Waiting Line Models For

Service Improvement
by:
Shannon Duffy
Katie McPartlin
Jason Jacklow
Amanda Holtz
B.J. Ko



Economic Optimization Model

EOM
Which has developed
using queuing analysis
EOM
Used at L.L. Bean for telemarketing
operations
To determine the optimal number of
telephone trunks for incoming calls
The number of agents scheduled
The queue capacity
The maximum number of customers who are
put on hold to wait for an agent
Queuing models are used

To determine the economic impact of busy
signals
Customer waiting time
Lost orders


Decision about waiting lines
Are based on averages for customer arrivals
and service times
They are used to computer operation
characteristics
Which are average of values for characteristics
that describe the performance of a waiting line
system
Elements of a waiting line

Basic element is queue
Which is a single waiting line
Which consists of arrivals, servers, and waiting
line structure
Single-channel queuing system
The Calling Population
Is the source of the customers to the
queuing system, and it can be either infinite
or finite
Infinite
Calling population assumes such a large
number of potential customers that is always
possible for one more customer to arrive to be
served
Ex. grocery store, bank
Finite calling population
Has a specific, countable
number of potential
customers
Ex. repair facility in a
shop

Arrival Rate
Is the rate at which customers arrive at the
service facility during a specified period of
time
This rate can be estimated from empirical
data derived from studying the system or a
similar system, or it can be average of these
empirical data
Service times
The time required to serve a customer, is
more frequently described by the negative
exponential distribution
Service must be expressed as a rate to be
compatible with the arrival rate
Customers must be served faster than they
arrive or an infinitely large queue will build
up

Queue Discipline and Length

Queue discipline is
the order in which
waiting customers
are served
Most common type
is first come, first
served
Infinite Queue
Can be of any size
with no upper limit
and is the most
common queue
structure
Ex. Movie theater
line
Finite Queue
Is limited to size
Ex. Driveway at
bank
Basic Waiting Line Structures
There are four basic structures according to
the nature of the service facilities
Single-channel, single-phase
Single-channel, multiple-phase
Multiple-channel, single-phase
Multiple-channel, multiple-phase
Channels and Phases
Channel
Is the number of parallel servers for servicing
arriving customers
Phases
Denotes the number of sequential servers each
customer must go through to complete service

Poisson Distribution
The Poisson Distribution is a discrete
distribution which takes on the values
X=0,1,2,3
It is often used as a model for the number of
events (such as the number of telephone calls at
a business or the number of accidents at an
intersection) in a specific time period.
Single-Channel, Single-Phase
Models
Most basic of the waiting line structures
Frequently used variation
Poisson arrival rate, exponential service times
Poisson arrival rate, general distribution of
service times
Poisson arrival rate, constant service items
Poisson arrival rate, exponential service times
with a finite queue and a finite calling
population
Basic Single-Server Model

Assume the following
Poisson arrival rate
Exponential service times
First-come, first-served queue discipline
Infinite queue length
Infinite calling population
Constant Service times
The single-server model with Poisson
arrivals and constant service times is a
queuing variation that is of particular
interest in operations management, since the
most frequent occurrence of constant
service times is with automated equipment
and machinery.
This model has direct applications for many
manufacturing operations
Finite Queue Length
Since some waiting lines systems the length
of the queue may be limited by the physical
area in which the queue forms;
Space may permit only a limited number of
customers to enter the queue
Variation of the single-phase, single-channel
queuing model
Finite Calling Populations
The population of
customers from
which arrivals
originate is limited,
such as the number
of police cars at a
station to answer
calls
Multi-Server Models
Two or more independent servers in parallel
serve a single waiting line
The number of servers must be able to serve
customers faster than they arrive
Definition of Variables
P
n
= Probability of n Units in System
= Mean Number of Arrivals per Time Period
= Mean Number of People or Items Served per
Time Period
L
s
= Average Number of Units in the System
W
s
= Average Time a Unit Spends in the System
(Wait Time + Service Time)

Definition of Variables
Continued
L
q
= Average Number of Units in the
Waiting Line
W
q
= Average Time a Unit Spends Waiting
in the Line
= Utilization Factor for the System
(Percent of Time the Servers are Busy)
P
0
= Probability of 0 Units in the System

Single Channel, Single Phase
L
s
= /( )
W
s
= 1/( )
L
q
=
2
/(( ))
W
q
= /(( ))
= /
P
0
= 1 (/)
P
n
= (/)
n
(1 (/))
In all Cases, >

Single Channel, Single Phase
Example
For = 2 cars arriving/hour
= 3 cars serviced/hour
L
s
= /( ) = 2/(3-2) = 2 cars in System
W
s
= 1/( ) = 1/(3-2) = 1 hour average
time spent in system
L
q
=
2
/(( )) = 2
2
/3(3-2) = 1.33 cars
waiting

Single Channel, Single Phase
Example Cont.
W
q
= /(( )) = 2/(3(3-2)) = 2/3 hour or 40
minute average time waiting
= / = 2/3 = 66.7% Utilization of Mechanic
P
0
= 1 (/) = 1-(2/3) = 1/3 = 0.33 probability
there are no cars in system = 0.33
P
1
= (/)
1
(1 (/)) = (2/3)
1
(1 (2/3)) = 1/9
probability there is 1 car in system = 0.11
The End
Any Questions?