POPULAR PUBLICATIONS

ros rua
QUEUING MODELS

Multiple Choice Type guestions

1. The steady state behaviour of the queueing system which is in operation for a
sufficient long period of time exists, it WBUT 2006, 2008, 2010, 2014, 2016
a) A< b) 2= c) A>p d) S
where 7 is the mean arrival rate and p is the mean departure rate.
Answer: (a)

2. If 2 is the arrival rate, u is the service rate, then the average waiting time of a
customer (in the queue) is given by WBUT 2010
(4-2)e (u+2) u-2) nai u-2
Answer: (a)

3. In an (M/M/1): (o/FIFO) model, the average number of customers E(n) is

given by WBUT 2011
a) p
Answer: (b)
bP 1-p
d)
1+P
4. What is the distribution of service time in a M/M/1 model?
a) Binomial
WBUT 2012(0DD)]
b) Poisson c) Normal d) None of these
Answer: (6)

5. When there are more than one server, customer behaviour in which he
from one queue to another is known as moves
[WBUT 2013]
a) balking b) jockeying c) reneging d) alternating
Answer: (a)

6. In (M/M/1): (N/FCFS) model, the steady state solution of the

queuing systen
exists even for A2p because arrivals at the system are controlled
by the syste
limit N not by the relative rates of arrival A and
departure WBUT 2014
a) True b) False
c) Not known
d) True only for A> 100
Answer: (6)

7. If in queuing theory p>1, then it is known as WBUT 2015]

a) queue elongation b) queue shortening
c) constant queue
d) queue burst
Answer: (a)

ORCE-134
 OPERATION RESEARCH

in a queu euing system is the average rate of custom arrival and is the
8. If
service rate then
[WBUT 2015]
a) A>u lev b) As c) =4 d) A+4=l
Answer:

9. Aqueuing model 1s generally expressed using [WBUT 2016]

a) Newton's Notation b) Fulkerson's Notation
c) Dantzing's Notation Kendall's Notation 2at t or
Answer: (d)

10. Multiple serves may be [WBUT 2017]

a) in parallel
t not b) in series
c) in combination of parallel and series d) none of these
Answer: (c)

Short Answer Type Questions

1. In a supermarket, the average arrival rate of customers is 10 every 30 minutes
following Poisson process. The average time taking by a cashier to list and

calculate customer's purchase is 2.5 minutes following exponential

the
exceeds 6 (six)? What is
distribution. What is the probability that the queue length
WBUT 2010]
the expected time spent by a customer in the system?
Answer:

Mean arrival rate of customer 1=10/30 =1/3 per minute.

Mean service rate of customer = per minute.

2.5
3 0Yo un betuoa os bn
erimunbe
mela2 p=2/u-25
1/25
zation factor=
The utilization factor =P= /2.56daeo ele ybueto srd i ctp ref
exceeds 6
) Probability that queue length
27]= = 0.279 nusbo
=Prob. queuelengthaborm A
of a customer in the system
(1) Expected waiting time
1 - 2 minutes.
) 3-2.5 0.5
youdo by (M/M/1):
mean (co/FIFO)?
Theory, what
2. With reference to Queuing
and "steady state" of queuing model? Write
"transient state"
What do you mean by WBUT 2010]
the formula for L, and L

ORCE-1335
POPULAR PUBLICATIONS
Answer: v1Te lo, 2e
1s Part:
(M/M/1/o-FIFO) is the simplest queue system where the arrivals follow
a Poissor
process, service times exponentially distributed and there is onls
are nly one server.
words, it is a system with Poisson input, exponential waiting time and Poio In ot
me and
Poisson output
with single channel.
Queue capacity of the system is infinite with first in first out mode. The fires
st Min the
notation stands for Poisson input, second M for Poisson output, 1 for the
servers and u for infinite capacity of the system.

