Simulation Lec#12 (Discrete Simulation Models, Model Times).pdf

Lecture Overview:
This lecture provides an in-depth exploration of discrete simulation models and model times, essential components of simulation analysis. Students will learn how to design, develop, and apply discrete simulation models to analyze complex systems and make informed decisions.

Introduction to Discrete Simulation Models
Discrete simulation models are a type of simulation model that represents systems that change at discrete points in time. These models are widely used in various fields, including manufacturing, logistics, finance, and healthcare, to analyze and optimize complex systems.

Characteristics of Discrete Simulation Models
Discrete simulation models have several key characteristics, including:

- *Discrete time*: The model changes at discrete points in time, rather than continuously.
- *Discrete state*: The model has a finite number of states, and the state changes at discrete points in time.
- *Event-driven*: The model is driven by events, such as arrivals, departures, and service completions.

Types of Discrete Simulation Models
There are several types of discrete simulation models, including:

- *Queuing models*: These models represent systems with queues, such as banks, hospitals, and supermarkets.
- *Inventory models*: These models represent systems with inventory, such as manufacturing systems and supply chains.
- *Reliability models*: These models represent systems with reliability considerations, such as fault-tolerant systems and maintenance scheduling.

Model Times
Model times are an essential component of discrete simulation models, as they determine when events occur and how the model changes over time. There are several types of model times, including:

- *Simulated time*: This is the time that elapses within the simulation model.
- *Real time*: This is the actual time that elapses during the simulation run.
- *Wall-clock time*: This is the actual time that elapses during the simulation run, including any pauses or interruptions.

Simulation Clock
The simulation clock is a mechanism that advances the simulated time and schedules events within the model. The simulation clock can be implemented using various techniques, including:

- *Fixed-increment clock*: This clock advances the simulated time by a fixed increment at each tick.
- *Variable-increment clock*: This clock advances the simulated time by a variable increment at each tick.
- *Event-driven clock*: This clock advances the simulated time based on the occurrence of events within the model.

Event List Management
Event list management is a critical aspect of discrete simulation models, as it determines the order in which events occur and how the model changes over time. There are several techniques for managing the event list, including:

- *First-in-first-out (FIFO)*: This technique processes events in the order

1. Discrete Simulation Model, Model Time
MODELING & SMINULATION

2. ModelingandSimulation:CourseOutline
A
r
f
a
n
S
h
a
h
z
a
d
.
c
o
m

3. DiscreteSimulationModels
UnderstandingSystemDynamics
A
r
f
a
n
S
h
a
h
z
a
d
.
c
o
m
•Discretesimulationmodelsareprecioustoolsusedinvariousfieldsto
understandandanalyzethebehaviorofcomplexsystems.
•Unlikecontinuoussimulationmodels,whichrepresentsystemswhere
changesoccurcontinuouslyovertime,discretesimulationmodels
focusonsystemswhereeventsoccuratdistinctpointsintime,
leadingtodiscretechangesinsystemstate.

4. DiscreteSimulationModelscont…
UnderstandingSystemDynamics
A
r
f
a
n
S
h
a
h
z
a
d
.
c
o
m
•Thesemodelsarehighlyversatileandcanbeappliedtoawiderange
ofdomains,includingmanufacturing,transportation,healthcare,
finance,andmore.

5. DiscreteSimulationModelscont… Key
Components
•Entities:Representobjects,individuals,orcomponentswithinthe
systembeingmodeled.
•Entitiesinteractwitheachotherandtheenvironment,leadingto
changesinsystemstate.
A
r
f
a
n
S
h
a
h
z
a
d
.
c
o
m

6. DiscreteSimulationModelscont… Key
Components
•Events:Representspecificoccurrencesoractionsthataffectthestate
ofthesystem.
•Eventstriggerchangesinthesystem'sbehaviororstate,suchas
entityarrivals,departures,resourceallocations,orstatetransitions.
A
r
f
a
n
S
h
a
h
z
a
d
.
c
o
m

7. DiscreteSimulationModelscont… Key
Components
A
r
f
a
n
S
h
a
h
z
a
d
.
c
o
m
•SimulationClock:Dictatestheprogressionoftimeinthesimulation.
•Thesimulationclockdetermineswheneventsoccurandinwhat
sequence,allowinganalyststostudysystembehaviorovertime.

