Waiting Line Management

Components of the Queuing System

Servicing System
Servers
Queue or
Customer
Arrivals

Waiting Line
Exit

Customer Service Population Sources

Population Source

Finite

Infinite

Example:
Example:Number
Numberof
of
machines
machinesneeding
needing
repair
repairwhen
whenaa
company
companyonly
onlyhas
has
three
threemachines.
machines.

Example:
Example:The
The
number
numberof
ofpeople
people
who
whocould
couldwait
waitin
in
aaline
linefor
for
gasoline.
gasoline.

Service Pattern

Service
Pattern

Constant
Example:
Example:Items
Items
coming
comingdown
downan
an
automated
automated
assembly
assemblyline.
line.

Variable
Example:
Example:People
People
spending
spendingtime
time
shopping.
shopping.

The Queuing System

Length

Queue Discipline

Queuing
System

Service Time
Distribution

Number of Lines &

Line Structures

Examples of Line Structures

Single
Phase
One-person
Single Channel
barber shop
Multichannel

Bank tellers
windows

Multiphase
Car wash
Hospital
admissions

Degree of Patience

No Way!

BALK

No Way!

RENEG

Suggestions for Managing Queues

1. Determine an acceptable waiting time for
your customers
2. Try to divert your customers attention
when waiting
3. Inform your customers of what to expect
4. Keep employees not serving the customers
out of sight
5. Segment customers

Suggestions for Managing Queues (Continued)

6. Train your servers to be friendly
7. Encourage customers to come during the
slack periods
8. Take a long-term perspective toward
getting rid of the queues

Waiting Line Models

Model Layout
1
Single channel

Source
Population
Infinite

Service Pattern
Exponential

Single channel

Infinite

Constant

Multichannel

Infinite

Exponential

Single or Multi

Finite

Exponential

These four models share the following characteristics:

Single phase
Poisson arrival
FCFS
Unlimited queue length

Notation: Infinite Queuing: Models 1-3

== Arrival
Arrivalrate
rate
==Service
Servicerate
rate
11
Average
service
time
Average
service
time

11
Average
time
between
arrivals
Average
time
between
arrivals

== == Ratio
Ratioof
of total
totalarrival
arrivalrate
rate to
tosevice
sevicerate
rate

for
for aa single
singleserver
server
Lg
Lg Average
Average number
number wai
waiting
tingin
in line
line

Infinite Queuing Models 1-3 (Continued)

LLss== Average
Average number
numberin
in system
system
(including
(including those
those being
beingserved)
served)
W
Wgg == Average
Average time
time waiting
waitingin
in line
line
Ws
Ws Average
Average total
total time
timein
in system
system
(including
(including time
time to
tobe
beserved)
served)
nn Number
Numberof
of units
unitsin
in the
thesystem
system
SS == Number
Numberof
of identical
identicalservice
servicechannels
channels
Pn
Pn Probabilit
Probabilityyof
of exactly
exactlynnunits
unitsin
in system
system
Pw
Pw Probabilit
Probabilityyof
of waiting
waitingin
in line
line

Example: Model 1
Assume a drive-up window at a fast food restaurant.
Customers arrive at the rate of 25 per hour.
The employee can serve one customer every two
minutes.
Assume Poisson arrival and exponential service
rates.
Determine:
Determine:
A)
A) What
Whatis
isthe
theaverage
averageutilization
utilizationof
ofthe
theemployee?
employee?
B)
B) What
Whatis
isthe
theaverage
averagenumber
numberof
ofcustomers
customersin
inline?
line?
C)
C) What
Whatis
isthe
theaverage
averagenumber
numberof
ofcustomers
customersin
inthe
the
system?
system?
D)
D) What
Whatis
isthe
theaverage
averagewaiting
waitingtime
timein
inline?
line?
E)
E) What
Whatis
isthe
theaverage
averagewaiting
waitingtime
timein
inthe
thesystem?
system?
F)
F) What
Whatis
isthe
theprobability
probabilitythat
thatexactly
exactlytwo
twocars
carswill
willbe
be
in
inthe
thesystem?
system?

