current revenue management at vietnam airlines

TABLE OF CONTENTS
Page
- Secondary data: 13
3.2.4. Overbooking application 38
Alliances 64
ACKNOWLEDGEMENT
I am pleased to send our sincerest thanks to all organizations and individuals
who have enthusiasm for helping the authors carried out and completed this thesis.
I would like to give my grateful appreciation to all the lectures, tutor;
classmate who helped me during my studying times at the EMBA program of
Business school of the national economy University.
Especially, I would like to express the deep gratitude to Doctor Tran Thi
Hong Viet for her enthusiastic guidance and encouragement that inspired the author
a lot during the process of writing the thesis.
I would like to acknowledge my colleges and experts; managers of
departments of Vietnam Airlines for providing the material needed for the thesis.
I owe a great deal to my family, relatives and friends, especially to my
parents, my wife and my children for their valuable encouragement and unceasing
supports in my academic pursuits
3
ABBREVIATION
CRM Customer relationship management
CX Cathay pacific airways
HAN Ha Noi
JL Japan airlines
O-D Origin – Destination
PAX Passenger
PNH Pnompenh
REP Siem Reap city
RM Revenue Management
SGN Ho Chi Minh city

SQ Singapore airlines
VNA Vietnam airlines
4
LIST OF TABLES
- Secondary data: 13
3.2.4. Overbooking application 38
Alliances 64
5
LIST OF FIGURES
Figure 1.1 Research Procedure 11
- Secondary data: 13
3.2.4. Overbooking application 38
Alliances 64
- Secondary data: 13
3.2.4. Overbooking application 38
Alliances 64
6
TERMINOLOGIES
In the thesis, there have been some Airlines terminologies. In order to agree upon
the meaning of these technical words and avoid misunderstanding, the following
words are understood as follow:
Airlines Alliances: is an agreement between two or more airlines to cooperate on a
substantial level. (The three largest passenger alliances are the Star Alliance,
SkyTeam and Oneworld).
Cancellation: guests make reservations for flight but then cancel the reservation.
Denied Boarding passenger: Passenger have confirmed ticket and reservation but
airlines can not arrange seats on flight due to overbooking and all seats of flight
occupied at departure times of flight.
Forecast: the process of predicting events and trends in business
Load factor: is a measure of the amount of utilization of the total available capacity

of a commercial transport vehicle. It is useful for calculating the average occupancy
on various routes of airlines.
Low-cost airlines :( also known as a no-frills, discount or budget carrier or airline)
is an airline that offers generally low fares in exchange for eliminating many
traditional passenger services
No show: a guest who made a reservation but not show up at check-in counter for
departure.
Overbooking: accepting more reservation than the seat of flight available.
Over-selling: Selling more seats on flight than those available.
Spoilage: results from the inability to sell seats for a certain period of time and thus
causes the permanent loss of revenue for that period.
7
EXECUTIVE SUMMARY
Airlines have faced with the empty seats after flights departure due to the no-
shows and cancellations of passengers at last minute, the empty seats of flight will
cause the loosing revenue of airlines. With a fixed cost that airlines have to spend
for every flight (fuel, aircraft…), filling up the seats and the load factor, fare mixes
and selling up are necessary requirements for the airlines to maximize the revenue
and lead to gain optimal profit. And to solve these problems, airlines apply the
Revenue management system. Airlines management (or yield management) is
defined as to maximize passenger revenue by selling the right seats to the right
customers at the right time.
In this thesis, base on the theory of revenue management and knowledge of
operation management subject studied, I analyzed Revenue Management activities
at Vietnam Airline.
Vietnam airlines have deployed the revenue management system for years in
its operation, such as forecast demand, seats inventory control, fare mix and
overbooking application, these also brings a given success for Vietnam airlines.
However, there some problems such as still exist the empty seat on plane after plane
takeoff although the booking system had sold over capacity of airplane or existing

