Welcome To The Discrete-Event Simulation Training Module: Self-Learning Anyone Developing A Training Program

Welcome to the Discrete-Event
Simulation Training Module

This training material is composed of a series of PowerPoint slides
complete with speaker notes. It is recommended for:
self-learning
anyone developing a training program
Numerous reviews have stated that TEMPLETs modules are both
high quality and relevant to students and professionals alike. We are
proud that companies and academics around the world have used
these materials extensively for learning advancement.
The first 34 slides of this module are provided as a preview. The
complete 63 slide module, which addresses the topic of Discreteevent Simulation in more depth and breadth, is available for purchase
(see last slide).
For customized versions of this module, such as inclusion of case
studies for your industry, please contact [email protected].
Introduction
to Discrete-Event Simulation
of Manufacturing Systems
If you are not seeing speaker notes with this version,
please right click and select edit slides.
Agenda
Purpose and applications of simulation

Comparison of alternative methods
Basic definitions and components of a system
Types of simulation models
Random variables and probability
Steps in a simulation study
Output analysis
Examples
Purpose and applications

To model and analyse the performance of a
manufacturing system
At the design stage
Material handling systems
Number and types of machines
Facility layout
At the operational stage

Scheduling
Labour assignment
Improvement efforts
System Performance Metrics

Simulation is used to determine desired
performance metrics, including:
Flow time
Utilization
Value-added time
Waiting time
Flow rate
Inventory levels
Yields
Etc.
Alternative methods
Rough hand calculations
Appropriate for simple problems only
Spreadsheet models
Cannot model variability and interactions
Queuing theory
Appropriate for simple problems only
Gives only steady state averages
Simulation
Can model complex systems with randomness and
interactions
Simulation vs. Analytical Methods

Each have their own merits for different
levels of analysis
Comparison of specific aspects
Macro vs. micro
Uniformity vs. randomness
Effects of interactions
Single answer vs. range of outcomes
Numberical vs. animation
Macro vs. Micro

Analytical methods ignore the differences
between individuals, and rely on the
analysis of average behaviour.
Simulation models, on the other hand, use
the distribution of population behaviour,
and can account accurately for peaking of
demand during short time intervals.
Uniformity vs. Randomness

Analytical models assume that processes
are distributed uniformly and behave
homogeneously over a period of time.
Simulation models recognize and account
for the randomness of processes.
Effects of Interactions
Analytical methods are limited to
evaluating the outcome of a stand-alone
process and cannot recognize the
interaction effects that parallel processes
may have on that process.
Simulation methods are able to explicitly
model such interaction effects, allowing the
analyst to measure the impact of process
interactions.
Single Answer vs. Range of Outcomes

The solution of analytical models yields
one answer to the question of interest.
Simulation models, in contrast, generate a
variety of statistics that can be used to
evaluate a systems performance.
Numerical vs. Animation

Analytical solutions give only numerical
results, which are simple, but less
persuasive.
Simulation models, however, may provide
computer animation in addition to
numerical results. This picture of the
result can be very persuasive in conveying
the result of analysis.
Basic Definitions
Simulation: the imitation of the operation of
a real-world process or system over time.
System: a collection of objects which act
and interact together toward the
accomplishment of some objectives.
Model: a representation of a system for the
purpose of studying the system.
Environment: anything outside the system
that could affect the state of the system.
Components of a System:
Entity: an object of interest in the system (e.g., workparts,
machine tools, workers, etc.).
Attribute: property of an entity (e.g., machine status,
workpart status, utilization, etc.).
Activity: a time period of specified length (e.g., processing
time to drill a hole).
State: the collection of variables necessary to describe the
system at any time, relative to the objectives of the study
(e.g., number of parts in queues, status of machines,
etc.).
Event: an instantaneous occurrence that may change the
state of the system (e.g., a machine breakdown).
Types of Activities and Events:

Endogenous:
Activities and events occurring within the
system.
Exogenous:
Activities and events occurring in the
environment that affect the system.
Types of Simulation Models:

Static vs. Dynamic
Deterministic vs. Stochastic
Discrete vs. Continuous
Static vs. Dynamic

Static: representing a system at a
particular point in time (Monte Carlo
simulation).
Dynamic: representing a system as it
changes over time.
Deterministic vs. Stochastic

Deterministic: has a known set of inputs
which results in a unique set of outputs.
Stochastic: has one or more random
variables as inputs. Random inputs lead to
random outputs.
4
3
2
1
Time
Figure 1.1. Discrete-system state variable.