Example: Model 1

A) What is the average utilization of the employee?

== 25
25 cust
cust// hr
hr
11 customer
customer
==
== 30
cust
// hr
30
cust
hr
22 mins
(1hr
/
60
mins)
mins (1hr / 60 mins)

25
cust
// hr
25
cust
hr = .8333
==
==
= .8333

30
cust
/
hr
30 cust / hr

Example: Model 1
B) What is the average number of customers in
line?

(25)
(25)
Lg
==
== 4.167
Lg ==
4.167
(( --)) 30(30
30(30--25)
25)
22

22

C) What is the average number of customers in the

system?

25
25
Ls
=
=
Ls =
=
=
= 55
-- (30
(30--25)
25)

Example: Model 1

D) What is the average waiting time in line?

Lg
Lg
Wg
== .1667
Wg ==
.1667 hrs
hrs == 10
10 mins
mins

E) What is the average waiting time in the system?

Ls
Ls
Ws
==.2
Ws ==
.2 hrs
hrs ==12
12mins
mins

Example: Model 1

F) What is the probability that exactly two cars

will be in the system (one being served and the
other waiting in line)?

pp =
= (1(1- )(
)( ))

nn

nn

25
25 25
25 22
pp2 == (1)(
)) == .1157
(1)(
.1157
2
30
30 30
30

Example: Model 2
An automated pizza vending machine heats and
dispenses a slice of pizza in 4 minutes.
Customers arrive at a rate of one every 6 minutes
with the arrival rate exhibiting a Poisson
distribution.
Determine:
Determine:
A)
A)
B)
B)

The
The average
average number
number of
of customers
customers in
in line.
line.
The
The average
average total
total waiting
waiting time
time in
in the
the system.
system.

Example: Model 2
A) The average number of customers in line.
22
22
(10)
(10)
Lg
=
=
==.6667
Lg =
=
.6667
22(( --)) (2)(15)(15
(2)(15)(15--10)
10)

B) The average total waiting time in the system.

10
10
Wg
=
=
==.06667
Wg =
=
.06667hrs
hrs==44mins
mins
22(( --)) 22((15
15)()(1155--10)
10)

11
11
Ws
==.1333
Ws==Wg
Wg++ ==.06667
.06667hrs
hrs++
.1333hrs
hrs==88mins
mins

15/hr
15/hr

Example: Model 3
Recall the Model 1 example:
Drive-up window at a fast food restaurant.
Customers arrive at the rate of 25 per hour.
The employee can serve one customer every two
minutes.
Assume Poisson arrival and exponential service
rates.
IfIf an
an identical
identical window
window (and
(and an
an identically
identically trained
trained
server)
server) were
were added,
added, what
what would
would the
the effects
effects be
be on
on
the
the average
average number
number of
of cars
cars in
in the
the system
system and
and the
the
total
total time
time customers
customers wait
wait before
before being
being served?
served?

Example: Model 3
Average number of cars in the system
Lg
Lg == 00..176
176

(Exhibit
(ExhibitTN6.10
TN6.10---using
-usinglinear
linearinterpolat
interpolation)
ion)

25
25
Ls
Ls == Lg
Lg ++ ==.176
.176 ++ ==1.009
1.009

30
30
Total time customers wait before being served
Lg
.176
customers
Lg
.176
customers = .007 mins ( No Wait! )
Wg
=
=
Wg =
=
= .007 mins ( No Wait! )
25
customers/
min
25 customers/min

Notation: Finite Queuing: Model 4

D
D == Probabilit
Probabilityy that
thatan
an arrival
arrivalmust
must wait
waitin
in line
line
FF Efficiency
Efficiencyfactor,
factor,aa measure
measureof
of the
theeffect
effect of
of
having
having to
to wait
waitin
in line
line
H
H Average
Average number
numberof
of units
units being
beingserved
served
JJ == Population
Populationsource
sourceless
lessthose
thosein
in queuing
queuing
system
system((NN --nn))
LL Average
Average number
numberof
of units
unitsin
in line
line
SS Number
Numberof
of service
servicechannels
channels