the cases that forced to deny boarding passengers due to can not arrange seats for
them because of deploying overbooking policies, this happened has affected to the
revenue and bad image of Vietnam airlines.
After analyzing the situations at Vietnam airline in terms of history, staffing,
facilities and specially the RM activities from fare structure, overbooking policies
and seats inventory control, annual reports gained from 2005 to 2008, and the
interviews with the experts and manager working at revenue management, the
research has identified some problems and shortcomings that Vietnam Airline is
8
facing as follow: 1) Unreasonable overbooking in low frequency routes , 2)
Revenue management Control system, 3) Flight Report system, 4) Unstable
schedule leads to difficult for forecasting demand and collecting data, 5)Lack of
capacity in peak season, 6) Compensation policy for denied boarding passenger, 7)
Human resources.
These problems and shortcoming have influenced the effective operation of
Vietnam airline especially in maximizing revenue and profits.
To help solves these problems and shortcomings some recommendations were put
forward. They are: 1)Applying appropriate overbooking ratio, 2) Increasing
compensation for denied boarding passenger,3) Maintain the stable flight schedule,
4)Good capacity and demand forecast, 5)Training human resources, Improving
sufficient and efficient information database system, 6)More flexible fare-mix
policies.
9
CHAPTER 1. INTRODUCTION
1.1 Rationale of research
For airlines industries, revenue management plays a very important role for
existence and development. With a fixed cost that airlines have to spend for every
flight (fuel, aircraft…), filling up the seats and the load factor, fare mixes and
selling up are necessary requirements for the airlines to maximize the revenue and
lead to gain optimal profit.

Vietnam airlines have deployed the revenue management system for its
operation such as controlling space and applying fare mixes in sale system
especially in applying the overbooking policies on reservation system, and its also
brings a given success for Vietnam airlines. However, there some problems such as
still exist the empty seat on plane after plane takeoff although the booking system
had sold over capacity of airplane or existing the cases that forced to deny boarding
passengers due to can not arrange seats for them because of deploying overbooking
policies, this happened has affected to the revenue and bad image of Vietnam
airlines.
So, there need to have an analysis revenue management to evaluate how well
it has been applied and has worked, to identify if there are ways for improvement or
necessary adjustment.
1.2 Research problem
The research is designed to analyze the revenue management system at
Vietnam airlines especially in operation as overbooking, Fare structure, and
Demand forecast, control revenue management system fields to identify possible
improvement or adjustment for that system operating more efficiency.
10
1.3 Research objectives
With the identified problem, the following objectives are set for the research:
• Provide an overview of Vietnam airlines business activities related to
Revenue management system.
• Analyze the revenue management at Vietnam airlines in operational field.
• Find out the problems of the existing Revenue management at Vietnam
airlines
• Propose some recommendations for improvement to revenue management of
Vietnam airlines.
1.4 Research scope and limitation
The research focuses on the working or revenue management in the period
from 2005 to 2008

Time and budget is limited, and the data collection in Vietnam airlines HDQ
from 2005 to 2008.
1.5 Research methodology
Figure 1.1 Research Procedures
Source: Author’s Summary
Secondary Data
report, statistic, publication,
previous study
by VNA
Secondary Data
report, statistic, publication,
previous study
by VNA
Primary Data
Qualitative: 4 deep
interviews, obrervation
Primary Data
Qualitative: 4 deep
interviews, obrervation
Analysis &
Findings
Analysis &
Findings
Recommendations
for improvement
Recommendations
for improvement
11
Research method
In this thesis, for the nature of the study about the operation of a system,