Head of water behind the dam
Discrete: the state

variables change only at a
discrete set of points in
time. (e.g., number of
customers in a bank)
Continuous: the state
variables change
continuously over time.
(e.g., the level of water
behind a dam)
Number of customers waiting in line
Discrete vs. Continuous
Time
Figure 1.2. Continuous-system state variable.
Discrete-Event System Simulation

The modeling of dynamic systems in
which state variables change
stochastically only at a discrete set of
points in time
Discrete-Event Simulation
Time
State 1
Start
simulation
State 2
Event 1
State 3
Event 2
Event 3
State n
Event n
Events are determined by system variables and interactions

Each event changes the system state
Example
Event:
Vertical machining centre (VMC) breaks down
7h 32m 14s after the simulation starts (the
time is a random variable)
Change in state:
VMC status changes to down
System Variables
Decision variables
These are input variables that can be modified (e.g.,
number of workers, number of machines, routing logic)
Response variables
These are the outputs representing the performance of
the system (e.g. WIP levels, utilization, etc.)
State variables
These completely define the state of the system at any
discrete point in time
Systems Approach to System

Simulation
Identify the problems and opportunities
E.g., reduce inventory by 20%
Develop alternative solutions

E.g., alternative scheduling methods (input variables)
Evaluate the solutions

Perform simulation studies
Analyse results (output variables)
Select and implement the best solution
Random Variables
Events are determined by random
variables
Machine breakdowns and repairs
Manual processing time
Part and order arrival times
Probability Distributions
Random variables are modelled using
probability density functions (PDF)
For example, machine breakdowns are
often approximated by an exponential PDF
The mean value of the PDF is equal to the
Mean Time Between Failures (MTBF),
which is usually known
Discrete vs Continuous PDFs

Discrete PDFs
Describe the likelihood of a particular discrete value
occuring
For example, there are only six possible outcomes
from rolling a die
Continuous PDFs
Describe the likelihood of a continuous value lying
within a specified range
For example, a machine can break down at any time.
We can use a continuous PDF to determine the
probability that it wont break down for at least 100
hours
Some Discrete PDFs

Poisson Distribution
Often used to model the number of arrivals (parts,
orders, customers) occurring within a given time
interval
Binomial Distribution
Often used to model the number of defects in a batch
of parts
Uniform Distribution
Used when each value is equally likely (e.g., rolling a
die)
Common Continuous PDFs

The following PDFs are most commonly
used:
Uniform distribution
Triangular distribution
Normal distribution
Exponential distribution
Often used as a first

approximation
For example, it can
represent task
completion times that
vary uniformly
between some
minimum and
maximum values
Probability
Uniform Distribution
Minimum
Value
Maximum
Value
Can model a variable

based on a minimum,
maximum and most
likely value
The probability is
highest at the most
likely value
Probability
Triangular Distribution
Most likely
Value
Maximum
Minimum
Value
Value
Normal Distribution
Mean
Probability
Many variables are

normally distributed
This is the classic
bell curve
It is described by the
mean (average) and
standard deviation
(amount of variation)
Standard
Deviation
This distribution is
often used to describe
the time between
arrivals, breakdowns,
etc.
Number of events in a
given time is
described by a
Poisson distribution
Probability
Exponential Distribution
Mean
The Full Presentation

For the full 63 slide Powerpoint presentation,
complete with speaker notes, please go to:
www.templet.org/imm/en/cart.html
For more information, contact [email protected]

Welcome To The Discrete-Event Simulation Training Module: Self-Learning Anyone Developing A Training Program

Uploaded by

Copyright:

Available Formats

Welcome To The Discrete-Event Simulation Training Module: Self-Learning Anyone Developing A Training Program

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Welcome To The Discrete-Event Simulation Training Module: Self-Learning Anyone Developing A Training Program

Uploaded by

Copyright:

Available Formats

Welcome to the Discrete-Event

Simulation Training Module

Purpose and applications of simulation

Purpose and applications

At the operational stage

System Performance Metrics

Simulation vs. Analytical Methods

Macro vs. Micro

Uniformity vs. Randomness

Single Answer vs. Range of Outcomes

Numerical vs. Animation

Types of Activities and Events:

Types of Simulation Models:

Static vs. Dynamic

Deterministic vs. Stochastic

Figure 1.1. Discrete-system state variable.

Discrete: the state

Number of customers waiting in line

Discrete vs. Continuous

Figure 1.2. Continuous-system state variable.

Discrete-Event System Simulation

Events are determined by system variables and interactions

Systems Approach to System

Develop alternative solutions

Evaluate the solutions

Select and implement the best solution

Discrete vs Continuous PDFs

Some Discrete PDFs

Common Continuous PDFs

Often used as a first

Can model a variable

Many variables are

The Full Presentation

You might also like