John Loucks: Slides Prepared by

Slides Prepared by
John Loucks
St. Edward’s University
Slides Modified by
Danny Cho
Brock University
© 2020 Cengage Learning. All Rights Reserved. Slide 1

Chapter 15
Waiting Line Models
 Structure of a Waiting Line System
 Queuing Systems
 Queuing System Input Characteristics
 Queuing System Operating Characteristics
 Analytical Formulas
 Single-Channel Waiting Line Model with Poisson
Arrivals and Exponential Service Times
 Multiple-Channel Waiting Line Model with Poisson
Arrivals and Exponential Service Times
 Economic Analysis of Waiting Lines

Structure of a Waiting Line System
 Queuing theory is the study of waiting lines.
 Four characteristics of a queuing system are:

• the manner in which customers arrive
• the time required for service
• the priority determining the order of service
• the number and configuration of servers in the
system.

 Distribution of Arrivals
• Generally, the arrival of customers into the system
is a random event.
• Frequently the arrival pattern is modeled as a
Poisson process.
 Distribution of Service Times

• Service time is also usually a random variable.
• A distribution commonly used to describe service
time is the exponential distribution.

 Queue Discipline
• Most common queue discipline is first come, first
served (FCFS or FIFO).
• An elevator is an example of last come, first served

(LCFS or LIFO) queue discipline.
• Other disciplines assign priorities to the waiting

units and then serve the unit with the highest
priority first.

 Single Service Channel System

Customer Waiting line Customer
arrives S1 leaves
 Multiple Service Channels System
S1
Customer Waiting line Customer

arrives S2
leaves
S3

 Single Service Multiple Phase Channels
Customer Waiting line WL Customer

arrives ST1 ST2 leaves
 Multiple Service Multiple Phase Channels
ST1
ST2
Customer Waiting line WL Customer
arrives ST1 leaves
ST2
ST1

Queuing Systems
 A three part code of the form A/B/k is used to

describe various queuing systems.
• A  the arrival distribution

• B  the service (departure) distribution
• k  the number of channels for the system.

Queuing Systems
 Symbols used for the arrival and service processes are:
 M - Markov distributions (Poisson/exponential)

 D - Deterministic (constant)
 G - General distribution (with a known mean and
variance).
 For example, M/M/k refers to a system in which

arrivals occur according to a Poisson distribution,
service times follow an exponential distribution and
there are k servers working at identical service rates.

Queuing System Input Characteristics
 = the average arrival rate

1/ = the average time between arrivals
µ = the average service rate for each server

1/µ = the average service time
 = the standard deviation of the service time

Queuing System Operating Characteristics
P0 = probability the service facility is idle

Pn = probability of n units in the system
Pw = probability an arriving unit must wait for
service
Lq = average number of units in the queue
awaiting service
L = average number of units in the system
Wq = average time a unit spends in the queue
awaiting service
W = average time a unit spends in the system

Analytical Formulas
 For nearly all queuing systems, there is a relationship

between the average time a unit spends in the system
or queue and the average number of units in the
system or queue.
 These relationships, known as Little's flow equations

are:
L = W and Lq = Wq

Analytical Formulas
 When the queue discipline is FCFS, analytical formulas

have been derived for several different queuing models
including the following:
• M/M/1
• M/M/k
• M/G/1
• M/D/1
• M/M/1 with a finite calling population
• M/G/k with blocked customers cleared – skip this
 Analytical formulas are not available for all possible
queuing systems. In this event, insights may be gained
through a simulation of the system.