qualitative research method was used intensively. The data were collected from
secondary and primary sources. Qualitative method applied to find out perception,
process and evaluation about Revenue management at Vietnam airlines. The
following sections presents in details the way of collecting and analyzing the data
sources used in this study.
Data collected
- Primary data:
To assist in the identification of causes mentioned earlier, 4 in depth
interviews with experts and managers in airline industry and RM system technicians
are to be done, the first interview with Ms Nguyen Minh Hien – Helpdesk Manager
of Space control center- Vietnam airlines, the second interview Mr Nguyen gia Loc
– Director of Space control – Vietnam airlines, the third interview Mr Hoang Thanh
Quy – deputy Director of Marketing and sales Department – Vietnam airlines, the
forth interview Mr Dang Anh Tuan – Director of Noibai Operation center –
Vietnam airlines.
The purpose of the interviews is to get the experts’ ideas and opinions about
the revenue management system of Vietnam airlines, what they have deployed for
the revenue management system and how they run the system and how it has
operated.
The interviews shall not go into technical aspect but to the managerial aspect
of using the system for decision making.
The interviews also aim at getting ideas from experts on strategies and
consideration in order to apply RM effectively.
With my ten years experiences working in Vietnam airlines at operational
field, I have analyzed about the revenue management system and discovered some
12
the limits and disadvantages of Vietnam airlines revenue management system need
to improve.
- Secondary data:
The second data composes a wide range of information published in relevant

material. The main source is from internal records of Vietnam airlines. Especially it
includes: Sales report, revenue management report and statistics collected by VNA
Space control center, RM department in five year (2005-2008).
1.6 Research Structure
Besides the parts of introductions, conclusion, reference and appendix, the
content of this thesis includes 3 main chapters starting from chapter 2
Chapter 2: Theoretical Background
In this chapter provide the concept of revenue management, definition of
revenue management, the process of revenue management in airlines industry’s
application. These the basic for analyzing the revenue management activities in
Vietnam airlines at chapter 3
Chapter 3: Current Revenue management at Vietnam airlines
In this chapter, provide the overview of Vietnam airlines business, revenue
management activities that Vietnam airlines have implemented, such seats
inventory control, fare structure, reservation system, overbooking application, the
organization of revenue management of Vietnam airlines. Provide identifications
problem in Vietnam airlines revenue management activities
Chapter 4: Recommendation to improve revenue management at
Vietnam airlines
In this chapter, provide the overview of strategic of Vietnam airlines to 2020
and draw the recommendation to improve revenue management activities
13
CHAPTER 2.THEORETICAL BACKGROUND
2.1 Concept of Revenue Management
2.1.1 Definition of Revenue Management
There are several definitions of revenue management (also refer to as yield
management) in the literature. American Airlines (1987) defined the goal of yield
management as to “maximize passenger revenue by selling the right seats to the
right customers at the right time.”[1, p22-p25] Pfeifer (in1989) described airline
yield management as “process by which discount fares are allocated to scheduled

flights for the purposes of balancing demand and increasing revenues.”[7, p149-
p157] From the hotel industry’s perspective it have been defined as “charging a
different rate for the same service to a different individual” [5] and “controlling the
trade off between average rate and occupancy” [6].
Weatherford and Bodily (1992) have concluded from the above definitions
that the term yield management is too limited in describing the broad class of
revenue management approaches.[11] After analyzing situations in which yield
management was used. They concluded that these situations had the following
characteristics in common:
• There is one date on which the product or service becomes available and
another after which it is either not available or it spoils. The product cannot
be stored for significant periods of time-it eventually perishes. In the grocery
store example, the fruit would spoil.
• There is a fixed number of units. Capacity cannot be charged in the short
term. In the hotel example, there are so many rooms that may be sold at a
given property location.
• There is the possibility of segmenting price-sensitive customers. In the
airlines example, vacation travelers are much more sensitive to price than
business travelers.
14
Weatherford and bodily proposed the term perishable-asset revenue management to
define this class of problems and described it as “the optimal revenue management
of perishable assets through price segmentation”
2.1.2 Origin of Revenue Management
The root of modern revenue management can be traced back to the early
days of the U.S. airlines industry. Prior to the Airlines Deregulation Act of 1979,
fares for airlines travel in the United States were regulated by the Civil Aeronautics
Board (CAB). The CAB ensured that the airlines operated in a highly controlled
environment designed to serve the public convenience and necessity.[2] The CAB
required economic justification for any fares proposed by the airlines. Thus, there

