AsiaPacific Symposium

1
Asia-Pacific Symposium
Optimizing the Airline Business Cycle
Macao, 16-17 September 2004
Revenue Management with Restriction Free
Pricing Vish A Viswanathan Vice President,
Consulting Sabre Airline Solutions
2
Presentation Outline

• Sabre Holdings Introduction
• Typical Pricing Structures
• Pricing Structures New paradigms
• Low Cost Carrier Effects
• Traditional Revenue Management approaches and
  effects
• New Revenue Management Solution
• Solution scalability and complexity
• Discussions

3
Who is Sabre Holdings?
A world leader in travel commerce, retailing
travel products, and providing distribution and
technology solutions for the travel industry
4
Sabre Holdings Businesses
5
Sabre Holdings Fast Facts

• Industry leader in multiple travel channels
• Revenues of 2.05 billion in 2003
• SP 500 company
• NYSETSG
• Headquarters in Dallas/Fort Worth, Texas
• 6,200 employees in 45 countries

6
Sabre Holdings Worldwide
Top 10 Largest Offices
1. Dallas/Fort Worth, TX 2. San Antonio,
TX 3. Plains, PA 4. Clintwood, VA 5. London,
England 6. Sydney, Australia 7. San
Francisco, CA 8. Sao Paulo, Brazil 9. Tulsa,
OK 10. New York, NY
5
3
9
7
10
1
4
2
8
6
7
Travelocity

• 1,000 customer servicerepresentatives available
  24/7
• 41 million members
• 2003 gross bookingsof 3.9 billion
• Travelocity Business

8
Sabre Travel Network

• Sabre Travel Network connects approximately
  53,000 travel agency locations around the world
  in 113 countries on six continents
• Includes content from more than 400 airlines,
  60,000 hotel properties, 41 car rental companies
  and 220 tour operators
• Worldwide leadership in booking share of 36

9
Sabre Airline Solutions

• Airline Products and Services
• Provides a broad rangeof software solutionsto
  airlines
• Reservations andDeparture Control
• Supports criticalairline functions sales,
  distributionand daily operations
• Hosts 90 airlines
• Consulting
• Focused industry experts to helpairlines achieve
  optimum results

Sabre Airline Solutions
Airline Reservations
Flight Operations
BusinessProcessManagement
Solutions at the Airport
CargoManagement
Maintenance and Engineering
Planning and Scheduling
Dining and Cabin Services
Crew Management
Fares Management and Revenue Management
Finance Systems
10
Sabre Consulting
A Trusted Advisor to the Travel
Industry Delivering Demonstrable Value
Decision Support Systems
Data Sources
Industry Experience
11
Sabre Consulting Solution offerings
Commercial Passenger Management Customer
Loyalty Planning Scheduling Market Data
Analysis Revenue Management Fare
Management Finance
Operations Flight Operations Crew Airport
Resources Maintenance Engineering In-flight
Services Cargo Management
12
Typical Pricing Approaches
How do you get to be a millionaire? Be a
billionaire and start an airline. Sir Richard
Branson
12
13
Typical Pricing Structure
Y Y6S KVE14N KSE14N
KFSE21N KUE14N SVVE14N SA3S WGE30N WSE30N
WGE14AN
Fare Class Alignment

• Availability is controlled by
• Validity of Fares
• Fare Restrictions
• Inventory Control

Fare Restrictions
14
Pricing Restrictions
Saturday Night Stay
Min Stay Requirements
Point of Sale Restrictions
Max Stay Requirements
Adv. Purchase Restrictions
Distribution Channel Restrictions
Routing Restrictions
Exchange, Cancellation Penalties
Simplification starts in a rather unexpected
manner
15
Pricing Structures New Paradigms
How do you get to be a millionaire? Be a
billionaire and start an airline. Sir Richard
Branson
15
16
Pricing Structure changes New Paradigm

• Multiple fares being offered with no restrictions
  affecting one or more leg/compartments
• Two main reasons for proliferation
• Low Cost Carrier Competition
• Research into LCC affiliates
• Various Types
• All Markets, All Fares
• Some Markets, All Fares
• Some Markets, Some Fares
• Point-to-point markets only
• Point-to-point and Connection Markets

17
Low Cost Carriers Effect
How do you get to be a millionaire? Be a
billionaire and start an airline. Sir Richard
Branson
17
18
Low Cost Carriers Effect
Only one fare in a market at any time

• A simpler pricing structure
• Changes to pricing fences
• None at some airlines
• None for a price range
• Non-refundable bookings
• Exchange fee
• Possibility of a higher fare
• Lower distribution costs
• Inventory Control with revenue management
• Price is the main differentiator in market
  segmentation

Fare changes are becoming very dynamic

• Price becomes the main differentiator of market
  segments
• Restriction-free Pricing becomes the norm