8. DiscreteSimulationModelscont… Key
Components
•SimulationEngine:Executesthesimulationbyprocessingeventsin
chronologicalorderandupdatingthestateofthesystemaccordingly.
•Thesimulationenginemaintainsascheduleofpendingeventsand
advancesthesimulationclocktothenexteventtime.
A
r
f
a
n
S
h
a
h
z
a
d
.
c
o
m

9. DiscreteSimulationModelscont…
Advantages
•Flexibility:Discretesimulationmodelscancaptureawiderangeof
systembehaviors,includingcomplexinteractions,variability,and
uncertainty.
•Granularity:Thesemodelsallowanalyststostudysystemdynamicsat
afinelevelofdetail,enablinginsightsintoindividualeventsandtheir
impactonsystembehavior.
A
r
f
a
n
S
h
a
h
z
a
d
.
c
o
m

10. DiscreteSimulationModelscont…
Advantages
A
r
f
a
n
S
h
a
h
z
a
d
.
c
o
m
•ScenarioAnalysis:Discretesimulationmodelsfacilitatethe
evaluationofdifferentscenarios,parametervalues,orinterventions,
helpinganalystsunderstandtheconsequencesofvariousdecisionsor
policies.

11. DiscreteSimulationModelscont…
Advantages
•PerformanceEvaluation:Thesemodelsenabletheassessmentof
systemperformancemetrics,suchasthroughput,resourceutilization,
waitingtimes,andservicelevels,providingvaluableinsightsfor
systemoptimizationandimprovement.
A
r
f
a
n
S
h
a
h
z
a
d
.
c
o
m

12. DiscreteSimulationModelscont… Model
Time
A
r
f
a
n
S
h
a
h
z
a
d
.
c
o
m
•Modeltimereferstotherepresentationoftimewithinasimulationmodel.
•Indiscretesimulationmodels,timeistypicallyrepresentedasasequence
ofdiscretetimesteps,whereeachstepcorrespondstotheoccurrenceof
aneventorthecompletionofaprocess.
•Thesimulationclockgovernstheprogressionoftime,advancingfromone
timesteptothenextaseventsoccur.

13. DiscreteSimulationModelscont… Model
Time:KeyAspects
A
r
f
a
n
S
h
a
h
z
a
d
.
c
o
m
•EventScheduling:Eventsarescheduledtooccuratspecificpointsin
modeltime,basedonpredeterminedeventtimesorconditions.
•Thesimulationenginemaintainsascheduleofpendingeventsand
processestheminchronologicalorder.

14. DiscreteSimulationModelscont… Model
Time:KeyAspects
•TimeAdvancement:Thesimulationclockdetermineswhenthenext
eventwilloccur,basedonthecurrentstateofthesystemandthe
scheduledeventtimes.
•Timeadvancementmaybedeterministicorstochastic,dependingon
thenatureoftheeventsandthesimulationenvironment.
A
r
f
a
n
S
h
a
h
z
a
d
.
c
o
m

15. DiscreteSimulationModelscont… Model
Time:KeyAspects
•SimulationHorizon:Thesimulationhorizondefinesthedurationof
thesimulation,specifyingthemaximumamountofmodeltimetobe
simulated.
•Analystsmaychoosethesimulationhorizonbasedontheobjectives
ofthesimulationandthedesiredlevelofdetail.
A
r
f
a
n
S
h
a
h
z
a
d
.
c
o
m

16. DiscreteSimulationModelscont… Model
Time:KeyAspects
•TimeUnits:Modeltimemaybeexpressedinvariousunits,suchas
seconds,minutes,hours,days,orotherrelevanttimeintervals,
dependingontheapplicationdomainandthegranularityofthe
simulation.
A
r
f
a
n
S
h
a
h
z
a
d
.
c
o
m

17. DiscreteSimulationModelscont… Model
Time:ApplicationsofModelTime
•DynamicSystemBehavior:Modeltimeenablesanalyststostudythe
dynamicbehaviorofsystemsovertime,capturingthesequenceof
eventsandtheireffectsonsystemstate.
A
r
f
a
n
S
h
a
h
z
a
d
.
c
o
m

18. DiscreteSimulationModelscont… Model
Time:ApplicationsofModelTime
•PerformanceAnalysis:Bytrackingtheprogressionoftime,analysts
canevaluatesystemperformancemetrics,suchasthroughput,
responsetimes,andresourceutilization,providinginsightsinto
systemefficiencyandeffectiveness.
A
r
f
a
n
S
h
a
h
z
a
d
.
c
o
m

19. DiscreteSimulationModelscont… Model
Time:ApplicationsofModelTime
A
r
f
a
n
S
h
a
h
z
a
d
.
c
o
m
•ScenarioComparison:Modeltimefacilitatesthecomparisonof
differentscenariosorpoliciesbysimulatingthebehaviorofthe
systemundervaryingconditionsandobservingtheeffectsonsystem
performanceandoutcomes.

20. DiscreteSimulationModelscont…
A
r
f
a
n
S
h
a
h
z
a
d
.
c
o
m
•Discretesimulationmodelsandmodeltimeareessentialconceptsin
thefieldofsimulationmodeling,enablinganalyststostudythe
behaviorofcomplexsystemsovertime.

21. Any Question?