were few fares for customers to choose from. In the 1930’s all airlines offered all
seats on a flight for the same price. But it was obvious to the airlines that passengers
could be divided into two broad categories, based on their travel behavior and their
sensitivity to prices. There were business travelers and leisure travelers. Business
passengers tended to make their travel arrangement close to their departure date and
stay at their destination for only a short time. There was little flexibility in their
plans and were willing to pay higher prices for tickets. Leisure travelers, on the
other hand, booked their flights well in advance of their travel date. They stayed
longer at their destinations and had much more flexibility in their travel plans. They
would often decide not to travel rather than pay high fares. Since there was only on
fare offered to both types of passengers, many of leisure passengers chose not to fly,
and many flights departed with empty seats.
Airlines managers saw an opportunity to increase revenue by lowering fares
in certain markets. The first experiment to offer low-fare service occurred in
California on the San Francisco – Los Angles route in 1994[2]. United airlines
began its Sky Coach Service using 10-passengers Boeing 247s and charging a one-
way fare of $13.90. The CAB approved the low fares based on the lower operating
cost of the B-247s and fewer amenities offered on board. The experiment was a
15
success but ended shortly thereafter when the airline’s fleet was turn over to the
armed forces during World War II.
Throughout the next few decades, other discount fares were offered with
varying degrees of success. First – class and coach – class became standard on all
airlines. But the airlines were not permitted to offered different fares within the
coach cabin and prices were set through a cost-plus pricing formula administered by
the CAB. Carriers gradually became less efficient at operating their airlines, and
coach fares rose over time as average costs increased.
During the 1960s, the CAB began approving new types of fares such as night
coach fares and 7-21 day excursion fares based on length of stay. However, the
airlines placed no limits on the number of seats that could be sold at these fares, and

all were available on a first-come, first served basis.
In the early 1970s, the CAB responded to demand for more discount fares by
easing regulations for charter airlines. With their lower operating costs, the charter
carriers were able to offer low fares and still earn a profit. For example, in the
winter of 1976, passengers could travel from New-York to Florida on a charter for
as little as $99.[2] This fare was less than the average cost for major airlines to fly
that market. So if the airlines matched the charter fare, then it would lose money on
the flight, even if it filled every seat.
This situation caused concern among the managers at the major airlines.
Their initial though was to figure out a way to reduce costs so they could remain
competitive. But that was impractical. The costs of operating a major airline with its
staffing and airport needs were simply much higher than cost of running a charter
operation. But then the executives at American Airlines realized something. On
average, their planes were departing with half their seats empty. While the average
cost of these seats were higher than the charter fares. The marginal cost was close
to zero. So if they could find a way to sell just the empty seats at the charter fares,
profits would increase dramatically. The challenge was to devise a plan that would
make the empty seats available at the lower fare, while preventing passengers who
16
were willing to pay the higher fare from buying low-fare seats. American airlines’
response to this challenge was the introduction of “Super Saver Fares” in 1977.
With these fares came the beginning of modern day revenue management [9].
2.2 Revenue Management in airlines industry.
2.2.1 Role of Revenue Management in airlines
The airline industry is one in which production is very inflexible. Essentially,
when committing to fly a flight from A to B, an airline both fixes the level of its
output (the number of seats) and, for all practical purposes, the total cost of that
output–independent of how many customer actually fly on the flight. It’s unit cost
per seat sold, therefore, varies tremendously with the volume of sales, and once the
capacity constraint is reached, no more production is possible. Worse yet, like all

