MIT ICAT. MIT ICAT M I T I n t e r n a t i o n a l C e n t e r f o r A i r T r a n s p o r t a t i o n

Transcription

1 M I T I n t e r n a t i o n a l C e n t e r f o r A i r T r a n s p o r t a t i o n BENEFITS OF REVENUE MANAGEMENT IN COMPETITIVE LOW-FARE MARKETS Dr. Peter Belobaba Thomas Gorin IATA REVENUE MANAGEMENT CONFERENCE Budapest, Hungary October 21, 2004 MOTIVATION Recent rapid growth of low-fare airline (LFA) competition worldwide LFAs now carry more than 25% of US domestic passengers Growing LFAs in Canada, Europe, Asia, South America, Australia and New Zealand Questions about RM system performance in face of LFA competition and changing fare structures How is the effectiveness of basic Leg RM affected for both network and low-fare airlines under lower less restricted fares? What are the impacts on revenue gains of Network RM (O+D Control) in hybrid networks with standard and LFA markets? 2 1

2 PODS RM RESEARCH AT Passenger Origin Destination Simulator simulates impacts of RM in competitive airline networks Airlines must forecast demand and optimize RM controls Assumes passengers choose among fare types and airlines, based on schedules, prices and seat availability Recognized as state of the art in RM simulation Realistic environment for testing RM methodologies, impacts on traffic and revenues in competitive markets Research funded by consortium of seven large airlines Findings used to help guide RM system development 3 BASELINE HUB NETWORK (NO LFA) H1(41) H2(42) Traffic Flows 4 2

3 BASELINE NETWORK CHARACTERISTICS 482 Origin-Destination Markets 20 cities on each side of two competing hubs 82 non-stop local markets to/from/between hubs 400 connecting O-D markets Two competing network carriers Each operates three daily flights between each spoke city and its own hub aircraft capacity 100 seats Connecting banks at hubs allow three connecting frequencies in each west to east O-D market Standard fare structure and restrictions Y, B, M and Q increasingly restricted (advance purchase and restrictions) Y is the full fare class, B, M and Q are discounted classes 5 REVENUE MANAGEMENT METHODS Leg RM: Fare Class Yield Management (FCYM) Demand forecasting by fare class and leg Leg-based EMSRb seat protection algorithm for nested booking limits applied to 4 fare classes Re-optimization of booking limits 16 times before departure Network RM: Displacement Adjusted Virtual Nesting (DAVN) Forecasting by origin destination, itinerary, fare class ( ODIF ) Network-based optimization to find leg displacement costs Mapping of ODIFs to virtual buckets based on total fare minus displacement costs; booking limits applied to virtual buckets Re-optimization and re-mapping of ODIFs 16 times before departure 6 3

4 LEG RM REVENUE PERFORMANCE INCUMBENT AIRLINES 1 AND 2 Leg RM revenue gains compared to both carriers using first come, first served (FCFS) Typical 5-8% revenue gains at load factors approximately 83% 9.0% 8.0% 7.0% 6.0% 5.0% 4.0% 3.0% 2.0% 1.0% 0.0% 8.63% vs. FCFS/FCFS 5.72% Airline 1 Airline 2 7 BASELINE NETWORK RM GAINS: INCUMBENT AIRLINES 1 AND 2 Network RM gains compared to both carriers using Leg RM only Network RM gains are substantially larger when only 1 airline implements O-D control 2.0% 1.5% 1.55% 1.0% 0.5% 0.0% 0.76% 0.70% Airline 1 Airline 2-0.5% -1.0% -0.76% Network RM vs. Leg RM Network RM vs. Network RM 8 4

5 NETWORK RM GAINS GIVEN COMPETITOR WITH NETWORK RM Network RM revenue gains when Airline 2 moves from Leg RM, given Airline 1 is using Network RM Airline 2 gains 1.5%, again partially at expense of Airline % 1.5% 1.0% 0.5% Airline 1 Airline 2 0.0% -0.5% -1.0% -0.78% 9 RM PERFORMANCE DEPENDS ON COMPETITIVE RM SITUATION RM revenue gains come at least in part from relative sophistication of competitor s RM approach Strong RM against weak RM leads to feedback loop Weaker RM is unable to protect for higher valued customers, opening up too many low-fare seats Under high demand situations, airline with weaker RM fills up, leaving highest revenue consumers to airline with stronger RM Stronger RM then further protects for future high-fare demands, forcing more low-fare traffic on the weaker competitor Competitive RM lessons become especially important when network carriers compete against low-fare airlines 10 5