Finite Queuing: Model 4 (Continued)

nn==Average
Averagenumber
numberof
of units
unitsin
inqueuing
queuingsystem
system
(including
(includingthe
theone
onebeing
beingserved)
served)
NN Number
Numberof
of units
unitsin
in population
populationsource
source
Pn
Pn Probabilit
Probabilityyof
of exactly
exactlynnunits
unitsin
inqueuing
queuingsystem
system
TT ==Average
Average time
timeto
toperform
performthe
theservice
service
U
U Average
Averagetime
timebetween
betweencustomer
customerservice
servicerequiremen
requirementsts
W
W Average
Average waiting
waitingtime
timein
inline
line
XX Service
Servicefactor,
factor,or
orproportion
proportionof
of service
servicetime
timerequired
required

Example: Model 4
The
The copy
copy center
center of
of an
an electronics
electronics firm
firm has
has four
four copy
copy
machines
machines that
that are
are all
all serviced
serviced by
by aa single
single technician.
technician.
Every
Every two
two hours,
hours, on
on average,
average, the
the machines
machines require
require
adjustment.
adjustment. The
The technician
technician spends
spends an
an average
average of
of 10
10
minutes
minutes per
per machine
machine when
when adjustment
adjustment is
is required.
required.
Assuming
Assuming Poisson
Poisson arrivals
arrivals and
and exponential
exponential service,
service,
how
how many
many machines
machines are
are down
down (on
(on average)?
average)?

Example: Model 4
N, the number of machines in the population = 4
M, the number of repair people = 1
T, the time required to service a machine = 10 minutes
U, the average time between service = 2 hours
T
10
min
T
10
min
X
=

X = T + U 10 min + 120 min == .077

.077
T+ U
10 min + 120 min

From
FromTable
TableTN6.11,
TN6.11,FF==.980
.980(Interpolation)
(Interpolation)
L,
L,the
thenumber
numberof
ofmachines
machineswaiting
waitingto
tobe
be
serviced
serviced==N(1-F)
N(1-F)==4(1-.980)
4(1-.980)==.08
.08machines
machines
H,
H,the
thenumber
numberof
ofmachines
machinesbeing
being
serviced
serviced==FNX
FNX==.980(4)(.077)
.980(4)(.077)==.302
.302machines
machines

Number
Number of
of machines
machines down
down == LL++H
H== .382
.382 machines
machines

Queuing Approximation

This approximation is quick way to analyze a queuing situation. Now, both

interarrival time and service time distributions are allowed to be general.
In general, average performance measures (waiting time in queue, number
in queue, etc) can be very well approximated by mean and variance of the
distribution (distribution shape not very important).
This is very good news for managers: all you need is mean and standard
deviation, to compute average waiting time

Define:
Standard deviation of X
Mean of X
Variance
2
Cx2 squared coefficient of variation (scv) = Cx
mean2
Cx coefficient of variation for r.v. X =

Queue Approximation
Inputs: S, , , Ca2 ,Cs2
(Alternatively: S, , , variances of interarrival and service time distributions)

Compute
S
2( S 1) Ca2 Cs2
Lq

1
2

Lq

Ls
as before, Wq , and Ws

Ls Lq S

Approximation Example

Consider a manufacturing process (for example making plastic parts) consisting of a single stage
with five machines. Processing times have a mean of 5.4 days and standard deviation of 4 days.
The firm operates make-to-order. Management has collected date on customer orders, and verified
that the time between orders has a mean of 1.2 days and variance of 0.72 days. What is the
average time that an order waits before being worked on?
Using our Waiting Line Approximation spreadsheet we get:
Lq = 3.154 Expected number of orders waiting to be completed.
Wq = 3.78 Expected number of days order waits.
= 0.9 Expected machine utilization.