Queuing Costs
CS = cost of service per server per time period

Cw = cost of waiting per customer per time period
k = number of servers (or channels)
Total Cost per time period:
TC = k CS + Wq Cw = k CS + Lq Cw (for queue)

or
TC = k CS + W Cw = k CS + L Cw (for system)
Note: Lq = Wq & L = W (Little’s flow equation)

M/M/1 Queuing System
 Single channel
 Poisson arrival-rate distribution
 Exponential service-time distribution
 Unlimited maximum queue length
 Infinite calling population
 Examples:
• Single-window theatre ticket sales booth
• Single-scanner airport security station

Example: SJJT, Inc. (A)
 M/M/1 Queuing System

Joe Ferris is a stock trader on
the floor of the New York Stock
Exchange for the firm of Smith,
Jones, Johnson, and Thomas, Inc.
Stock transactions arrive at a mean
rate of 20 per hour. Each order received by Joe
requires an average of two minutes to process.
This is an M/M/1 queue with  = 20 per hour and

 = 30 per hour.

 Average Time in the System

Question
What is the average time an order must wait from
the time Joe receives the order until it is finished being
processed (i.e. its turnaround time)?
Answer
The average time an order waits in the system is:
W = 1/(µ -  ) = 1/(30 - 20)

= 1/10 hour or 6 minutes

 Average Length of Queue

Question
What is the average number of orders Joe has
waiting to be processed?
Answer
Average number of orders waiting in the queue
is:
Lq = 2/[µ(µ - )]
= (20)2/[(30)(30-20)]
= 400/300
= 4/3

 Utilization Factor
Question
What percentage of the time is Joe processing
orders (or that he is busy)?
Answer
The percentage of time Joe is processing orders is
equivalent to the utilization factor, /. Thus, the
percentage of time he is processing orders is:
/ = 20/30 = 2/3 or 66.67%

M/M/k Queuing System
 Multiple channels (with one central waiting line)

 Exponential service-time distribution
 Examples:
• Four-teller transaction counter in bank
• Two-clerk returns counter in retail store

Example: SJJT, Inc. (B)
 M/M/2 Queuing System

Smith, Jones, Johnson, and Thomas, Inc. has begun
a major advertising campaign which it believes will
increase its business 50%. To handle the increased
volume, the company has hired an additional floor
trader, Fred Hanson, who works at the same speed as
Joe Ferris.
Note that the new arrival rate of orders,  , is 50%
higher than that of problem (A).
Thus,  = 1.5(20) = 30 per hour.

 Sufficient Service Rate

Question
Why will Joe Ferris alone not be able to handle the
increase in orders?
Answer
Since Joe Ferris processes orders at a mean rate of µ
= 30 per hour, then  = µ = 30 and the utilization factor
is 1.
This implies the queue of orders will grow infinitely
large. Hence, Joe alone cannot handle this increase in
demand.

 Probability of n Units in System

Question
What is the probability that neither Joe nor Fred will
be working on an order at any point in time?

 Probability of n Units in System (continued)

Answer
Given that  = 30, µ = 30, k = 2 and ( /µ) = 1, the
probability that neither Joe nor Fred will be working is:
1
P0 
k  1 (  /  )n ( /  )k k
  ( )
n 0 n! k! k  
= 1/[(1 + (1/1!)(30/30)1] + [(1/2!)(1)2][2(30)/(2(30)-30)]
= 1/(1 + 1 + 1) = 1/3 = .333

 Average Time in System

Question
What is the average turnaround time for an order
with both Joe and Fred working?

 Average Time in System (continued)

Answer
The average turnaround time is the average
waiting time in the system, W.
 (  )k (30)(30)(30 30)2 1
Lq  ( P0 )  (1/3) 
( k  1)!( k    ) 2
(1!)(2(30)  30) 2
3
L = Lq + ( /µ) = 1/3 + (30/30) = 4/3
W = L/(4/3)/30 = 4/90 hr. = 2.67 min.

 Average Length of Queue

Question
What is the average number of orders waiting to
be filled with both Joe and Fred working?
Answer
The average number of orders waiting to be filled
is Lq. This was calculated earlier as 1/3.

Example: SJJT, Inc. (C)
 Economic Analysis of Queuing Systems

The advertising campaign of Smith, Jones, Johnson
and Thomas, Inc. (see problems (A) and (B)) was so
successful that business actually doubled. The mean
rate of stock orders arriving at the exchange is now 40
per hour and the company must decide how many
floor traders to employ. Each floor trader hired can
process an order in an average time of 2 minutes.

