COMPUTER SIMULATION Flashcards
answer the question “What if?”
computer simulation
answer the question “What is?”
computer simulaton
enables studies of more complex systems because it creates observations by “moving forward” into the future
computer simulation
involves modeling the organizational system as
a set of entities evolving over time according to the availability of resources and the triggering of events.
Discrete event simulation
involves identifying the key “state” variables that define
the behavior of the system, and then relating those variables to one another through coupled, differential equations.
System dynamics
involves agents that attempt to maximize their
fitness (utility) functions by interacting with other agents and resources
Agent-based simulation
seven purpose of simulation
prediction, performance, training, entertainment, education, proof, theory discovery
By comparing different output
obtained via different structures and governing rules, researchers can infer what might
happen in the real situation if such interventions were to occur.
prediction
s a substitute for experimentation and intervention on
the actual system.
Simulation for prediction
undertaken when such experimentation is too dangerous, costly, untimely, or inconvenient
Simulation for prediction
Simulation can uncover phenomena that in turn focus theoretical attention
Theory discovery
simulation can be used to perform real tasks for an organization, such as diagnosis or decision-making.
performance
A simulation environment makes it quick, easy, and safe for users to make
decisions that mimic the decisions they (will) make in reality.
training
three different approaches to simulation
discrete event simulation, system dynamics, ad agent-based simulations
are best used when the organizational system
under study can be adequately characterized by variables and corresponding
states, and events occur that change the value of these variable states in some
rule-oriented but stochastic manner.
Discrete event simulation
not appropriate
when state variables interact with one another and change on a continuous basis,
Discrete event simulation
best fit for situations
where the variables in question are numerous, and can be related to one another in
terms of how their rates of change interact with one another.
System dynamics
tend to treat systems rather mechanistically
System dynamics
a “top-down” modeling
approach, and thus requires fairly extensive knowledge about how the state
variables of the system interact with one another.
System dynamics
best fit for situations when the organizational
system is best modeled as a collection of agents who interpret the world around
themselves and interact with one another via schema
agent-based simulations
emphasize change in agents’ schema via learning and adaptation, and also
highlight the phenomena of emergent
agent-based simulations
considered “bottom-up” models
in that one describes individual agents and their patterns of connectivity and
interaction
agent-based simulations
