McGraw-Hill Education Copyright 2017 by McGraw-Hill Education. All rights reserved.

SERVICE PROCESSES
Chapter 7
Learning Objectives
1. Understand the characteristics of service processes
2. Analyze simple service systems
3. Understand waiting line (queueing) analysis

7- 2
Operational Classification of Services

Service organizations are classified according to the

customers they service and the service they provide
to those customers

Customer Contact Creation of the

service

The physical presence Service systems with a The work process

of the customer in the high degree of involved in providing
system customer contact are the service itself
more difficult to control

7- 3
Service-System Design Matrix

7- 4
Virtual Services
¨ Web-platform businesses – Facebook, Uber,
Alibaba ) they connect the buyer and the seller
(they don’t sell their products)
¨ Non-platform Web businesses – Netflix HBO,
Walmart (they sell their own products as well)

7- 5
Service Blueprinting and Fail-safing
¨ The standard tool for service process design is the
flowchart
¤ Called a service blueprint
¨ A unique feature of the service blueprint is the
distinction made between the high customer contact
aspects of the service and those activities that the
customer does not see
¤ Made with a “line of visibility” on the flowchart
¨ Fail-safing involves using the service blueprint to
identify opportunities for failure and then establishing
procedures to prevent mistakes from becoming defects
(poka-yokes)

7- 6
Service Organization Design
¨ Services cannot be stored in inventory
¨ In services, capacity becomes the dominant issue
¤ Too much capacity leads to excessive costs
¤ Insufficient capacity leads to lost customers

¨ Waiting line models provide a powerful

mathematical tool for analyzing many common
service situations

7- 7
Waiting Line Problems (Queues)
¨ A central problem in many service settings is the
management of waiting time
¤ Reducing waiting time costs money, but raises customer
satisfaction and throughput
¨ When people waiting are employees, it is easy to
value their time
¨ When people waiting are customers, it is more
difficult to value their time
¤ Lost sales is one value (often a low estimate)

7- 8
 Arrival and Service Profiles

Arrivals often
vary greatly
over a time Service
period capacity is
usually fixed

7- 9
Practical View of Waiting Lines
The number of arrivals over Customers demand varying
the hours that the service amounts of service, often
system is open exceeding normal capacity

Waiting Line
Factors

We can control arrivals We can affect service time by

• Short lines using faster or slower servers,
• Specific hours for specific customers or by using creative methods
• Specials

7- 10
Operations and SCM in Practice
EFFECTIVE MANAGEMENT OF QUEUES DURING BOARDING CAN INCREASE AIRLINE
BOTTOM LINE

An important question for airlines is “What is the

best way to board passengers?”Avoiding queues
and getting passengers on a plane quickly can
greatly affect an airline’s costs. Southwest says
that if its boarding times increased by 10 minutes
per flight, it would need 40 more planes at a cost
of US$40 million each to run the same number of
flights it runs now.

7- 11
14-1
Managing Queues

Segment the customers

Train your servers to be friendly
Inform your customers of what to expect
Try to divert the customer’s attention when waiting

Encourage customers to come during slack periods

7- 12
Queuing System Components

Service System 7- 13
 Queuing System Analysis
Queuing Systems essentially consists of three major components
1) Source Population and Customer Arrivals
a) Source Population - who are your customers?
n Population size – finite or infinite?
b) Customer Arrivals
n Arrival Rate Characteristics
n Random; Exponential and Poisson Distributions
n Constant rate
n Customer Arrival Characteristics
n Arrival patterns (steady or seasonal)
n Size or arrival rates (individuals or groups)
n Degree of patience (will they wait?)
7- 14
Queuing System Analysis
2) Service Systems
a) Waiting Lines
n Line Length, Number of Lines, Queue Discipline (Priority Rules)
b) Servers and Service Rate (Time)
n Service Rate Characteristics
n Random; Exponential distribution
n Constant rate; machine controlled operations

c) Service Line Structures

n Single channel: single phase or multiphase
n Multichannel: single phase or multiphase
n Mixed structures
3) Condition of Exiting Customers
a) Low Probability of reservice
b) High Probability of reservice
7- 15
Waiting Line Models

(a)
Customer
population
Source Service system

Waiting line Served

(d)
Service customers
(c) line
Priorit structures
(b) (e)
y rule
Arrival Service
Rate Time
 a) Customer Population Sources
¨ The source of customers to the waiting
line system, and it can be either Population Source
infinite or finite.

