SSCM4833
Discrete Event Simulation
OBJECTIVES
Understand discrete event
simulation
Develop a simulation
Objectives of this See how it applies to real
model and program to
course systems.
solve a real-life problem.
Understand its strengths and
weaknesses.
SIMULATION
Imitation of the system of a real-world process.
To understand and estimate behaviour of a
system.
Observation is done by developing a simulation
model.
Can be ‘continuous’ time or discrete event.
WHY
SIMULATE ?
WHY SIMULATE?
▪ System does not exist – unable to experiment on the real system.
Product production line
▪ Cost-effective - to avoid unnecessary spending.
Adding a machine in product production.
▪ Avoid high-risk experimentation
Chernobyl nuclear reactor incident (1986)
Human mistakes.
▪ Observe highly complex systems – precluding the possibility of
analytical solution.
WHY SIMULATE?
▪ Complex analytical solutions – require vast computing
resources.
▪To investigate and improve the weakness in the current systems.
▪ To determine the important variables and the relationship
between variables in the systems. Important in order to predict
the system performance if the variables are changed.
▪ To test any policy or system design experimentally before
applying on the real system.
AREAS OF
APPLICATION
Corporate planning
Inventory control systems
Job-shop scheduling
Banking sector
Hospital
Traffic system
Transportation
AREAS OF
APPLICATION
Manufacturing
Human
Systems Construction
Areas of
Application
Business
Process Military
Logistics,
Transportation,
Distribution
SYSTEM
&
MODEL
SYSTEM
▪ A collection of entities that act and interact together towards the
accomplishment of some logical end in order to achieve a certain
objective.
▪ Discrete System – state variables changes instantaneously at
separated point in time.
▫ i.e. a queuing system in a bank
▫ number of customers (state variables) change when a customer
arrives or when a customer finishes being served and departs.
▪ Continuous System – state variables change continuously with respect
to time.
▫ i.e. airplane moving through the air
▫ position and velocity (state variables) change continuously with
respect to time.
SYSTEM
Experiment with
actual system
System Physical model
Experiment with a
model of actual Analytical solution
system
Mathematical model
Simulation
COMPONENTS IN A
SYSTEM
Activities
State
Attributes
Variables
Entity SYSTEM Events
COMPONENTS IN A
SYSTEM
• An object of interest in the system to be used for system modelling
purposes : e.g. machines in factory
Entity
• Entity to be chosen must has impact to the objective function
• have certain properties called attributes.
Attribute • characteristic / property of an entity : e.g. speed, capacity
Activities • A job that has a time period of a specified length : e.g. welding, stamping
• collection of variables necessary to describe the system at any time.
State Variables
• e.g. status of machines (busy, idle, down)
• instantaneous occurrence that may change the state of the system.
Events • an event initiates activities.
• e.g. machine breakdown.
EXAMPLES OF
COMPONENTS IN A
SYSTEM
System Entities Attributes Activities Events State Variables
Banking Customers Checking Making Arrival; Number of busy
account deposits departure tellers, number of
balance customers waiting.
Production Machines Speed; Welding; Breakdown Status of machines
capacity; stamping (busy, idle, or
breakdown down)
rate
Communications Messages Length; Transmitting Arrival at Number waiting to
destination destination be transmitted
Inventory Warehouse Capacity Withdrawing Demand Levels of inventory;
backlogged
demand.
Name several entities, attributes,
activities, events, and state variables
for the following systems:
Class
Exercise A cafeteria
A laundromat
A hospital emergency room
A grocery store
A fast-food restaurant
MODEL
▪ Model is a representation of a system.
▪ Can be used as a surrogate of a system.
▪ Good model – experiments will give results close to the actual
system.
▪ Main factor to the successful simulation analysis.
▪ Types of model: iconic, analog & symbolic.
THE TASK:
1. Define iconic model, analog model & symbolic model.
2. State the differences between these models.
3. Give examples.
SYMBOLIC MODEL
Static
Deterministic Continuous
Dynamic
System Model Discrete
Static (Monte Carlo
Simulation)
Stochastic
Continuous
Dynamic
Discrete (Discrete-Event
Simulation)
ADVANTAGES &
DISADVANTAGES
OF SIMULATION
▪ Advantages ▪ Disadvantages
New hardware designs, physical Model building requires special
layouts, transportation systems and training & expertise.
so on can be tested.
Simulation results can be difficult
Hypotheses about how or why to interpret - random nature of
certain phenomena occur can be simulation models.
tested for feasibility.
Simulation modeling and analysis
New policies, decision rules, can be time consuming and
organizational procedures and so on expensive.
can be explored without disrupting
ongoing operations of the real Difficult to get cooperation and
system. involvement from the user in data
collection and model
Simulation time can be controlled – development.
can be set longer or shorter.
Can control the state variables.
Class When does the
Discussion
simulation is not
appropriate?
