Professional Documents
Culture Documents
8/20/2017
12
S IMULATION M ODEL T YPES
(T IME A NGLE )
Static representation of a system at a
particular time
Monte Carlo models
Robots in a factory
8/20/2017
13 M ONTE C ARLO S IMULATION
8/20/2017
15
S IMULATION M ODEL T YPES
(G RANULARITY A NGLE )
Discrete modeling of system as it evolves over
time where state variables change
instantaneously at specific and separate points
in time
Continuous modeling of system as state
variables change continuously with respect to
time
Certain cases when aggregation happens, then
discrete systems become continuous
8/20/2017
16
C OMPONENTS O F D ISCRETE
E VENT S IMULATION
System State collection of state variables to describe system
System Clock current value of simulation time
Event List (Calendar) next time when each event type will occur
Statistical Counters statistical information on system performance
Initialization Routine start simulation at time = 0
Timing Routine advance time to the next event from calendar
Event Routine update system state when particular event occurs
Library Routine generate random variates
Report Generation compute estimates of performance measures
Main Program invoke timing routine & check for end simulation
8/20/2017
17 A DVANTAGES O F S IMULATION
8/20/2017
19 P ITFALLS O F S IMULATION
Absence of well-defined objectives before study
Inappropriate model detail
Misunderstanding of simulation by management
Treating simulation as a computer programming exercise
Failure to collect and analyze good system data
Belief on easy-to-use simulation package that requires no
programming
Misuse of animation
Failure to use accurate distributions and analysis methods to account
for randomness and independence
Comparing results based on one run of simulation
8/20/2017