¨ Infinite population source: a large

number of potential customers that it is
always possible for one more customer Finite Infinite
to arrive to be served; that is the
whole town or geographic area. Example: a Example: a bank and a
repair person in grocery store.
¨ Finite population source: a specific, a shop who is
countable number of potential responsible for
customers. a fixed # of
machines to
work on.
Customer Arrival Factors

Balking

Reneging
7- 18
 (b) Customer Arrival Rate
¨ The frequency at which customers arrive at a waiting line
according to a probability distribution.
¤ Generally, arrival of customers into the system is a random
event.
¤ Frequently arrival pattern is modeled as a Poisson process
where customer arrivals occur continuously and
independently at a constant average rate, l number of
customers per unit time.
¨ Poisson Process Properties:
¤ The number of arrivals in any time interval is independent
of the number of arrivals in any other time interval.
¤ The probability of an arrival in an interval is the same for
all equal-size intervals

7- 19
 (b) Customer Arrival Rate
¨ Poisson Probability Distribution:
¤ Probability of n customers arrive when average (mean)
arrival rate is l customers per unit time within a time
length (T):
(lT)n -lT
PT(n) = e
n!

¨ Mean and Variance of Poisson Distribution over any

time interval T is equal to the rate, lT.

7- 20
(b) Customer Arrival Rate (Cont.)

If the Arrival rate = 2/hour

What is the Probability that 4 customers will
arrive within a one-hour period?
[2(1)]4 -2(1)
P1(4) = e
4!

7- 21
Arrival Distributions - Poisson
¨ Describes the number of arrivals (n) in some time period (T)
¨ E.g. With a mean arrival rate of three per minute (λ=3),
what is the probability of exactly five (n=5) arrivals in the
next minute?

𝜆𝑇 " 𝑒 #$!
𝑃! 𝑛 =
𝑛!

3×1 & 𝑒 #'×%

𝑃% 5 =
5!
!
' ) "#
= %*+
= 0.101
= 10.1%
7- 22
Arrival Distributions - Exponential
¨ The exponential distribution (also known as
the negative exponential distribution) is
the probability distribution that describes the time
between events (arrivals) in a Poisson process
¨ Describes the probability distribution of time
between arrivals (inter-arrival times)
¨ It has the infamous memoryless property
¤ Itdoes not matter when last arrival happened,
the probability of a customer arriving in the next
time unit is always the same
7- 23
 Arrival Distributions - Exponential
¨ Properties of Exponential Distribution
¨ Let 𝜆 denote arrival rate per unit time
¨ Mean = 1/ 𝜆
¨ Variance = 1/(𝜆2)

𝑃𝑟𝑜𝑏. 𝐷𝑒𝑛𝑠𝑖𝑡𝑦 𝐹𝑢𝑛𝑐𝑡𝑖𝑜𝑛: 𝑓 𝑡 = 𝜆𝑒 !"#

Distribution Function: Area = 𝐹 𝑡 = 1 − 𝑒 !"#

7- 24
 Arrival Distributions - Exponential
¨ For any given 𝜆 , in order to find the probability that the
next customer arrival will happen within a specified period
of time t, use F(t)
𝑓 𝑡 = 𝜆𝑒 !"#

𝐹 𝑡 = 1 − 𝑒 !"#

For 𝜆 = 1, P (an arrival in the next t mins)

= 1 − 𝐹 𝑡 = 𝑒 !"# = 𝐹 𝑡 = 1 − 𝑒 !"#

7- 25
Queuing System (Waiting Line) Factors

¨ Length – how much

waiting room space is
available?
¨ Number of lines – how
many servers are
working?
¨ Queue discipline –
priority rule or set of
rules that determine
the order of service
for customers who are
waiting in line

7- 26
(c) Priority Rule
¨ Queue discipline (PRIORITY RULES) – how do new arrivals
enter the line? How do you decide which customer to
serve next?

¨ Most common queue discipline is first come, first

served (FCFS) (We will consider this discipline in our
models.)

¨ Other disciplines assign priorities to the waiting units and

then serve the unit with the highest priority first.
Copyright © 2013 McGraw-Hill Ryerson Limited 7- 27
(d) Service Line Structure
¨ Service Line Structure – what does the process look like?

¨ The service facility can be classified in terms of # of

service channels and # of service phases.

