Review from Last Time Motivations to do simulations Modeling - PDF document

Review from Last Time � � Motivations to do simulations � � Modeling characteristics CSCI 8220 Simulation & Modeling � � Time and event driven simulations Process Oriented Simulation 2 Maria Hybinette, UGA Maria Hybinette, UGA Today Event-Oriented World View Event handler procedures state variables � � � Event-Oriented Simulation (review) Departure � Arrival � Landed � Integer: InTheAir; � � � Process-oriented simulation Integer: OnTheGround; � Event � Event � Event � Boolean: RunwayFree; � { � { � { � » � Fundamental concepts: Processes, resources … � … � … � » � Simulation primitives } � } � } � Simulation application � » � Example Simulation executive � » � Implementation Event processing loop while(simulation not finished) � Now = 8:45 � � E = smallest time stamp event in PEL � Pending Event List (PEL) � � Remove E from PEL � 9:00 � 10:10 � � Now := time stamp of E � 9:16 � � call event handler procedure � 3 4 Maria Hybinette, UGA Maria Hybinette, UGA Example: Event-Oriented Air traffic Execution Example Simulation State � R=3 � Variables � Now: current simulation time � G=4 � InTheAir : number of aircraft landing or waiting to land OnTheGround : number of landed aircraft RunwayFree : Boolean, true if runway available InTheAir � 0 � 1 � 2 � 1 � 0 � Arrival Event: � OnTheGround � 0 � 1 � 2 � 1 � 0 � InTheAir := InTheAir+1; � if( RunwayFree ) � RunwayFree � true � false � true � � RunwayFree:=FALSE; � 0 � 1 � 2 � 3 � 4 � 5 � 6 � 7 � 8 � 9 � 10 � 11 � � Schedule Landed event @ Now + R; � Simulation Time � Landed Event: � Processing: � Arrival F1 � Arrival F2 � Land F1 � Land F2 � Depart F1 � Depart F2 � InTheAir := InTheAir-1; � Time � Event � Time � Event � Time � Event � Time � Event � Time � Event � Time � Event � Time � Event � OnTheGround := OnTheGround + 1; � 1 Arrival F1 � Schedule Departure event @ Now + G; � 3 Arrival F2 � 3 Arrival F2 � if( InTheAir > 0 ) Schedule Landed event @ Now + R; � 4 Land F1 � 4 Land F1 � 7 Land F2 � else RunwayFree := True; � 8 Depart F1 � 8 Depart F1 � Departure Event: � 11 Depart F2 � 11 Depart F2 � OnTheGround := OnTheGround - 1; � Now=0 � Now=1 � Now=3 � Now=4 � Now=7 � Now=8 � Now=11 � 5 Maria Hybinette, UGA

Event-Oriented World View Process Oriented Event handler procedures state variables � � � A simulation process models a specific entity Departure � Arrival � Landed � Integer: InTheAir; � with a well defined behavior. Integer: OnTheGround; � Event � Event � Event � Boolean: RunwayFree; � » � It describes the action performed of the process { � { � { � through out its lifetime. … � … � … � – � Models a specific entity with well defined behavior } � } � } � Simulation application � and it is encapsulated within the process. � � Event-oriented simulation programs may be difficult to – � Example: an aircraft understand and modify: � � Event oriented view: lifetime of an event is a » � Program organized around state transitions SINGLE instant in time. » � Behavior of an aircraft distributed across multiple event handlers � � Process oriented view: lifetime is a time » � Flow of control among event handlers not obvious period of the ‘process’ or ‘thread’ » � Suppose you want to model: Different aircrafts, airlines, pilots – imagine events for each segment (volume) of airspace 7 8 Maria Hybinette, UGA Maria Hybinette, UGA Event versus Process Oriented Views Process Oriented Execution Model Event Oriented View � � Focus simulation program around behavior of entities state variables � » � Aircraft: arrives, waits for runway, lands, departs Arrival � Landed � Departure � Integer: InTheAir; � Event � Event � Event � � � Process-oriented simulation Integer: OnTheGround; � Boolean: RunwayFree; � { � { � { � » � Process : Thread of execution describing entity behavior over time … � … � … � » � Resources : Shared resource used by entities (e.g., the runway) } � } � } � � � Execution: alternate between Focus of model is on EVENTS and how they affect the state of the simulation. » � simulation computations at a single instant of simulation time, and Process Oriented View » � advances in simulation time (no computation) Simulation time advances state variables � Aircraft1 � Aircraft2 � AircraftN � Computation at a single (no computation) Integer: InTheAir; � Instant of simulation time { � { � { � Integer: OnTheGround; � Arrive � Arrive � Arrive � Boolean: RunwayFree; � Land � Land � Land � Computation Time advance Computation Time advance Depart � Depart � Depart � } � } � } � 9 10 Entities modeled by processes. Wall clock time Maria Hybinette, UGA Maria Hybinette, UGA Simulation Primitives Process Model Example: Aircraft Primitives needed to advance simulation time A new aircraft process is created with each Arrival event � � AdvanceTime( T ) : advance T units of simulation time /* simulate aircraft arrival, circling, and landing */ » � Also called “hold” Integer: InTheAir; Integer: OnTheGround; » � Example: AdvanceTime(R) to model using runway R units Boolean: RunwayFree; of simulation time � � WaitUntil( p ) : simulation time advances until predicate 1 InTheAir := InTheAir + 1; p becomes true 2 WaitUntil( RunwayFree ); /* circle */ » � Predicate based on simulation variables that can be modified 3 RunwayFree := FALSE; /* land */ by other simulation processes 4 AdvanceTime( R ); » � Example: WaitUntil(RunwayFree) to wait until runway 5 RunwayFree := TRUE; becomes available for landing /* simulate aircraft on the ground */ 6 InTheAir := InTheAir - 1; � � Other combinations 7 OnTheGround := OnTheGround + 1; » � WaitUntil(p,T) : Wait up to T units of simulation time for 8 AdvanceTime( G ); predicate p to become true /* simulate aircraft departure */ » � Not used in the air traffic example 9 OnTheGround := OnTheGround - 1; 11 12 Maria Hybinette, UGA Maria Hybinette, UGA