 Economic Analysis of Queuing Systems

Based on a number of factors the brokerage firm
has determined the average waiting cost per minute
for an order to be $.50. Floor traders hired will earn
$20 per hour in wages and benefits. Using this
information compare the total hourly cost of hiring 2
traders with that of hiring 3 traders.

 Economic Analysis of Waiting Lines

Total Hourly Cost
= (Total salary cost per hour)
+ (Total hourly cost for orders in the system)
= ($20 per trader per hour) x (Number of traders)
+ ($30 waiting cost per hour) x (Average number
of orders in the
system)
= 20k + 30L.
Thus, L must be determined for k = 2 traders and
for k = 3 traders with  = 40/hr. and  = 30/hr. (since
the average service time is 2 minutes (1/30 hr.).

 Cost of Two Servers

1
P0 
k  1 ( /  )n (  /  ) k k
  ( )
n 0 n! k! k  
P0 = 1 / [1+(1/1!)(40/30)]+[(1/2!)(40/30)2(60/(60-40))]
= 1 / [1 + (4/3) + (8/3)]
= 1/5

 Cost of Two Servers (continued)
Thus,
 (  )k (40)(30)(40 30)2 16
Lq  ( P0 )  (1/5) 
( k  1)!( k    ) 2
(1!)(2(30)  40) 2
15
L = Lq + ( /µ) = 16/15 + 4/3 = 2.40
Total Cost = (20)(2) + 30(2.40) = $112.00 per hour

 Cost of Three Servers

1
P0 
k  1 ( /  )n (  /  ) k k
  ( )
n 0 n! k! k  
P0 = 1/[[1+(1/1!)(40/30)+(1/2!)(40/30)2]+
[(1/3!)(40/30)3(90/(90-40))] ]
= 1 / [1 + 4/3 + 8/9 + 32/45]
= 15/59

 Cost of Three Servers (continued)
 (  )k (30)(40)(40 30)3
Lq  ( P0 )  (15/59)  .1446
( k  1)!( k    ) 2
(2!)(3(30)  40) 2
Thus, L = .1446 + 40/30 = 1.4780

Total Cost = (20)(3) + 30(1.4780) = $104.35 per hour

 System Cost Comparison

Wage Waiting Total
Cost/Hr Cost/Hr Cost/Hr
2 Traders $40.00 $82.00 $112.00
3 Traders 60.00 44.35 104.35
Thus, the cost of having 3 traders is less than that of
2 traders.

M/D/1 Queuing System
 Single channel
 Constant service time
 Examples:
• Single-booth automatic car wash
• Coffee vending machine

Example: Ride ‘Em Cowboy!
 M/D/1 Queuing System

The mechanical pony ride
machine at the entrance to a
very popular J-Mart store provides
2 minutes of riding for $.50. Children
wanting to ride the pony arrive
(accompanied of course) according to a
Poisson distribution with a mean rate of 15 per hour.

 What fraction of the time is the pony idle?

l = 15 per hour
m = 60/2 = 30 per hour
Utilization = l/m = 15/30 = .5
Idle fraction = 1 – Utilization = 1 - .5 = .5

 What is the average number of children waiting

to ride the pony?
2 (15) 2
Lq = =  .25 children
2 ( - ) 2(30)(30 - 15)
 What is the average time a child waits for a ride?

 15
Wq = =  .01667 hours
2 ( - ) 2(30)(30 - 15)
(or 1 minute)

M/G/1 Queuing System
Poisson Arrivals and Arbitrary Service Times
The Cutler University Student Union has one self

service copying machine. On the average, 12
customers per hour arrive to make
copies. (The arrival process follows a
Poisson distribution.) The time to copy
documents follows approximately
a normal distribution with a mean
of 2.5 minutes and a standard
deviation of 30 seconds.