¨ Channels: # of parallel servers for serving customers

¨ Phases: # of servers in sequence a customer must go

through

Copyright © 2013 McGraw-Hill Ryerson Limited 7- 28

(d) Service Line Structure (Cont.)
Single channel, single phase
(We will study this scenario in detail.)

Service
facility

E.g. One-person barber shop,

A postal office with only one
postal clerk working
Copyright © 2013 McGraw-Hill Ryerson Limited 7- 29
 (d) Service Line Structure (Cont.)
Multiple channel, single phase
(We will study this scenario in detail.)

Service
facility 1

Service
facility 2

E.g. Bank tellers’ windows

(d) Service Line Structure (Cont.)
Single channel, multiple phase
(We will NOT study this scenario in detail.)

Service Service
facility 1 facility 2

E.g. Car wash,

Hospital admission with a single person in the counter
and a single doctor
 (d) Service Line Structure (Cont.)
Multiple channel, multiple phase
(We will NOT study this scenario in detail.)

Service Service
facility 1 facility 3

Service Service
facility 2 facility 4

E.g. Hospital admissions with several persons in the

counter and several doctors
 (d) Service Line Structure (Cont.)
¨ In a college registration process, several department heads have to approve
an individual student's semester course load. What is the queuing system line
structure?
A. Single channel, single phase
B. Single channel, multiphase
C. Multichannel, single phase
D. Multichannel, multiphase
E. None of the above

¨ Buying food at a large food store with multiple checkout counters features
which type of queuing system line structure?
A. Single channel, single phase
B. Single channel, multiphase
C. Multichannel, single phase
D. Multichannel, multiphase
E. None of the above
Copyright © 2013 McGraw-Hill Ryerson Limited 7- 33
Service Time Distribution
¨ Constant
¤ Service is automated and not customized to individual
customers (automatic car wash)
¨ Variable
¤ Service is provided by humans
¤ Can be customized to individual customers

¤ Described using exponential distribution

7- 34
Service Time Distribution
Service
Pattern

Constant Variable
Example: machine Example: People
controlled operations, spending time
such as vending shopping.
machines

7- 35
 (e) Service Time (Cont.)
¨ Service time distribution
¤ A distribution commonly used to describe random service time
is the exponential distribution
¨ E.g. Probability that a customer requires less than time t
for finishing the service when service rate is µ (average
# of customers served per unit of time). Defining T as
the time it takes to serve a customer, P(T ≤ t) = 1 – e-µt
¨ Probability that a customer requires less than 0.167
hours (10 minutes) when the service rate is 3 customers
per hour
P(T ≤ 0.167 hr) = 1 – e-3(0.167) = 1 – 0.61 = 0.39
7- 36
Some Specific Waiting Line Models

7- 37
Notation: Infinite Queuing: Models 1-3

l = Arrival rate
µ = Service rate
1
= Average service time
µ
1
= Average time between arrivals
l
Lq = Average number waiting in line

7- 38
Infinite Queuing Models 1-3 (Continued)
Ls = Average number in system
(including those being served)
Wq = Average time waiting in line
Ws = Average total time in system
(including time to be served)
n = Number of units in the system
S = Number of identical service channels
Pn = Probabilit y of exactly n units in system
Pw = Probability of waiting in line 7- 39
Model 1: Single Channel and Exponential Service Time

average utilization of the system l

(probability that the server is busy and a r =
customer has to wait.) µ

probability that no customer exists

l
in the system
p0 = (1 - )
µ
probability that exactly n
customers will be in the system l l n
p = (1 - )( )
n
µ µ
 Model 1: Single Channel and Exponential Service Time

average number of customers in average waiting time in line

line (i.e., the customers in the
waiting line) Lq
l 2 Wq =
Lq = l
µ (µ - l )
Note the use of Little’s Law:
Inventory=Throughput Rate*Flowtime.
Same here: Lq= l * Wq
average number of customers in the
system (i.e., the customers being average waiting time in the system
serviced and in the waiting line) (i.e., waiting and being serviced)

l Ls
Ls = Ws =
µ -l
l
Model 2: Single Channel and Constant Service Time

average number of customers in average waiting time in line

line (i.e., the customers in the
waiting line) Lq
Wq =
Lq =
l 2
l
2µ ( µ - l )

average number of customers in the average waiting time in the system

system (i.e., the customers being (i.e., waiting and being serviced)
serviced and in the waiting line)

l Ls
Ls = Lq + Ws =
µ l
 Model 3: Multiple Channel and Exponential Service Time

average number of customers in

average waiting time in line
line (i.e., the customers in the
waiting line, use Exhibit 7.12
instead of the following formula)
Lq
Wq =
lµ (l / µ ) s
l
Lq = p
( s - 1)!( sµ - l ) 2 0

average number of customers in the average waiting time in the system

system (i.e., the customers being (i.e., waiting and being serviced)
serviced and in the waiting line)

l 1
Ls = Lq + Ws = Wq +
µ µ
Model 3: Multiple Channel and Exponential Service Time

average utilization of the system (probability that a server is busy)

