Deck 13: Capacity Planning and Queuing Models

Question 1

In analytical terms, the modern self-serve supermarket would be approximated best by the M/M/C model, where C is the number of servers.

Answer 1

False

Question 2

A system is said to be in statistical equilibrium when the number in queue assumes a distribution that is independent of the starting condition.

Answer 2

True

Question 3

Excess capacity is required in a service system because variability in customer arrivals and service times creates idle capacity.

Answer 3

True

Question 4

When Xerox Corporation introduced the Model 9200 Duplicating System, the level of service dipped because technical representatives were assigned to territories.

Question 5

In the M/M/1 model the calling population is finite.

Question 6

Capacity usually is measured in terms of outputs (e.g. guest nights) rather than inputs (e.g. number of hotel rooms) in service firms.

Question 7

A system is said to be in a transient state when the values of its governing parameters in this state are expected to change uncontrollably with time.

Question 8

The main advantage of analytical models is that they can be used almost anywhere regardless of the underlying assumptions. They still will provide good results.

Question 9

For a system to reach steady state, the arrival rate must always be greater than the service rate, i.e., the traffic intensity SYMBOL 114 \f "Symbol" must always be greater than unity.

Question 10

A Poisson process describing the distribution of departures from a busy server with an average rate of 15 per hour is equivalent to the negative exponential distribution of service times with a mean of 5 minutes.

Question 11

The average time a customer should expect to wait can be calculated using just the mean arrival rate and the mean service rate.

Question 12

The graph depicting the economic trade-off in capacity planning has cost on the x-axis and the capacity to serve on the y-axis.

Question 13

Planning for service capacity involves prediction of consumer waiting associated with different levels of capacity.

Question 14

A queue is said to be finite if there is an upper limit on the number of consumers who are allowed to be in it. Consumers who come after the queue is full, either balk or are refused entrance.

Question 15

For a standard service system to operate over the long run, the servers must be idle part of the time.

Question 16

Pooling of services is an approach toward achieving economies of scale in services through better utilization of services.

Question 17

Reducing the variability either in service time or arrival time can reduce most of the waiting involved for the consumers.

Question 18

In the A/B/C classification of analytical queuing models, the term G refers to a general distribution with mean and variance.

Question 19

Analytical queuing models are classified mainly on the basis of the calling population characteristics and the queue discipline.

Question 20

Capacity planning decisions deal implicitly with decisions on the cost of making consumers wait and the extent to which these costs can be borne.

Question 21

In the A/B/C analytical queuing model classification, M is used to denote:

A) exponential inter-arrival time distribution.
B) deterministic inter-arrival time distribution.
C) general distribution with mean and variance.
D) Erland distribution.

Question 22

Pseudo-random numbers do not permit an exact replication of experimental conditions.
Multiple Choice

Question 23

A queuing system is described by all but one of the following:

A) balking behavior
B) calling population
C) arrival process
D) queue configuration

Question 24

Simulation allows time compression, doing in seconds what would otherwise require years of actual observation.

Question 25

In general, for infinite queues, the expected number of people in the system is equal to:

A) the average arrival rate minus the average service rate.
B) the expected number in the queue plus the expected number in service.
C) the expected arrival rate times the expected waiting time.
D) alternatives (a) and (b) above.

Question 26

A queuing system is said to be in a transient condition when:

A) its parameters change with time.
B) its parameters remain constant.
C) it is still under the influence of the initial state.
D) it grows beyond bound.

Question 27

Which one of the following is not part of the queuing system features?

A) calling population
B) queue configuration
C) queue discipline
D) departure process

Question 28

The A/B/C notation classifies queuing systems on the basis of:

A) arrival rates and queue discipline.
B) queue discipline alone.
C) arrival rates and service rates.
D) arrival rates, service rates, and number of servers.

Question 29

Service capacity decisions have all but one of the following strategic characteristics:

A) the decision to invest in capacity will attract competitors.
B) financial investment in capacity must be balanced against the costs of lost sales.
C) inadequate short-term capacity needs can generate customers for the competition.
D) hostile customers can result from inadequate capacity following a marketing campaign.

Question 30

In the long run, if service capacity is inadequate for the arrival rate, one of the following will likely occur:

A) servers will slow down.
B) customers will demand less.
C) reneging will occur.
D) prices will be increased.

Question 31

The cost of service capacity is determined by

A) the number of customers waiting.
B) the total number of customers.
C) waiting time.
D) the number of servers on duty.

Question 32

Simulation can be viewed as a laboratory for experimenting with service systems.

Question 33

The ___________ queuing model is often used in situations when little if no waiting occurs.

A) M/M/1
B) M/M/c
C) M/G/1
D) M/G/ $\infty$

Question 34

Verification refers to determining whether or not the structure of a simulation model is consistent with the real system.

Question 35

All but one of the following need to be analyzed for capacity requirements when setting up a "Cookies and Ice Cream" shop:

A) technology requirements.
B) facility requirements.
C) equipment requirements.
D) labor requirements.

Question 36

A shortcoming of simulation is the use of pseudo-random numbers.

Question 37

Little's Law states that a relationship exists between

A) line length and waiting time.
B) multiple and single queues.
C) waiting in queue and waiting in system.
D) exponential time between arrivals and Poisson arrival rate.

Question 38

For the system to operate satisfactorily over the long run,

A) the capacity to serve must be greater than the average demand.
B) the expected waiting time for each customer must not exceed average service time.
C) the arrival rate must be less than the service rate.
D) the number of servers must equal the average number of arrivals per unit time.

Question 39

Queuing models can help explain all but one of the following:

A) predicting results of adding servers to a multiple server system.
B) effect of reducing service time variation on waiting time.
C) determining the size of a parking lot.
D) mean arrival rate.

Question 40

Pooling of servers results in reduced waiting time because:

A) the mean service time for the pool as a whole is now constant.
B) servers are utilized better.
C) multiple queues are replaced by a single queue.
D) alternatives (a) and (b) above.

Question 41

In the equation L_S = L_Q + ρ, ρ refers to:

A) number of servers.
B) traffic density.
C) mean number of customers being served.
D) number of customer that have been served.

Question 42

All but one of the following are advantages of using one large computer system to serve an entire university community:

A) short turnaround time.
B) large memory.
C) economics of scale in services.
D) short wait in queue.

Question 43

Which of the following is not true of pseudo-random numbers?

A) They exhibit the characteristics of true random numbers.
B) They are generated using a mathematical function.
C) They are stored in a computer file.
D) An identical sequence can be replicated with the same seed value.

Question 44

A parking lot is a special variation of the _________ queuing model.

A) infinite queue M/G/∞
B) finite-queue M/M/c
C) infinite queue M/M/c
D) finite-queue M/G/1

Question 45

The symbol ρ is used to denote which one of the following parameters:

A) number of customers in the system.
B) number of servers.
C) traffic intensity.
D) finite size of queue.

Question 46

Which of the following is not a criterion for capacity planning?

A) average customer waiting time
B) probability of excessive waiting
C) the rate of departure
D) probability of sales lost because of inadequate waiting area