2nd Part:
A system is said to be in transient state when its operating characteristics are d
on time. penden
A steady state condition is said to prevail when the behaviour of the system hero
independent of time.
ecomes

3d Part:
Formulas

Average number of units in the system (L) =

S 4 -

Average queue length (L,) = -

AA-2)

LLong Answer Type Questions

1. Find the expected number of customers waiting in the queuing system L,, When

the queue is in the steady state corresponding to the model

(M/MI):(o/FCFS|.
WBUT 2006, 2014
Answer:
Model I: (M/M/1): (0 /FCFS) [Birth and Death Model]
This model is also referred to as birth and death model. dr
To obtain the system
of steady state equations:
The probability that there will be n units (n> 0) in the system at time (t + maybe
Au the
expressed as the sum of four independent compound probabilities by U
fundamental properties of probability, Poisson arrivals and exponential service tu
The following are the four cases:
S. No. of units in the units in the
No. of units No. of units served ofuni
No. of
No. system in time t arrivedin timeAt duringtime At systemintime
1. 0 0
2.
3. ntl
n-1 Il

ORCE-136
 OPERATION RESEARCH
The product
of the probabilities for
that there
,

are n
units in the
system at time t is P().
thatthere is no arrival in time At is
and that there is no service in time AtPo(At)=1-AAt).
is 1- uAt ,is given by
P,CX1-AArXl - uAr)= P.1-(2+uar]+O(ar)
2. The product of the probabilities for
.that there are n units in the
system at time t is Pn(t)
that there is 1 unit arrived in time At is AAt
and that there 1 unit served in time At is (uAt), is
given by
P,XaAruAr)=o(ar)
3. The product of the probabilities for
that there is (n +1) units in the system is Pn + 1)
that there is no arrival in time At is (1 -At)
and that there is one unit served in time At is (LuAt), is given by
Pr.1-AIXuAr) =P.uAr)+ O(Ar)
4. The product of the probabilities for
that there is (n 1) units in the system is Pa-1()
-

that there is one arrival in the system in time At is (4t)

and that there is no service in the system in time At is (1 uAt), is given by
Pp-Aar)X1-uAI)= P,XAar)+O(Ar)
Adding the above four independent compound probabilities above, we obtain the
probability of n units in the system at time (t + At), i.e.,

P.(+r)-P.)1-(2+u)ar]+P()AAr+P, ) uas+O(4r)
=-(3+4)P.(1)+AP (t) +uPt)+a)
At Af
lim2+ Ar)-P.(t)
Ar-+0 At

+AP()+uP., (t)+ O(Ar)

= lim-(2+u)P,(1)
Ar-0 Ar

--(2+z)P.(0)+ AP 0)+4P,)+
dt
0)

n> and O(Ar) 0as Ar>0, as O(A) is of order higher than the 1st
At
 OPERATION RESEARCH

P Po, n20...6)

Since, PoP,P2, are probabilities, thenP. =1

n=0

P+ =1

+ 1

1
P weffo

1-
P-1-=1-p, wherep (7) g arus
From (6) and (7), we get P, =
p"(1-p) with p <l and n 2 0..(8)
Thus, equations (7) and (8) give the required probability distribution of queue length.

Measures of Model I
1. Expected (average) number of units in the system (L):
n-1

L-. -{E1-EGE

-Pp-<l. L, =

. In a public telephone booth the arrivals are on the average 15 per hour. A call on
ne average takes 3 minutes.
If Is Just
there time phone, find (i) the expected
one the
and () proportion of the time the
numbers of callers in the booth at any
ooth is expected to be idle. [WBUT 2006, 2016]
ORCE-139
POPULAR PUBLICATIONS

Answer:
) mean arrival rate (A)=15callers/h
mean service rate (u)=20callers/h
traffic intensity (p)= 0.75

a) Expected no. of callers in the booth at any time (L,)= 3callers

-P
b) The booth will remain idle ofthe total time.

3. In a railway marshalling yard, goods trains arrive at the rate of 30

slday.
Assuming that inter-arrive time follows exponential distribution and senvice
Rate distribution is also exponential with an average 36 minutes. Calculate a
following
Average number of train in the yard
ii) Probability if the queue exceeds 10.
OR,
WBUT 2007,201
In railway marshalling yard, goods trains arrival time follows an exponenia
a
distribution and the service time (the time taken to load a train in the hump
distribution is also exponential with an average of 36 minutes.
yard
average number of trains in the queue.
Calculatethe
WBUT 2013
Answer:
Here, =- 301 trains/minute and u= trains/minute
60x 24 48
) Here, p==0.75
Expected number of trains in the yard (system)=L, p/1-p)= 0.75/0.25 =3 tras
=