1 l
r= ´
s µ

probability that a customer arriving in the system must wait for service

sµ
Pw = Lq ( - 1)
l
Waiting Line Model Notation

7- 45
Waiting Line Model Equations

7- 46
Question Bowl

If the service rate is 2 customers per hour, what is the

“average service time” for this queuing situation?
a) 40.00 minutes
b) 0.5 hours
c) 0.0667 hours
d) 16% of an hour
e) Can not be computed from data above

Answer: (b)

7- 47
 Question Bowl

If the arrival rate is 2 customers per hour, what is

the “average time between arrivals” for this
queuing situation?
a) 30.00 minutes
b) 0.6667 hours
c) 0.0667 hours
d) 16% of an hour
e) Can not be computed from data above

Answer: (a)
7- 48
Example 7.1
¨ Western National Bank is considering opening a drive-
through window. Management estimates that customers
will arrive at a rate of 15 per hour, the teller staffing
the window can serve customers at a rate of one every
three minutes (20 per hour).
¨ Management would like to know
¤ Utilization rate of the teller
¤ Average number in the waiting line
¤ Average number in the drive-through system
¤ Average time in line
¤ Average time in the system, including service

7- 49
Example 7.1 – Model 1: Simple Model
𝜆 15
¨ Average utilization is rho (ρ) 𝜌= =
𝜇 20
= 75%

𝜆% 15%
¨ Average number in the line 𝐿$ = = = 2.25
𝜇(𝜇 − 𝜆) 20(20 − 15)
is Lq
¨ Average number in the 𝜆 15
𝐿& = = =3
𝜇 − 𝜆 20 − 15
system is Ls
𝐿$ 2.25
¨ Average waiting time in line 𝑊$ = = = 0.15 ℎ𝑜𝑢𝑟
𝜆 15
is Wq
¨ Average waiting time in 𝑊& =
𝐿&
=
3
= 0.2 ℎ𝑜𝑢𝑟
system is Ws 𝜆 15

7- 50
Example 7.1 – Excel Solution

Excel: Queuing
Calculations
Instructor Slides 7- 51
 Example 1: (Model 1)

Assume a drive-up window at a fast food restaurant. Customers arrive

at the rate of 25 per hour. Only one employee works at the
restaurant. The employee can serve at the rate of one customer every
two minutes. Assume Poisson arrival and exponential service rates.

Determine:
A) What is the average utilization of the employee (probability that the
employee is busy and a customer has to wait)?
B) What is the average number of customers in line per hour?
C) What is the average number of customers in the system per hour ?
D) What is the average waiting time in line?
E) What is the average waiting time in the system?
F) What is the probability that exactly two cars will be
in the system? 7- 52
Example 2: Model 2
An automated pizza vending machine heats and
dispenses a slice of pizza in 4 minutes (constant service time) .

Customers arrive at a rate of one every 6 minutes with

the arrival rate exhibiting a Poisson distribution.

Determine:

A) The average number of customers in line.

B) The average total waiting time in the system.

7- 53
Example 7.3: Good application of Model 3

¨ Study this example to see how Exhibit 7.12 is used.

¨ Good application to see how to determine how
many servers should be in a system based on costs

7- 54
Computer Simulation of Waiting Lines

Some waiting line problems are very complex

• Servers with different capabilities, multiple customer types

Equations assume that waiting lines are independent

• When one service is the input to the next, we can no longer use the
simple formulas
Some problems have conditions that do not meet the
requirements of the equations
• Finite populations, specific arrival/service distributions

7- 55
Next Time

¨ Test 1coming – Watch for Announcement on

Moodle.
¨ We have class after the exam, please attend
class
¨ Read and Review today’s chapter, solve the
exercise problems
¨ Solve end of chapter suggested problems 16, 17
¨ Read Chapter 10 Quality Management for next
class

7- 56