chapter1_sqqp3033
TRANSCRIPT
-
8/3/2019 Chapter1_SQQP3033
1/32
Introduction to SimulationChapter 1
-
8/3/2019 Chapter1_SQQP3033
2/32
Objectives
i. Define simulation, system and modelii. List types of simulation modelsiii. List types of constructive simulationiv. Introduce Discrete Event Simulation (DES)
conceptsv. List suitable problems for DESvi. Overview of a simulation studyvii. Identify different tools for building simulation
modelsviii.List advantages and disadvantages of simulationix. List paths to failure and success in conducting a
simulation project
-
8/3/2019 Chapter1_SQQP3033
3/32
Educational perspective: Simulation is a digitallearning material which allows learners toperform hands-on activities (e.g., mouseclicking, text entering, etc.) in order to receive
additional tasks or information. Engineering perspective: Simulation refers to
the process of designing a model of a realsystem for the purpose of understanding the
behaviour of the system and/or evaluatingvarious strategies on its operations based onspecified objectives of a study.
Simulation
-
8/3/2019 Chapter1_SQQP3033
4/32
System & Model
System a group or collection of interrelatedelements that cooperate to accomplish somestated objective; e.g., bank, airport, etc.
Modeling the process of producing a model
Model a representation of a system s operations
or processes similar to but simpler than the system it
represents enables the analyst to predict the effect of
changes to the system
-
8/3/2019 Chapter1_SQQP3033
5/32
live simulations (or role playing) - real peopleoperate simulated operations of a real systemusing real equipment (e.g., training exercise of afire drill).
virtual simulations - real people operatesimulated equipment in a simulated environment(e.g., flight and vehicle simulators).
constructive simulations - real people operatecomputerised models by performing interactionsand get feedbacks.
Types of Simulation Models
-
8/3/2019 Chapter1_SQQP3033
6/32
Type Learning support Example
Soft skill Simulation Exposing learners with workexperiences before they actually deal
with real situations in order to improvelearners communication and decisionskills
Software applicationsimulations
Situation-based simulations;e.g., Business andManagement training,customer and sales training,doctor-patient interactionsand customer services
Procedural Simulationsor Virtual products
Understanding the physicalcharacteristics of real life equipmentand learning to use complicatedequipment
Mechanical devicesimulations; e.g.,medical/manufacturing /homeelectronic equipment
Discrete Event
Simulations (DES)
Understanding the operation of a
system which is represented by achronological sequence of events
Queuing systems
Manufacturing systems Logistic models
System Dynamics (SD) Understanding the behaviour ofcomplex systems over time whichinvolves internal feedback loops, stocks(entities which accumulate or depleteover time) and flow (the rate change of
the stock)
Policy analysis and design Population systems Ecological systems Economic systems
Types of Constructive Simulations
-
8/3/2019 Chapter1_SQQP3033
7/32
Limit our study to Discrete Event Simulation(a.k.a. DES)
DES operations of a system is represented by achronological sequence of events (i.e., discretepoints of times where state variables changevalues)
State variables variables that affect systemperformance; e.g., resource utilization, totalproduction, number in a queue, etc.
Discrete Event Simulation (DES)
-
8/3/2019 Chapter1_SQQP3033
8/32
A model intended for a DES study is amathematical model developed using DES tools(languages or software).
Mathematical model classifications include:
deterministic input and output variables arefixed values
stochastic at least one of the input or outputvariables is probabilistic
static time is not taken into account dynamic time-varying interactions among
variables are taken into account Typically, DES models are stochastic and
dynamic
DES (contd.)
-
8/3/2019 Chapter1_SQQP3033
9/32
Entity any objects that enter a system, move
through a series of processes, and then leavethe system; e.g., customers, patients, parts,
etc. can have individual characteristics
(attributes)- a common characteristic of all
entities but with a specific value that candistinguish the entity; e.g., name, priority, duedate, account number, etc.
DES Concepts
-
8/3/2019 Chapter1_SQQP3033
10/32
DES Concepts (contd.)
Resources any objects that process entities;e.g., workers, machines, etc.
Queue a space to wait a resource.