(i) Probability (queue size exceeds 10)

-
Probability (queue size 2 11)
=1-Probability (queue size < 11)
=1-(P,+R+Pt. .+ Po)
1-
=

P,(1+p+p' t. .tp")

=1-0-p)P(since
(-P) P=1-p)
=
(1-p") =p" (0.75)" =0.042
1- =

(Similar to example-1 of Chapter: Queuing models)

ao

ORCE-140
 OPERATION RESEARCH

4.A ch of a nationalized bank has only one typist. Since typing work varies
branch ributed in
4. gth (number of pages to be typed), the typing rate is omly d i s t r i b u t e
randomiy
jengimating a Poisson distribution with a mean service rate of 8 letters per hour.
apletter arrives at a rate of 5 per hour during the entire 8-hour work- day. If the
e t is valued at RS. 1.50 per hour, determine: t e e pho p r
ypis

i) Equipment utilization,
The percent time an arriving letter has to wait,
til Average system time and e begyi
iv Average idle time cost of the typewriter per day. nuued WBUT 200
Answer:

5 per hour
Arrival rate (2)
=

8 per hour.
Service rate (u)
=

Equipment utilization (p) = 0.625 o teo er it lo opae

time that an arriving letter has to wait =

the percent time that the type
in The percent
writer remains busy =62.5%.
1
1 hour 20 minutes. sun
ii) Average system time (W) = -
8x(1-)x1.50 Rs.4.50
of the type writer per day
=

iv) Average Idle time cost

time spent on his jobs has an exponential

5. T. V. repairman finds that the sets in the order in which they
distribution with mean 30 minutes. If he repairs rate of
come in and if the arrival of sets
is approximately Poisson with an average
repairman's expect idle time each dayy?
10 minutes per 8-hours day, what is the
set just brought in?
How many jobs are ahead of the average [WBUT 2008, 2010, 2011]
OR,
time spent on his jobs has an exponential
A television repairman finds that the order in which they
with mean 30 mins. if he repairs the sets in the
distribution a
with an average rate
in and if the arrivals of sets are approximately Poisson each
come the repairman's idle time day? Find the
of 10 per 8 hrs a day, what is
the system and in
the queue? [WBUT 2017]
expected number of units in
Answer:
10 1
Here, u 130 8x 60 48
Therefore, expected no. ofjobs on the system=Ls

L, 1-3/4-2= 48 jobs
1-2/u 30 48
No. ofjobs ahead to the set brought= 5/3 is busy (i.e., the traffic intensity) is
is equal
equal to
to
the repairman
Since the
nce the fraction of the time,which the repairman remains busy in a 8-hour day is
A/4), number of hours for
ORCE-141
POPULAR PUBLICATIONS

8(2/u)=s hrs 3e
Therefore, the time for which therepairnman remain idle in8-hour day 1s (8-5) hrs=3hrs
No. of jobs ahead to the set brought in- Average no. of jobs in the system =L5/3.

6. A branch of a Nationalized bank has only one typist. Since typing work variae :.
length (number of pages to be typed), the typing rate is randomly distribu n
approximating a Poisson distribution with a mean service rate of 8 letters perhot
The letter arrives at a rate of 5 per hour during the entire 8-hour workday, f 6
typist is valued at Rs. 1.50 per hour, determine. the
) Equipment utilization
n The per cent time of an arriving letter has to wait
ii) Average system time, and
Average idle time cost ofthe typewriter per day. WBUT 2008]
Answer
Here arrival rate is à =
5 per hour, and Service rate is u= 8 per hour. es
) Equipment utilization is P===0.625
8
(i) The percent time an arriving letter has to wait
the percent time the typewriter remains busy= 62.5%
(11) Average system time, i.e., expected time a unit spends in the system is

hr 20min, 0 T97 dud. fee

(iv) Average cost due to waiting on the part of the typewriter per day3 1 2s urim
=8x1-P)x Rs.1.50 Rs.4.50:pspt tobio a1s ado,vusireot
7. a) On average 96 patients per 24, hour day require the service of an
an
emergency clinic. Also on an average, a patient requires 10 minutes of
attention. Assume that the facility can handle only one acuve
emergency at a time
Suppose that it costs the clinic Rs. 100 per patient treated to obtain an averdg
servicing time of 10 minutes and that each minute of decrease in this average
would cost Rs. 10 per patient treated. How much would have
to be budgeted Dy
clinic to decrease the average size of the queue from 1 nt?
patients toa patie
WBUT 2010
Answer:
96
Here 2=-
24x60 patients/min,