The Union manager has received several
complaints from students regarding the long lines at
the copier. Based of this information,
determine:
1) the average number of customers
waiting or using the copier.
2) the probability an arriving
customer must wait in line.
3) the proportion of time the copier is idle.
4) the average time a customer must wait in line
before using the copier.

This situation can be modeled as an M/G/1
system with:
l = 12 per hour = 12/60 = .2/minute

1/µ = 2.5 minutes
µ = 1/(2.5) = .4 per minute
s = 30 seconds = .5 minutes

 Average number of customers waiting or using
the copier
    (  )
2 2 2
L  Lq  where Lq 
 2(1    )
Substituting the appropriate values gives:
(.2) 2 (.5)2  (.2 .4) 2

Lq   .26
2(1  .2 .4)
Thus L = .26 + (.2/.4) = .76 customers

 Probability an arriving customer must wait in line
PW     .2 .4  .5
 Proportion of time the copier is idle
P0  1     1  .2 .4  .5
 Average time a customer must wait in line before
using the copier
Wq  Lq   .26 .2  1.3 minutes

M/G/k Queuing System
Poisson Arrivals, Arbitrary Service Times,
and No Waiting Line
Several firms are over-the-counter (OTC) market
makers of MegaTech stock. A broker wishing to
trade this stock for a client will call on
these firms to execute the order. If the
market maker's phone line is busy, a
broker will immediately try calling
another market maker to transact the
order.

and No Waiting Line
Richardson and Company is one such OTC
market maker. It estimates that on the average, a
broker will try to call to execute a stock
transaction every two minutes. The time
required to complete the transaction
averages 75 seconds. The firm has four
traders staffing its phones. Assume calls
arrive according to a Poisson distribution.

and No Waiting Line
What percentage of its potential business will be
lost by Richardson?
What percentage of its potential
business would be lost if only three
traders staffed its phones?

and No Waiting Line
This problem can be modeled as an M/G/k
system with block customers cleared with:
1/l = 2 minutes = 2/60 hour

l = 60/2 = 30 per hour
1/µ = 75 seconds = 75/60 minutes = 75/3600 hours
µ = 3600/75 = 48 per hour

and No Waiting Line
 Percentage of potential business that be lost when
using 4 traders (k = 4)
The system will be blocked when there are four
customers in the system. Hence, the answer is P4.
1
P0  k
where k = 4
 (
i 0
  ) i
/ i!
1
P0   .536
1  (30/ 48)  (30/ 48) / 2! (30/ 48) / 3! (30/ 48) / 4!
2 3 4
continued

and No Waiting Line
(  ) 4 (30/48)4
Now, P4  P0 = (.536)  .003
4! 24
Thus, with four traders 0.3% of the potential
customers are lost.

and No Waiting Line
The system will be blocked when there are three
customers in the system. Hence, the answer is P3.
1
P0   .537
1  (30/ 48)  (30/ 48) / 2! (30/ 48) / 3!
2 3
(  ) 3
(30/48) 3
P3  P0 = (.537)  .022
3! 6
Thus, with three traders 2.2% of the potential
customers are lost.
with Finite Calling Populations
Biff Smith is in charge of maintenance for four of
the rides at the Rollerama Amusement Park. On the
average, each ride operates four
hours before needing repair.
When repair is needed, the
average repair time is 10 minutes.
Assuming the time between
machine repairs as well as the
service times follow exponential
distributions, determine:
1) the proportion of time Biff is idle, and
2) the average time a ride is "down" for repairs.

This problem can be modeled as an
M/M/1 queue with a finite calling population
of size N = 4 (rides).
l = 1/(4 hours) = .25 per hour

µ = 60/(10 minutes) = 6 per hour
l/m = .25/6 = 1/24

 Proportion of time Biff is idle
1
P0  N
N!  n
n 0 ( N  n)!
( )

1
P0   .841
4! 4! 4! 4! 4!
(1/ 24)  (1/ 24)  (1/ 24)  (1/ 24)  (1/ 24) 4
0 1 2 3
4! 3! 2! 1! 0!
Hence, Biff is idle approximately 84% of the time.