DES Model a computer program thatrepresents the logic of a system as entities withattributes arrive, join queues to be allocated and
processed with a number of required resources,and then exit the system.
-
8/3/2019 Chapter1_SQQP3033
11/32
Conclusion
AlteredSystem
Real World Simulation Study
DES Study Schematic
SimulationAnalysis
SimulationExperiment
Simulation
Model
SystemUnderStudy
-
8/3/2019 Chapter1_SQQP3033
12/32
DES Projects
DES modeling involves two fields: Art system analysis and modeling Science the programming and statistical
components
-
8/3/2019 Chapter1_SQQP3033
13/32
DES Project (contd.)
Most are conducted by team
Need two types of people:1) people who know and understand the
considered system; e.g., designers, systems,
manufacturing or process engineers2) people who know how to formulate and model
the system as well as program the model(simulation specialists)
If we do not have (2) hire people with the necessary skills contract the modeling to outside consultants
-
8/3/2019 Chapter1_SQQP3033
14/32
train some of our own people data collection,statistical skills and programming skills
some combination of the above
DES Project (contd.)
-
8/3/2019 Chapter1_SQQP3033
15/32
Suitable Problems For DES
Generally, when there is a need to model andanalyze randomness in a system. Morespecifically: It is impossible or extremely expensive to
modify certain processes in the real world; e.g.,large manufacturing systems
Problems in which mathematical model can be
formulated but analytic solutions are eitherimpossible or too complicated (e.g., complexsystems like large scale queuing models)
-
8/3/2019 Chapter1_SQQP3033
16/32
It is impossible or extremely expensive tovalidate the mathematical model describingthe system; e.g., due to insufficient data
Suitable areas:
Computer and communication systems Manufacturing Transportation Service systems
Suitable Problems For DES (contd.)
-
8/3/2019 Chapter1_SQQP3033
17/32
A Simulation Study
Steps that should be present in any simulation
study:1) Problem DefinitionClearly defining the goals of
the study The purpose why are we studying the problem What questions do we hope to answer?
2) System DefinitionDetermining the boundariesand restrictions to be used in defining the system(or process) and investigating how the systemworks Divide the system into logical subsystems. Define the entities and resources Define the basic flow patterns of entities through the
stations using flow diagrams
-
8/3/2019 Chapter1_SQQP3033
18/32
3) Conceptual Model FormulationDeveloping apreliminary model either graphically (e.g. blockdiagram or process flow chart) or in pseudo-codeto define the components, descriptive variables,
and interactions (logic) that constitute the system4) Collect and process real system dataIdentifying and collecting the input data neededby the model. Example:
Time between arrival (Inter-arrival Time) Processing time at each station Route time (travel time) between stations
-very time consuming and difficult (1/3 of the total
time used in the study)
A Simulation Study (contd.)
-
8/3/2019 Chapter1_SQQP3033
19/32
Potential sources of input data Historical records Observational data Similar systems Operator estimates Vendor's claims Designer estimates Theoretical considerations
5) Model Translation Formulating the model in an
appropriate simulation language
A Simulation Study (contd.)
-
8/3/2019 Chapter1_SQQP3033
20/32
6) Verification and ValidationVerification make sure the computer programperforms as expected and intended.Validation the process of reaching an
acceptable level of confidence that the inferencesdrawn are correct and applicable to the real-worldsystem being represented. Does the model adequately represent the real-world
system. Is the model generated behavioral data characteristic of
the real-world system's behavioral data? Does the simulation model user have confidence in the
model's results?
A Simulation Study (contd.)
-
8/3/2019 Chapter1_SQQP3033
21/32
A Simulation Study (contd.)
7) Design and Run Experiment A simulationexperiment is a test or a series of tests in whichmeaningful changes are made to the inputvariables of a simulation model so that we may
observe and identify the reasons for changes inthe performance measures
8) Analysis and Interpretation Drawing inferencesfrom the data generated by the simulation runs
-
8/3/2019 Chapter1_SQQP3033
22/32
9) Implementation and DocumentationReportingthe results, putting the results to use, recordingthe findings, and documenting the model and itsuse. Choosing an appropriate vocabulary (no
technical jargon). Length and format of both written and verbal
reports (short and concise). Must address the issues that the sponsor or
user consider important.