services, output cannot be inventoried, so production of air transport output in one
period cannot be used to satisfy demand in later periods (e.g., an unsold seat on
Monday cannot be used to supply the need of an excess passenger on Tuesday). All
these factors combine to create extreme inflexibility in the technology of air
transport service, and this is one of the key driving factors in the importance of RM
in this industry to maximize revenue.
2.2.2 Contents of Revenue management in airlines
2.2.2.1 Forecast demand
Forecast demand is play a important role in RM, accurate forecasts are
crucial to a RM system. Poor estimates of demand lead to inadequate inventory
controls and sub-optimal revenue performance. Forecast for airlines RM system is
inherently difficult. Competitive action, seasonal factors, the economic
environment, and contrast fare changes are a few of the hurdles that must be
overcome; in addition, the fact that most of the historical demand data is censored
further complicates the problem.
The number of seats an airline can sell on a flight is determined by the
booking limits set by the RM system. An airline continues to accept reservations in
17
a fare class until the booking limit is reached. At the point, the airlines stops selling
seats in that fare class-it also collecting valuable data. Demand for travel in that fare
class may excess the booking limit, but the data does not reflect this. So the data is
censored or “constrained” at the booking limit.
Forecast at the level required by RM system is extremely difficult. Mc Gill
and van Ryzin (1992) list the following factor as contributing to this difficulty:
Seasonality: Passenger are more likely to fly to some destinations based on
the time of year. For example, there is greater demand for flight to Europe in the
summer than in the winter.
Day-of-week and time of day variations: business traveler are more likely on
weekdays than weekends, Early morning and evening flights are desired by
business travelers who want to accomplish a day’ work at their destination and

return the same day.
Special events: Event such as the world cup cause a temporary increase in
demand at specific location
Sensitivity to pricing actions: Price increases and decrease result in demand
decreases and increases respectively, but different passenger types have different
elasticity.
Demand dependencies between fare classes: Passenger who book full-fare
seat might have met the restriction for lower fare seats. But there was no
availability.
Group booking: Groups tend to book and cancel reservations in large
numbers.
Cancellation: a RM system requires a forecast of how many passengers will
book and travel in each fare class. Since som passengers make reservation and
subsequently cancel them, this behavior must be considered.
Censoring of historical demand data: An aircraft capacity and booking limits
constrain the demand seen in the historical data.
18
No-show: some passengers make reservations, decide not to travel and do
not cancel their reservation.
Types of forecast
There are three types of forecast used in the airlines industry: Macro level,
Passenger choice modeling, and micro level.
Macro level forecast are usually made for aggregate forecasts of total airlines
passenger demand. For example macro level forecast might be a projection of total
annual domestic air travel or future travel between us and Europe.
Passenger choice models attempt to predict future demand by modeling
current passenger behavior based on socioeconomic factors and the characteristic of
travel alternative options.
Finally, micro level forecasting is used to predict passenger behavior at
disaggregate level such as flight, date, and fare.

So, based on historical data, forecast demand for airlines concentrate on the
following types:
• Passengers demand
• Revenue for a flight
• Booking class of flight
• Overbooking ratio of flight
Data for forecasting:
o Past flight data.
o Current flight performance.
o Market factors.
o Other factors affect to demand.
19
POD model:
Figure 2.1 Airlines Passenger Origin Destination simulation flow (POD model)
Sources: Hoppers tad, the Boeing Company.
According to POD model, the historical booking database collects
information from previous flights and feeds this historical data into the forecaster,
whose historical database is manually initialized at the beginning of each simulation
run. The forecaster then uses this data along with booking currently on hand
provided by the Revenue management optimizer to forecast future demand for a
given flight. These expected future bookings are the fed into the Revenue
management optimizer. With this data and actual path and class bookings and
cancellations, the Revenue management optimizer then determines seat protection
and availability. Finally, this data is fed into the passenger choice model, which
uses it to assign new prospective passengers to available path fare combination
20
according to their decision window and budget. All information is then input back
into Revenue management optimizer as historical data to be used by the airlines for
future flight departure.
2.2.2.2 Seat inventory control