19
Segmentation of Low Cost Carriers
20
From LCC to Majors

• LCC arent the only ones offering fares with
  fewer or no restrictions
• Majors Alitalia, Singapore, Lufthansa, British
  Airways
• Majors within US
• New one-way fares with minimal restrictions from
  Delta, United, US Airways, Hawaiian
• Two main reasons
• pressure from Southwest, jetBlue, AirTran
• Research into launching their own LCC

Time to take a closer look into applicability of
traditional revenue management techniques
21
Traditional Revenue Management Approaches
If I can get a 7 flight to somewhere within 200
miles of Venice, well Ill _at_ take it. Seven
quid, I dont care where I _at_ go. Sir Bob
Geldof
21
22
Traditional RM Approach

• Assumption of demand independence among all
  classes within a cabin
• Protection for higher classes assumes
  Infeasibility of Sell-down
• Market Segmentation Assumptions
• Segment restrictions included in pricing
  definitions
• Independence among market segments (distribution
  channel, point-of-sale)
• Effect of external conditions on market
  segmentation

23
Traditional RM Approach

• Save seats for High Paying Late Coming
  passengers
• No restriction to FORCE Late Coming
• Passengers either buy lower fare or move to other
  airlines (if lower fare is closed)
• Seats saved for Late Coming demand is spoilt
  resulting in revenue dilution

• For low load flights, lots of space available for
  lower fare
• No restrictions to avoid sell-down of business
  demand
• Results in revenue dilution

Restriction-free pricing structure violates basic
market segmentation assumptions related to
traditional RM
24
Revenue Dilution
High Load Flights
25
Revenue Dilution
Low Load Flights
26
Inventory Controls

• Initially No Revenue Management System Support
  Available
• Assumption of demand independence is not valid
  anymore
• Airlines had to resort to manual controls using
  data from existing databases
• Development of new algorithms needed

27
Inventory Controls

• Types of leg-based controls
• Time to departure Close-out (Leg-based)
• Easy to manage flights
• Works well in a point-to-point environment
• Market segmentation becomes a continuous exercise
• Frequent performance measurements needed
• Monitoring of competitive landscape becomes very
  important
• Allocation-based controls with final Close-out
  period
• Typical in a mixed compartment environment
• Subjective Allocation overrides
• Loss of opportunity based on final close-out DCP

28
Revenue Management Solution
How do you get to be a millionaire? Be a
billionaire and start an airline. Sir Richard
Branson
28
29
Revenue Management Solution
The Inventory Control Question is
When to move up from the current available fare
in any market? After x number of bookings? At y
days before departure?
Fare changes related to booking changes?
30
RM Solution Outline
Class Closure W either at day 84 or after10
bookings M either at day 35 or after 27
bookings K either at day 14 or after 60
bookings B either at day 5 or after 72 bookings
Fare Structure / Exp Bookings
B
K
M
Fare Structure
W
Cum. Exp. Bookings
31
RM Solution Outline

• Given Price-Demand Relationship with respect to
  Days Prior to Departure
• Calculate Optimal Booking Limits
• Closure point for each fare

Optimization Step

• Evaluation of Price-Demand Relationship with
  respect to Days Prior to Departure

Price Sensitivity Step
32
RM Solution Outline

• Historic Price points and related bookings
• Inventory controls open/close time period
• Grouping of historical data Seasonality

Assuming only one price point, expected bookings
at days prior to departure
Available Data for Calibration
33
RM Solution Outline
Assuming only one price point, expected bookings
at days prior to departure

• Concept of Sell-up and Sell-down
• Demand for a 50 fare deduced from observed
  bookings at all
• Demand for a 50 fare deduced from observed
  bookings at all 50 fare in a time-band
  Sell-down
• Calibration of Sell-up Sell-down curves

34
RM Solution Outline
Booking Profile for each price point, assuming
that was the only one in that compartment
35
RM Solution Outline
Inventory control in this period affects price
sensitivity in future
Optimization Step
36
Optimization Step

• Calibrated price-demand curve
• Calculated available space
• Identified open price points

• Optimization Allocation at a price point to
  maximize total expected revenue
• Moving from lowest eligible price point to the
  highest

Demand at a future time is a function of current
allocation
37
RM Solution Outline
Fare Curve Update
Price-demand relationship updated dynamically
Dynamic Fare Curve
Cum. Exp. Bookings
38
Solution Scalability

• Population data vs. one airline sample
• Estimation based on competitor fares
• Flexible Data Grouping
• Schedule, Seasonality
• Dynamic Pricing Capability
• Alternate/Advanced models to estimate
  price-demand relationship

39
Solution Scalability

• Complexity
• Refundable/Non-refundable fares
• Biased Data
• Bookings based on previous definition of classes
• Marketing Promotions Competitive Issues
• Code-share Alliance Partner Bookings
• Group Bookings (Dedicated Class vs. All Classes)
• Nesting Structure

40
Discussions

• THANK YOU !!

41
Contacts

• Vish A Viswanathan
• Vice President, Sabre Airline Solutions
• 91 98404 96765
• vish.viswanathan_at_sabre.com