A Simulation Study (contd.)
-
8/3/2019 Chapter1_SQQP3033
23/32
DES Tools
A simulation model can be built using: General-purpose programming languages General-purpose simulation language more
flexible
Special purpose simulation packages areeasier to learn
Some of the advantages of using a simulationpackage Reduction of the programming task Provision of conceptual guidance Increased flexibility when changing the model Fewer programming errors Automated gathering of statistics
-
8/3/2019 Chapter1_SQQP3033
24/32
Types Of Simulation Packages Examples
Programming Languages FORTRAN, C++, Visual BasicSimulation Languages/packages AweSIM , Extend , GPSS/H , Micro Saint,
SIMSCRIPT , SLX, MODSIM III, SIMPLE++
High Level Simulators Arena, SIMUL8, Flexim
Application-Oriented Simulators Manufacturing:
AutoMod, ProModel, QUEST,WITNESS
Communication/computer:
COMNET III, NETWORK II.5, OPNET Modeler
Business:BP$IM, Extend+BPR, ProcessModel,ServiceModel, SIMPROCESS
Health Care:
MedModel
DES Tools (contd.)
-
8/3/2019 Chapter1_SQQP3033
25/32
We can test new designs, layouts, etc. withoutcommitting resources to their implementation
It can be used to explore new staffing policies,operating procedures, decision rules,organizational structures, information flows, etc.without disrupting the ongoing operations
Simulation allows us to identify bottlenecks in
information, material and product flows and testoptions for increasing the flow rate
It allows us to test hypothesis about how or why
certain phenomena occur in the system
Advantages of Simulation
-
8/3/2019 Chapter1_SQQP3033
26/32
Simulation allows us to control time. Thus we canoperate the system for several months or years ofexperience in a matter of seconds allowing us toquickly look at long time horizons or we can slow
down phenomena for study It allows us to gain insights into how a modeled
system actually works and understanding of which
variables are most important to performance Simulation's great strength is its ability to let us
experiment with new and unfamiliar situations andto answer "what if" questions
Advantages of Simulation (contd.)
-
8/3/2019 Chapter1_SQQP3033
27/32
Simulation modeling is an art that requiresspecialized training and therefore skill levels ofpractitioners vary widely. The utility of the studydepends upon the quality of the model and the
skill of the modeler
Gathering highly reliable input data can be timeconsuming and the resulting data is sometimes
highly questionable. Simulation cannotcompensate for inadequate data or poormanagement decisions
Disadvantages of Simulation
-
8/3/2019 Chapter1_SQQP3033
28/32
Simulation models are input-output models, i.e.they yield the probable output of a system for agiven input. They are therefore "run" rather than
solved. They do not yield an optimal solution,rather they serve as a tool for analysis of thebehavior of a system under conditions specifiedby the experimenter
Disadvantages of Simulation (contd.)
-
8/3/2019 Chapter1_SQQP3033
29/32
Failure to define a clear and achievable goal.
Inadequate planning and underestimating theresources needed
Inadequate user participation
Writing code too soon before the system is reallyunderstood
Inappropriate level of included detail (usually toomuch)
Wrong mix of team skills
Paths to Failure
-
8/3/2019 Chapter1_SQQP3033
30/32
Paths to Failure (contd.)
Lack of trust, confidence and backing bymanagement
-
8/3/2019 Chapter1_SQQP3033
31/32
Have clearly defined and achievable goals Be sure we have adequate resources available to
successfully complete the project on time
Have management's support and have it knownto those who must cooperate with us in supplyinginformation and data
Assure that we have all the necessary skillsrequired available for the duration of the project
Be sure that there are adequate communicationchannels to the sponsor and end users
Paths to Success
-
8/3/2019 Chapter1_SQQP3033
32/32
Paths to Success (contd.)
Have a clear understanding with the sponsor andend users as to the scope and goals of theproject as well as schedules
Have good documentation of all planning andmodeling efforts