Seat inventory control problem in airlines revenue management concern the
allocation of the finite seat inventory to the demand that occur over time. In order to
decide whether or not accept a booking request, the opportunity costs of losing the
seats taken up by the booking have to be evaluated and compared to the revenue
generated by accepting the booking request. Solution method for seat inventory
control problem are concerned with approximating these opportunity and
incorporating them in a booking control policy such that expected future revenue
are maximized.
Comeback to “Super saver fare” which American airlines had deployed in
1977(referred at beginning of this chapter), the Super saver fare were the first
capacity-controlled restricted discount fare. That is, they were offered in limited
numbers and certain conditions had to be met for the fare to be valid. For example,
the tickets needs to be purchased at least 21 days in advance of travel and the
itinerary had to include a Saturday night stay. These restrictions were meant to
prevent the high-fare passengers from purchasing the low-fare tickets. It reflected
the airlines’ belief that business travelers did not have enough flexibility in their
plans to meet the restrictions and, therefore, would continue to pay higher fares.
American began its Super saver fares by offering approximately 30% of the
seats on each flight to these fares [2, chapter 47]. But they soon found the number
of seats needed to be controlled carefully to increase total revenue. If too many
discount seats were sold, then the airlines would turn away late-booking high-fare
business passengers. If too few seats were sold to discount passengers, then the
planes would depart with empty seats. The correct number of seats to be allocated
to the discount passengers could only be calculated from an accurate forecast of
21
demand for high-fare ticket. Research thus began to develop the appropriate models
to forecast demand and calculate discount seat allocations.
Since the first super saver fare appeared on the market, the airlines pricing
structure has changed dramatically. Airlines publish a variety of fare in an attempt
to segment the market. Their goal is to design what is referred to as fare products

are differentiated by advance purchase restriction, minimum stay requirement and
penalties for refunds. The fare products correspond to the price elasticity airlines
have identified among their customers. For example, discount passengers who
desire a low price must be willing to purchase their tickets weeks in advance of
travel and stay at their destination for at least one Saturday night. If they cancel or
change their plans then they will be charged a penalty. On the other hand, business
travelers place a high value on flexibility. They may purchase their tickets at any
time and change their reservation without penalty. There are no restrictions on the
amount of time they must stay at their destination. For this flexibility, the business
traveler is willing to pay a higher fare than the leisure passenger [12, p92-p93]. So
while these two types of passengers may be seated next to each other on a flight.
They are paying different and receiving different products.
Airlines using single letter class codes to distinguish between fare products.
For example Y might be used for full-fare class coach, M and Q for discounts. V for
deeper discounts and others classes which vary by airlines [9, p8-p31]. There are
often six class or more different fare classes offered by a singe airlines in a given
origin and destination (O&D) market in the coach cabin of aircraft. An example of a
typical airline fare class structure is given in table 2.2
Table 2.1. Typical airline fare class structure
Fare class Fare product type
Y Full coach fare with no restrictions
B Unrestricted discount fares
M Seven-day advance purchase with minimum stay requirement
Q Fourteen date advance purchase with minimum stay requirement
V Deeply discount or industry fare use for airlines staffs
Source: Richard H.Zeni
22
Modern revenue management systems forecast demand for each one of these
fare classes by using historical booking data from the same fare class of similar
flight departures. This data is usually aggregated by departure time, day of week

and time of day [4]. These forecast are them used as inputs to optimization models
that calculate booking limits and control the number of seat available at various fare
levels.
Obviously, an airline would like to carry as many of the high-fare business
passenger as possible. Only those seats that cannot be sold to business passengers
should be made available to the leisure passengers. The problem is that leisure
passengers tend to book their reservation first. And even if they did not the advance
purchase restrictions most airlines place on leisure-fare tickets often force this
behavior. So, before any seats are sold, the revenue management system must
forecast how many business passengers will still want to book on a flight after the
leisure passengers have made their reservations. Then it must set aside or “protect”
these seats so that they will be available when the business passengers request them.
The seat inventory control problem has been approached from a variety of
perspective. Seat inventory can be controlled over individual flight legs (take off
and landing of one flight) or over the airline’s entire network. Most airlines manage
seat inventories by fare class at leg level. That is they attempt to maximize revenue
on each individual flight leg. Reservation request are evaluated by airlines based on
the availability of a particular fare class on each flight leg. A passenger’s entire
origin and destination itinerary is not taken into account when the decision is made.
2.2.2.3 Overbooking
Booking limits and Nesting
The objective of a revenue management system is to set booking limit at
different levels of control in an attempt to maximize revenue. As noted above, first
a forecast of demand is made, and then optimization is performed to calculate
protection levels. Thus, certain number of seats is protected from being sold to low-
fare passengers. The logic is that if a certain number of high-fare passengers are
23
expected to book, then seats should be set aside so that they will be available when
the request are made. For example, suppose there is a forecast for highest value fare
class(Y) and the subsequent optimization produces a protection level of 40. at least

