Simulation Probability of occurrence of an event = total favorable events/total number of experiments. -et us consider a simple e periment of tossing a coin. Can one predict# whether head or tail will come. 4o# it is not possible to predict. In order to +now the answer# we ha&e to toss the coin at least hundred times. Suppose we get 56 times head and 71 times tail. 'hen probability of getting head in hundred trials is defined as 8total number of heads9total number of toss:. 'hus getting head in tossing a coin is an e&ent.

/asic "robability Concept

HARAMAYA UNIVERSITYDEPARTMENT OF IT

Page 5

SIMULATION AND MODELING -- IV INFORMATION TECHNOLOGY

'he probability that a customer has to wait in the )ueue is ;.<7. 'his is determined in the following manner!

Arrival Process
The Arrival Process:Arri&al process for infinite-population models is usually characteri0ed in terms of inter arri&al times of successi&e customers. Arri&als may occur at scheduled times or at random times. Page 7

HARAMAYA UNIVERSITYDEPARTMENT OF IT

SIMULATION AND MODELING -- IV INFORMATION TECHNOLOGY

3hen at random times# the inter arri&al times are usually characteri0ed by a probability distribution E#AMP!E $% Sin le&Channel 'ueue A small grocery store has only one chec+out counter. Customers arri&e at this chec+out counter at random from 1 to = minutes apart. Each possible &alue of inter arri&al time has the same probability of occurrence. 'he ser&ice times &ary from 1 to < minutes with the probabilities shown in table 7. 'he problem is to analy0e the system by simulating the arri&al and ser&ice of .; customers.

E ample-Arri&al "rocess and Ser&ice times

HARAMAYA UNIVERSITYDEPARTMENT OF IT

Page <

SIMULATION AND MODELING -- IV INFORMATION TECHNOLOGY

Ser&ice times for all .; customers are shown in table >. 'hese ser&ice times were generated based on the methodology described abo&e# together with the aid of table 7. 'he first customer1s ser&ice time is 5 minutes because the random digits =5 fall in the brac+et <1-=7# or alternati&ely because the deri&ed random number ;.=5 falls between the cumulati&e probabilities ;.<1 and ;.=7.

HARAMAYA UNIVERSITYDEPARTMENT OF IT

Page >