6 LFA COMPETITOR ADDED TO NETWORK Third carrier added in the highest-demand local markets of Airline 1, between hub and 10 spokes Airline 3 operates the same schedule as Airline 1 in these 10 markets, to remove time of day schedule effects Airline 3 offers strictly point-to-point local services (no connections) Focus of this presentation is on Airlines 1 and 3 Airline 2 is much less affected by LFA entry than Airline 1 In simulation scenarios, Airline 3 s capacity is set to 30 seats per flight Compared to 100-seat aircraft operated by network carriers 11 ENTRY BY LFA INTO 10 MARKETS H1(41) H2(42) Traffic Flows 12 6

7 LFA UNRESTRICTED FARE STRUCTURE LFA enters with a lower unrestricted fare structure Unrestricted Y fare 50% lower than existing full fare Lower Q fare to stimulate demand ($10 cheaper than the preentry Q fare) B and M fares spread evenly between Y and Q fares Fare Class Original Fares Restrictions Advance Purchase Fare Class New Entrant Fares Restrictions Advance Purchase Y B M Q $400 $150 $100 Min. Sat. Sat, NR * Sat, NR *, CF* * 7 days 14 days 21 days Y B M Q $163 $127 $90 * Non-refundable * * Change Fee 13 LEG RM FOR UNRESTRICTED FARES Leg RM systems developed for segmented demand Assumed independent fare class demands, because restrictions kept full-fare passengers from buying lower fares With unrestricted fares, passengers buy lowest available fare EMSRb (without adjustments for sell-up) will not perform well in unrestricted fare structures Previous buy-down is recorded as lower fare demand Unless demand forecasts are adjusted to reflect potential sell-up, high-fare demand will be consistently under-forecast EMSRb then under-protects, allowing more buy-down to occur To replicate simplest LFA RM approaches, booking threshold methods were simulated 14 7

8 REVENUE MANAGEMENT METHODS LEG RM (EMSRb) Used by incumbents and/or low-fare new entrant NETWORK RM (DAVN) Used only by incumbent Airline 1 FIXED THRESHOLD New Entrant Only Simple closure of fare classes based on booked load factor Threshold Levels: 75%, 60%, 45% ADAPTIVE THRESHOLD New Entrant Only Target load factor used to compute Threshold multiplier Multiplier applied to threshold levels once per flight departure Used target LF of 70% which proved to be best performer 15 RESPONSE 1: LIED FARE STRUCTURE MATCH BY INCUMBENT Incumbents match lowest available fare in 10 LFA markets, but in most restrictive fare class (Q) only Incumbents maintain all advance purchase requirements and restrictions Fare Class Restrictions (A/P) Original Fares New Entrant Fares Limited Match Fares Y () $400 $400 B Sat (7) $163 M Sat, NR (14) $150 $127 $150 Q Sat, NR, CF (21) $100 $90 $

9 IMPACT OF ENTRY ON REVENUES LEG RM USED BY INCUMBENTS $1,250,000 Incumbent revenues decrease 5-8% with LFA entry and unrestricted fares, even with limited match Use of Threshold and Adaptive Threshold RM by LFA increases its revenues and further decreases incumbent revenues Airline 1 Absolute Revenues $140,000 Airline 3 Absolute Revenues $1,200, % -8.40% -8.45% $120,000 $100, % +54.7% $1,150,000 $80,000 $1,100,000 $60,000 $40,000 $1,050,000 $20,000 $1,000,000 Eb-Eb Eb-Eb-Eb Eb-Eb-Thres. Eb-Eb-Adapt. Thres. 70% $0 Eb-Eb-Eb Eb-Eb-Thres. Eb-Eb-Adapt. Thres. 90% 17 EFFECT OF ENTRY ON AIRLINE 1 S PASSENGER MIX Airline 1 loses more Y class passengers as Airline 3 moves from EMSRb to Fixed and Adaptive Threshold Airline 3 rejects more and more Q (with Fixed Threshold and Adaptive Threshold), and accommodates higher fare traffic Airline 1 Loads by Fare Class - LM No LFA EMSRb All EMSRb vs. Eb & Thres. EMSRb vs. Eb & Adapt. Thres Y B M Q 18 9

