BGSU BA 3800 Chapter 12. Computer Simulation: Basic Concepts

To generate random observations from a normal distribution, an analyst must know

- the mean - the standard deviation

Which of the following statements are TRUE?

-Simulation programming languages require less programming time but are less flexible - General - purpose programming languages offer high levels of flexibility but require more programming time

In a computer simulation of a queuing system, which of the following measures of performance can be gathered?

-average waiting time -maximum waiting time -number of customers in the system

The Queuing Simulator included with this text

-can use the uniform distribution for service times -can use the exponential distribution for arrivals -can have up to 25 simulated servers

When simulation results are presented to management, the information presented should include

-numerical results of the simulation -documentation of the validity of the simulation model -how the study was conducted

A computer simulation of a stochastic system uses a computer to

-record the performance of the simulated system -generate random occurrences of probabilistic events

To generate random observations from a uniform distribution, it is necessary for the analyst to know

-the minimum value -the maximum value

In a simulation, the sum of the probabilities of all possible outcomes for a particular even must be equal to

1

In Excel, a random observation from a uniform distribution with a known mean value (u) can be generated using the formula

=-u *LN(RAND())

The Excel function that generates a uniformly distributed number between 0 and 1 is

=RAND()

True or False: Because of the complexity of simulations, recommendations are generally presented to management without additional information.

False

True or False: Computer simulations are always more useful that mathematical models.

False

True or False: Since simulations are only used for very complex situations, testing the model for validity is not possible.

False

Match the simulation study step with its definition

Formulate the Problem and Plan the Study - Establish overall objectives of the simulation Collect the Data and Formulate the Simulation Model - Determine the distribution of inter arrival times and service times. Determine how to generate random observations of these values Check the Accuracy of the Simulation Model - Conduct a structured walk-through of the proposed simulation model. Plan the Simulation to be Performed - Determine how to obtain statistically valid results for the performance measures.

Place each characteristic in the appropriate category

Mathematical analysis: May require simplifying assumptions that limit usefulness. Provides precise results. Computer simulation: Inexpensive and quick approach to testing proposed systems. Imitates performance of a real system. Physical Prototype: Often expensive and time consuming. Provides exact model of the proposed system.

Match each approach to management science modeling with the best description.

Mathematical models - Precise solutions that reveal the underlying structure of the system. Computer simulations - useful for complex systems that are difficult to describe precisely

True of False: One way to track the results of many different random numbers when they are used as the input of a model is to use a data table in Excel.

True

Simulation models of stochastic system include all of the following EXCEPT

a real time clock

A system that is always changing because one or more important elements are probabilistic is

a stochastic system

In a simulation of a stochastic system, the simulation clock

can run much faster than real time

_________ __________ give a range where the true value is likely to be located.

confidence interval

One way to track the results of many different random numbers when they are used as the input of a model is to use a ______ table in Excel.

data

When values follow a uniform distribution

every value in the range is equally likely to occur

Compared to mathematical analysis, the use of computer simulations is

less precise but more flexible

In a simulation, the simulation clock is advanced whenever an event occurs. This is known as

next-event time advance

A stochastic system has one ore more important elements that are

probabilistic

A computer simulation of a stochastic system generates __________ events and measures how the system would perform.

random

The inverse transformation method can be used to generate _____ observations from continuous or discrete distributions.

random

When simulation results are presented to management, the information presented should include how the study was conducted, numerical ___________ of the simulation and documentation of the __________ of the simulation model.

results, validity

A computer ____________ uses a computer to imitate the elements of a system

simulation

Instead of building a physical prototype of a proposed system, a firm could use a computer __________ to investigate the process more quickly and at a lower cost.

simulation

In a simulation model of a stochastic system, the _______ of the system is updated each time an even occurs.

state

In a simulation, the _______ of the probabilities of all possible outcomes for a particular event must be equal to 1.

sum

A simulation of a queuing system can be checked for validity by

using inputs that cause the simulation to model a queuing system that has an analytical solution and then comparing results

After constructing a simulation model, the _________ of the model should be tested to ensure accuracy of results

validity