Pricing in Supply Chains: Airline Revenue Management

1
Pricing in Supply ChainsAirline Revenue
Management

• Salih ÖZTOP
• 20202801
• 14.05.2003

2
1. Introduction

• Aim of the paper
• To give definitions and insights about the
  pricing of seasonal products and give some future
  research directions such as pricing decision with
  substitute or complementary products.

3
Revenue Management

• Selling a fixed stock of items over a finite
  horizon for maximizing profit (revenue)
• Examples
• Hotels renting rooms
• Seasonal goods of retailers (swimsuits)
• (your additional orders cannot reach you during
  the seasonfixed stock)
• Airlines seats (you cannot sell todays flight
  ticket tomorrowfinite horizon)

4
Airline Revenue Management

• Short term costs are mostly fixed
• ex.Fuel usage for one flight
• Variable costs per passenger are small
• ex.Beverage cost per passenger
• Pricing policy (booking) is the
• most important tool for
• profit maximization.

5
First Airline Pricing Problem

• In 1970s British Airways offered bookings with
  discounted fares bought at least 21 days before
  the departure.
• Problem Determine the number of seats
  (protected)reserved for late booking with full
  fare
• Low Protection Full fare demand is not satisfied
• High Protection Departing with empty seats

6
Airline Revenue Management

• Two fare classes Discounted fare and full fare
• Decide whether accept or reject the low fare
  booking request when you have left some amount of
  seat capacity.
• Littlewoods Rule(1972) Two fare classes, Accept
  the discount fare customers as their revenue
  value exceeded the expected marginal revenue of
  future full fare sales
• RdiscountRfull P(Full fare demand remaining
  seat )

7
Factors Effecting Booking Decisions

• Relative fare prices
• Reservation price Maximum price that a customer
  is willing to pay for a product
• Itineraries (leg combinations,Origin-Destination)
• Season
• Day of the week,Time of the day
• Cancellation Penalties and Restrictions
  (Ex.Saturday night stays) To prevent full fare
  passenger taking the advantage of promotional fare

8
Network Effect

• First airline pricing studies considered single
  legs only
• Today's airline network structure is more
  complicated and multiple legs forms one itinerary
  and different combinations can be used for the
  same origin destination pair.

9
Example Airline Network

• Leg 1Leg2Itinerary 1
• Leg 3Itinerary 1

London
New York
Istanbul
Leg 2
Leg 1
Paris
Ankara
Leg 3
Destination
Origin
10
Demand Models used

• Random variable Demand for different legs
  assumed as random variables
• Stochastic(Depends on t)
• Demand of leg depends on
• Price of leg
• Price of other legs
• Remaining time before departure

11
Pricing Decision
When making pricing decision seller knows only
the probability distribution of the reservation
prices. Therefore seller faces a trade off
between Losing sales due to high prices Losing
consumer surplus due to low prices Consumer
SurplusDifference between reservation price and
product price (Bitran Mondschein 1997)
12
Models

• Dynamic Programming used (Gallego Van Ryzin
  1997)
• Price Set (P1,P2..,Pn)
• Null Price For closing one fare class, price of
  that fare assigned to extremely large value say
  infinity.
• Deterministic demand case was used to determine a
  bound
• Heuristics were developed

13
Models

• Switch price
• 1) Switch when a certain amount of time passed
• 2) Switch when certain amoun of seat left
• Feng Gallego 1995 (DP)
• They cannot be optimal since revenue depends both
  the time-to-go and the remaining seat number.
• Result Increase the price as soon as the
  remaining time-to-go is above a time threshold
  depends on number of unsold seats

14
Substitutate/Complementary

• Complementary Products Ex. Razor and Blade,
  price of Razor determines the Blade
  (complementary) sales amount
• Substitute Products Different leg combinations
  gives same Origin-Destination pair and price of
  them effects the capacity usage of each leg.

15
Effects on Demand

• In general Price is a function of items demand.
• Price of Substitute products
• Price of Complementary products
• Bundling
• Pure Bundling
• Mixed Bundling

16
ConclusionResearch Directions

• Although the airline revenue management is as
  early as 1970s they mostly consider the problem
  under single legs and given prices. However
  determination of prices and increasing network
  effect makes this issue still hot.
• Including Group Booking

17
ConclusionResearch Directions

• Effect of Substitutable/Complementary
  Products(Bundling)
• Including other firms price decisions (Game
  theoretical approaches)

18
References
Belobaba 1987 Airline Yield ManagementAn
overview of seat inventory control Weather,
Bodily 1992 Taxonomy and Research overview of
Perishable-Asset Revenue Management Yield
management Overbooking and Pricing McGill, Van
Ryzin 1994, Revenue Management Research
Overview and Prospects Bitran, Mondschein 1997
Periodic pricing of Seasonal Products in
Retailing Feng, Gallego 1995 Optimal Starting
Times for End-of-Season sales and Optimal
Stopping Times for Promotional sales Gallego,
Van Ryzin 1997 A Multiproduct Dynamic Pricing
problem and its Applications to Network Yield
Management
19
5. Questions Answers
Thank you