40 seats should be protected for these Y-fare passengers and not sold to any one
else. However, that happens if more than 40 Y-fare passengers request seats? The
airlines would not want to deny these requests. To eliminate the possibility of
turning away high-fare passengers when there are seats available, airlines
reservation systems usually nest the booking limit.
Nesting allows high-fare passengers to book seats that are available to lower-fare
passenger. Any seat available at a particular fare also be available at a higher fare.
In example in table 2.2 there are 100 seats available to be sold on the flight leg.
Forty seats are being protected for Y passenger. But the entire inventory of
seats(100) is available to booked by these passengers. So the booking limit for Y is
100. to arrive at the booking limit for M, the protection level for Y is subtracted
from the total capacity on the aircraft(100-40=60). Now suppose the protection
level for the Y/M nest is 65, So 65 seats are protected for sale to Y or M passenger.
To arrive at the booking limit for V, the Y/M protection level is subtracted from the
remaining capacity(100-65=35), therefore, the booking limit is given by:
Booking limit
i
=(C- Ө
i-1
) [ 8, p16]
Where C is the remain capacity in the aircraft canbin and
Ө
I
is the protection level of the i
th
fare class
Table 2.2. Nesting on a flight leg
Fare class Total protection level Ө
i
Nested booking limit(C- Ө

i-1
)
1(Y) 40 100
2(M) 65 60
3(V) 35
Source: Richard H.Zeni
with Ө
0
=0 and C= 100
Overbooking.
24
Overbooking practice of intentionally selling more reservations for a flight
than there are actual seats on the aircraft, Airlines use overbooking to offset the
effects of passengers cancellations and No-show.
Formular:
Where:
P(d<x) = Cumulative Probability
x = seats reserved for full fare
d = demand for full-fare tickets
C
u
= Lost of revenue associate with reserving to few seats at full-
fare(underestimate)
C
o
= Lost of revenue associate with reserving to few seats at full-
fare(overestimate).
2.2.3 Process of Revenue Management in airlines
A brief description of the generic operations of a RM system, This introduces
the key components and gives an overview of the information flows, controls and

design of a RM system.
RM generally follows four steps:
1. Data Collection Collect and store relevant historical data (prices, demand, causal
factors, no-show passengers).
2. Estimation and Forecasting Estimate the parameters of the demand model;
forecast demand based on these parameters; forecast other relevant quantities like
no-show and cancellation rates, based on transaction data.
3. Optimization Find the optimal set of controls (allocations, prices, mark-downs,
discounts, overbooking limits, scheduling) to apply until the next re-optimization.
C
u
C
u
+ C
o
P(d<x)
≤
[3, p367]
25
4. Control system the sale of inventory using the optimized control. This is done
either through the firm’s own transaction-processing systems or through shared
distribution systems (e.g., GRSs, GDSs).
The RM process typically involves cycling through these steps at repeated
intervals. The frequency with which each step is performed is a function of many
factors such as the volume of data, how fast business conditions change, the type of
forecasting and optimization methods used, and the relative importance of the
resulting decisions. For example, most RM systems in airline and hotel applications
stagger the dates (called Data Collection Points (DCPs) when they collect data,
reforecast and re-optimize, with the cycle occurring more frequently (at least daily)
as the service time nears. This is because in these industries, a substantial portion of

the reservations occurs during the last few days before the time of service.
Figure 2.2: Revenue Management process flow
Source: American airlines
26