1 patients/min.
Expected no. of patients in the waiting line

patients o 13 a
intredertrtd
ORCE-142
 OPERATION RESEARCH

want to reduce
Lq to L =
Ruf we

set L in the formula

Thereforewe 15

1 -)
Thus, we get

=patient/ min.

mins.
Hence the average
rate of treatment required
decrease in the average rate of treatment
Consequently, the
5
=10 mins 2
mins.,
5
100+x10=Rs. 125.
s o s t a i

itad
budget per patient
=

andthe should be increased from Rs.

size of the queue, the budget
So, in order to get the required
100 to Rs. 125 per patient.
five sub-ordinates V,
department has
five jobs A, B, C, D, E and each job is as
b) The head of the of hours each
man would take to perform
W, X, Y, Z. The number
follows:
10 15 8
5
15 18 8
7
12 20 20 12
8 10 6
5 5
10 15 25 10
E 10 sub-ordinate in such a way that total
time is
one
55ign each job to one
and only WBUT 2010]
on bio iisritg
minimized. 3 ut
od be

ORCE-143
 OPERATION RESEARCH

Arrivals at a telephone booth are considered to be Poison with an average time

call iss
minutes distributed
between one arrival and the next. 3 The length of phone
of
Oeumed to
assumed to be with
exponentially,
mean minutes.
a What is the probability that person arriving at the booth will have to wait?
a
that an
a second booth when convinced
The telephone departmentforwillat install
least 3 minutes for a phone call. By how
much
arrival would expect waiting a second booth?
ehould the flow of arrivals increase in order to justify than 10 minutes altogether to
that it will take him more
cil What is the probability WBUT 2014]
wait for the phone and complete his call?
Answer:
1/3
Given =1/10, u
=

=:
Prob. (w> 0) =1-P = 0.3

)We know that the average waiting time is W l -2

booth.
Here, W=3, u 1/3,l =1', for thesecond
=

l=0.16.
Therefore, 3=-

be 0.16-0.10-0.06 arrival per minute.

Hence, increase in the arrival rate should
e(u-2)eu-a)dr
(in) Prob. (waiting time in the system 10)= 10
- 2 3 x0.1003 = 0.03

and the service rate is 4 , then prove that the expected

10. If the arrival rate is 2

2 [WBUT 2016]
queue length is
a2-u)
Answer:
Kefer to Question No. 1 of Long Answer 1ype Questions.
in bank according to Poisson
. Customers arrive at a one-window drive
1
customer is exponential with
aistribution with 10 cars per hour. Service ume perincluding that for the serviced
ean 6 minutes. The space in front or he window
car can accommodate a maximum of 3 cars. Others can wait outside its space,
an arriving customer can drive directly to the
i) What is the probability that
window? an arriving customer will have to wait outside
i) What is probability that the
indicated space?
ii) How long is the arriving
customer expected to wait before starting sent
e?
WBUT 2017]
ORCE-147
 POPULAR PUBLICATIONS

Answer:6va f w ndlos od of heeb a32 976 oog soorgnt

Given 1 0 per hour a a u t6VTIE 0710 9941s
n
i u x60= l10 per hourg e fo 1 hidsdose

) The probability that an arriving customer can drive directly to the space in
window 2

R+R+R-
-a12 1 3
R=1-4
(i) Probability that an arriving
customer will have to wait outside of the
indicated space
=1-0 1
111) Average waiting time of a customer in a
queue
1
- 20-10hours.

1810s Tu8

rio22io oi prib10008 Arsd a 9vb

tiw el19109 es ariotaio
bscioe stti iot et vobtiw-sh0 a s evE
16 oi 6aivno8
unibiioi wobeiw sitt 1a0s 103 21s
o ot Jaot o
yoonil o15 i s3sua 0
1irttoieuo privisst ns 36i
ilidsdor
90t 196 únitthisia
ole
be.osuke aniotus 6062 oste
paivs si