Scribd
Upload
42 views

Example Questions For Queuing Theory and Markov Chains: Ations Research

This document provides 11 problems related to queuing theory and Markov chains. The problems cover topics such as deriving formulas for expected queue length and wait time, calculating probabilities and expected times for single and multi-server queues, proving properties of exponential distributions, modeling argument chains as Markov chains, and more. Additional relevant solved problems can be found in other textbooks.

Uploaded by

Ronnie
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
42 views

Example Questions For Queuing Theory and Markov Chains: Ations Research

This document provides 11 problems related to queuing theory and Markov chains. The problems cover topics such as deriving formulas for expected queue length and wait time, calculating probabilities and expected times for single and multi-server queues, proving properties of exponential distributions, modeling argument chains as Markov chains, and more. Additional relevant solved problems can be found in other textbooks.

Uploaded by

Ronnie
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Example Questions for Queuing Theory and

Markov Chains
Read: Chapter 14 (with the exception of chapter 14.8, unless you are in-
terested) and Chapter 15 of Hillier/Lieberman, Introduction to Oper-
ations Research

Problem 1: Deduce the formula Lq = λWq intuitively.

Problem 2: A two-server queueing system is in a steady-state condition


1 4 6
and the steady state probabilities are p0 = 16 , p1 = 16 , p2 = 16 , p3 =
4 1
16 , p4 = 16 , pn = 0 if n > 4. Calculate

1. L (the expected number of customers in the system) and Lq


(the expected number of customers in the queue) (Answer: L =
2, Lq = 3/8),
2. the expected number of customers being served (Answer: 13/8).

Suppose the arrival rate is 2 customers per hour. Calculate

1. W (the mean waiting time in the system) and Wq (the mean


waiting time in the queue) (Answer: W = 1 hr, Wq = 3/16 hrs),
1
2. µ (the mean service time per customer) (Answer 13/16 hrs/customer),
3. ρ (the traffic intensity) (Answer: 13/16).

Problem 3: Verify the following relations for a single server queue:

1. L = Lq + 1 − p0 ,
2. L = Lq + ρ,
3. p0 = 1 − ρ.

Problem 4: Show that the chance that an exponentially distributed vari-


able takes on a value below its mean is more than 60% and that the
chance that the value will be below half of the mean is almost 40%.

Problem 5: When are exponentially distributed interarrival (service) times


reasonable/unreasonable? Discuss this questions and provide exam-
ples.

1
Problem 6: Suppose a queueing system has two servers, exponential inter-
arrival times with mean of 1 hour, and exponential service times with
mean of 1 hour per customer. Suppose a customer has just arrived at
12.00 noon.

1. What is the probability that the next arrival will come before
1.00 pm (between 1.00 pm and 2.00 pm, after 2.00 pm)? (Answer:
0.6321, 0.2325, 0.1353, resp.)
2. Suppose no customer arrives before 1.00 pm. What is the prob-
ability that the next arrival will come between 1.00 pm and 2.00
pm? (Answer: 0.6321)
3. What is the probability that the number of arrivals between 1.00
pm and 2.00 pm will be zero (one, more than one)? (Answer:
0.3679, 0.3679, 0.2642, resp.)
4. Suppose that both servers are serving customers at 1.00 pm.
What is the probability that neither customer will have service
completed before 1.01 pm (before 1.10 pm, before 2 pm)? (An-
swer: 0.9672, 0.7165, 0.1353, resp.)

Problem 7: A dental surgery has two operation rooms. The service times
are assumed to be independent, exponentially distributed with mean
15 minutes. Andrew arrives when both operation rooms are empty.
Bob arrives 10 minutes later while Andrew is still under medical treat-
ment. Another 20 minutes later Caroline arrives and both Andrew and
Bob are still under treatment. No other patient arrives during this 30-
minute interval.

1. What is the probability that Andrew will be ready before Bob?(Answer:


1/2)
2. What is the probability that Caroline will be ready before An-
drew?(Answer: 1/4)
3. What is the probability that Caroline will be ready before Bob?
(Answer: 1/4)
4. Determine the distribution function of the waiting time in the sys-
tem for Caroline. Find the mean and variance. (Answer: F (t) =
1 − 2e−λt + e−2λt , Exp.=22.5 minutes, Var.=(16.8)2 minutes2 )

2
Problem 8:

a) Prove that the exponential distribution has both the lack of mem-
ory and the minimum property.

b) What are the characteristic properties of the Poisson input process?


Discuss their limitations by giving examples from practical queueing
systems.

c) Suppose there are 2 types of customers arriving at the same server


according to independent Poisson input processes. Show that the ag-
gregated arrival process (of customers without regard of type) is a
Poisson process.

Problem 9:

Complicated arguments often involve series of implications of the form


”A0 therefore A1 therefore A2 therefore...”, where the Ai ’s are propo-
sitions. The truth of any proposition in such an argumentation chain
depends on the validity of the foregoing implications. Often such im-
plications are not watertight but only true with a high probability. For
the purpose of this question suppose if An is true then An+1 is true
with a 90% chance while if An is false then An+1 is true or false with
equal probability (”ex falso quodlibet”).

1. Formulate the change of the truth values of the propositions as a


Markov chain and give the transition matrix
2. What does the Markov property mean in this setting? Discuss
this assumption. When does it apply?
3. What is the probability that A4 is true, provided A0 was true
with 50% probability (0.8248).
4. Calculate the stationary probabilities and interpret them ((1/6, 5/6)).
5. If an adversary checks the argumentation chain and finds one
false statement, then she would argue that the argumentation
chain is invalid and would reject any further conclusions. We
may therefore say that an argumentation chain breaks down if

3
the first false statement is made. Change the Markov chain to
model this situation.
6. Calculate the probability that the argumentation chain is still
valid after the fourth statement, provided the initial statement
was true (0.6561).
7. Calculate the stationary probabilities for the new Markov chain
and interpret them ((1, 0)).
8. Calculate the probability that, starting from a true proposition,
the argumentation chain breaks down after precisely 3 iterations
(8.1%).
9. The adversary of part 5 will only examine the argumentation
chain if she can be 50 % sure that she will find a false proposition.
How long does the argumentation chain have to be in order for
this to be the case, provided A0 is true. Use a spreadsheet to
make your calculations. (7 implications)
10. Calculate the expected number of statements until the argumen-
tation chain breaks down if the initial statement is true. (10
implications)
11. Give a formula for the expected number of statements until the
argumentation chain breaks down in terms of the probability p
that the next statement holds, provided the current statement is
true.

Additional relevant solved problems can be found in Chapters 19-24


of Richard Bronson, Schaum’s outline of theory and problems of oper-
ations research (to be available on the G12 shelf soon).

You might also like