 Average time a ride is "down" for repairs
1
W  Wq 

where
Lq 
Wq  Lq  N  (1- P0 ) L  Lq  (1  P0 )
( N  L ) 

 Average time a ride is "down" for repairs
Substituting the appropriate values gives:

Lq = 4 - ((.25+6)/.25)(1-.840825) = .020625
L = .020625 + (1-.840825) = .1798

Wq = .020625/((4-.1798).25) = .021596
W = .021596 + 1/6 = .18826 hours or 11.3 minutes

End of Chapter 15

John Loucks: Slides Prepared by

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

John Loucks: Slides Prepared by

Uploaded by

Copyright:

Available Formats

Slides Prepared by

© 2020 Cengage Learning. All Rights Reserved. Slide 1

© 2020 Cengage Learning. All Rights Reserved. Slide 2

 Queuing theory is the study of waiting lines.

 Four characteristics of a queuing system are:

© 2020 Cengage Learning. All Rights Reserved. Slide 3

 Distribution of Service Times

© 2020 Cengage Learning. All Rights Reserved. Slide 4

• An elevator is an example of last come, first served

• Other disciplines assign priorities to the waiting

© 2020 Cengage Learning. All Rights Reserved. Slide 5

 Single Service Channel System

 Multiple Service Channels System

Customer Waiting line Customer

© 2020 Cengage Learning. All Rights Reserved. Slide 6

 Single Service Multiple Phase Channels

Customer Waiting line WL Customer

 Multiple Service Multiple Phase Channels

© 2020 Cengage Learning. All Rights Reserved. Slide 7

 A three part code of the form A/B/k is used to

• A  the arrival distribution

© 2020 Cengage Learning. All Rights Reserved. Slide 8

 Symbols used for the arrival and service processes are:

 M - Markov distributions (Poisson/exponential)

 For example, M/M/k refers to a system in which

© 2020 Cengage Learning. All Rights Reserved. Slide 9

 = the average arrival rate

µ = the average service rate for each server

 = the standard deviation of the service time

© 2020 Cengage Learning. All Rights Reserved. Slide 10

P0 = probability the service facility is idle

© 2020 Cengage Learning. All Rights Reserved. Slide 11

 For nearly all queuing systems, there is a relationship

 These relationships, known as Little's flow equations

© 2020 Cengage Learning. All Rights Reserved. Slide 12

 When the queue discipline is FCFS, analytical formulas

© 2020 Cengage Learning. All Rights Reserved. Slide 13

CS = cost of service per server per time period

Total Cost per time period:

TC = k CS + Wq Cw = k CS + Lq Cw (for queue)

Note: Lq = Wq & L = W (Little’s flow equation)

© 2020 Cengage Learning. All Rights Reserved. Slide 14

© 2020 Cengage Learning. All Rights Reserved. Slide 15

 M/M/1 Queuing System

This is an M/M/1 queue with  = 20 per hour and

© 2020 Cengage Learning. All Rights Reserved. Slide 16

 Average Time in the System

W = 1/(µ -  ) = 1/(30 - 20)

© 2020 Cengage Learning. All Rights Reserved. Slide 17

 Average Length of Queue

© 2020 Cengage Learning. All Rights Reserved. Slide 18

/ = 20/30 = 2/3 or 66.67%

© 2020 Cengage Learning. All Rights Reserved. Slide 19

 Multiple channels (with one central waiting line)

© 2020 Cengage Learning. All Rights Reserved. Slide 20

 M/M/2 Queuing System

Thus,  = 1.5(20) = 30 per hour.

© 2020 Cengage Learning. All Rights Reserved. Slide 21

 Sufficient Service Rate

© 2020 Cengage Learning. All Rights Reserved. Slide 22

 Probability of n Units in System

© 2020 Cengage Learning. All Rights Reserved. Slide 23

 Probability of n Units in System (continued)