10 FARE CLASS MIX ON AIRLINE 3 With EMSRb, Airline 3 carries mostly Q traffic As Airline 3 switches to Fixed Threshold and Adaptive Threshold, its mix of passengers improves EMSRb Fixed Threshold Adapt. Threshold Airline 3 Loads by Fare Class Y B M Q 19 IMPACTS OF NETWORK RM ON INCUMBENT CARRIER REVENUES Use of Network RM by Airline 1 leads to revenue gains in all competitive situations Relative revenue gains vary between 1.35% and 1.43%, slightly lower than without low-fare competition, due to lower load factors Airline 1 Revenues $1,260,000 $1,240,000 $1,220,000 $1,200,000 $1,180,000 $1,160,000 $1,140,000 $1,120, % +1.35% Airline 1 - EMSRb Airline 1 - DAVN +1.40% +1.43% $1,100,000 vs. Eb vs. Eb-Eb vs. Eb-Thres. vs. Eb-Adapt. Thres. 70% 20 10

11 EFFECT OF ENTRY AND NETWORK RM ON AIRLINE 1 PASSENGERS 100 Local/Connecting Mix in Affected Markets - DAVN vs. Eb Locals Connecting 90-45% -37% -27% -47% -39% -30% % +7% +5% +12% +11% +10% Eb (No LFA) Eb & Eb Eb & FT Eb & Adat. Thres. Eb (No LFA) Eb & Eb Eb & FT Eb & Adat. Thres. Eb on Airline 1 vs. DAVN on Airline 1 vs. 21 LOCAL vs. CONNECTING TRAFFIC Airline 1 s carries fewer local market passengers with Network RM in LFA affected markets Network RM booking controls limit negative impacts on total loads by increasing connecting traffic more than Leg RM Connecting traffic on legs affected by LFA entry increase by 10-12%, up from 5-7% with Leg RM Network RM trades less attractive (lower fare) local market passengers for higher revenue connecting passengers Under LFA competition and lower local fares, Network RM protects seats for connecting passengers with higher total network revenue to airline 22 11

12 RESPONSE 2: FULL FARE STRUCTURE MATCH BY INCUMBENT Incumbents match all fares in these ten markets, including the removal of restrictions and advance purchase requirements Incumbents maintain fares, advance purchase requirements and restrictions in all other 472 markets Fare Class Original Restrictions (A/P) Original Fares New Entrant Fares Full Match Fares Y () $400 B Sat (7) $163 $163 M Sat, NR (14) $150 $127 $127 Q Sat, NR, CF (21) $100 $90 $90 23 AIRLINE 1 REVENUES FULL MATCH Airline 1 revenues also decrease under full match Revenues decrease slightly more than under the limited match case when the entrant uses EMSRb, but less in other cases Thousands $1,260 $1,240 $1,220 $1,200 $1,180 $1,160 $1,140 $1,120 $1,100 $1,080 $1,060 Airline 1 Absolute Revenues Full Match Limited Match Eb-Eb No LFA Eb All Eb-Eb-Thresh. Eb-Eb-Adapt. Thresh. 70% 24 12

13 AIRLINE 1 REVENUE GAINS FROM NETWORK RM Network RM gains increase under full match Gains over Leg RM increase to as much as 2.15% depending on the competitive RM situation $1,260,000 $1,240, % Airline 1 Revenues $1,220,000 $1,200,000 $1,180,000 $1,160,000 $1,140,000 Airline 1 - EMSRb Airline 1 - DAVN +1.54% +2.15% +1.93% $1,120,000 $1,100,000 vs. Eb vs. Eb-Eb vs. Eb-Thres. vs. Eb-Adapt. Thres. 70% 25 AIRLINE 3 REVENUES UNDER LIED AND FULL MATCH SCENARIOS Airline 3 revenues decrease 25% to 46% when the incumbent carriers match the LFA fare structure $140,000 $120,000 Limited Match Full Match $100,000 $80,000 $60,000-25% -44% -46% $40,000 $20,000 $- Eb All Threshold vs. Eb Adapt. Thresh. 70% vs. Eb 26 13

14 SUMMARY: RM IN LOW-FARE COMPETITIVE MARKETS LFA entry leads to revenue losses for incumbents Matching the LFA fare structure reduces Airline 1 revenue losses by limiting the loss of local market passengers Much evidence that % revenue gains of effective RM are greater in less restricted fare structures Segmentation power of fare restrictions is replaced by inventory control power of RM limits RM is also critical to LFA revenues when they compete against network carriers that use RM, especially Network RM Network RM gains increase for Airline 1, trading off local versus connecting passengers Improved ability to provide greater seat availability to connecting passengers with higher revenue contribution to total network 27 14