Mba Finance

Published on May 2016 | Categories: Book Excerpts | Downloads: 37 | Comments: 0 | Views: 425
of 62
Download PDF   Embed   Report

Comments

Content

1

Revenue Management System In Airline Industry

Submitted to

xxxxxxx

Submitted by xxxxxxxxx

A Dissertation report submitted in partial fulfillment of requirements for MBA (Aviation Management) April 2 !

2

"#apter$% Introdu&tion
Revenue Management was invented by U.S. airlines in the 1980’s in response to a newly deregulated industry and to the increased competition that was created. Since then it has been adopted by a variety o! industries and the list is constantly growing. "ut the basic concepts have been around !or #uite a long time $t was developed by the airlines to improve revenue per!ormance in the !ace o! increasing competition. $t was obvious to the airlines that passengers could be divided into two broad categories based on their travel behavior and their sensitivity to prices. %here business and leisure travelers. "usiness passengers tended to ma&e their travel arrangements close to their departure date and stay at their destination !or only a short time. %here was usually little !le'ibility in their plans and they were willing to pay higher prices !or tic&ets. (eisure travelers on the other hand boo&ed their !lights well in advance o! their travel date. %hey stayed longer at their destinations and had much more !le'ibility in their plans. %hey would o!ten decide not to travel rather than pay high !ares and !lights o!ten departed with empty seats. %he challenge to the innovators o! revenue management was to devise a plan that would ma&e the empty seats at the lower !are while preventing passengers who were willing to pay the higher !are !rom buying low)!are seats. Since low)!are passengers typically boo& in be!ore higher !are customers the revenue management system must !orecast how many business passengers will want to boo& on a !light. %hen it must set aside or protect these seats so that they will be available when the business passengers re#uest them.

3

'b(e&tive of Dissertation
%his research pro*ect e'amines the concept o! Revenue Management in the airline industry. %he ob*ective o! this dissertation is to understand the current and !uture trends in revenue management. +lso e'ploring new revenue management challenges and strategies

Revenue Management (RM) Ba&)ground
$n the early 19,0s some airlines began o!!ering restricted discount !are products that mi'ed discount and higher !are passengers in the same aircra!t compartments. -or e'ample ".+/ 0now "ritish +irways1 o!!ered earlybird boo&ings that charged lower !ares to passengers who boo&ed at least twenty) one days in advance o! !light departure. %his innovation o!!ered the airline the potential o! gaining revenue !rom seats that would otherwise !ly empty2 however it presented them with the problem o! determining the number o! seats that should be protected !or late boo&ing !ull !are passengers. $! too !ew seats were protected the airline would spill !ull !are passengers2 i! too many were protected !lights would depart with empty seats. 3o simple rule li&e protecting a !i'ed percentage o! capacity could be applied across all !lights because passenger boo&ing behavior varied widely with relative !ares itineraries season day o! wee& time o! day and other !actors. $t was evident that e!!ective control o! discount seats would re#uire detailed trac&ing o! boo&ing histories e'pansion o! in!ormation system capabilities and care!ul research and development o! seat inventory control rules. ($%%(45..6 019,71 o! ".+/ proposed that discount !are boo&ings should be accepted as long as their revenue value e'ceeded the e'pected revenue o! !uture !ull !are boo&ings. %his simple two !are seat inventory control rule 0hence!orth (ittlewood’s rule1 mar&ed the beginning o! what came to be called yield management and later revenue management. $n 3orth +merica the beginning o! intensive development o! revenue management techni#ues dates !rom the launch o! +merican +irlines’ Super Saver !ares in +pril o! 19,, shortly be!ore the deregulation o! U.S. domestic and international airlines. .ver the last twenty years development o! revenue management systems has progressed !rom simple single

4

leg control through segment control and !inally to origin8 destination control. 4ach o! these advances has re#uired investment in more sophisticated in!ormation systems but the return on these investments has been e'cellent. $n 1999 most o! the world’s ma*or air carriers and many smaller airlines have some level o! revenue management capability. .ther small airlines and international airlines in newly deregulated mar&ets are beginning the development process. %he success o! airline revenue management was widely reported and this stimulated development o! revenue management systems !or other transportation sectors and in other areas o! the service sector.

Meaning
Revenue Management is the process o! analysing and !orecasting consumer demand in order to optimi9e inventory and pricing levels to ma'imi9e pro!itability.

-ig 1 :rocess o! RM $n other words to constantly analyse and !orecast the remaining demand !or a certain !uture product or event and subse#uently ad*usting the pricing levels in order to sell the rig#t produ&t at the rig#t pri&e to the rig#t &ustomer at the rig#t time to ma'imi9e pro!itability. $t is important to note that Revenue Management addresses the revenue side o! the e#uitation not the cost side. %here e'ists an inverse proportionality between (oad -actor 0Units Sold1 and +verage ;ield 0Unit Revenue1 and Revenue Management will thus !ind the optimum balance between these !actors in order to ma'imise Revenue or $ncome.

5

:ro!it < .perating Revenue 8 .perating e'pense < R:= > ;ield ) +S= > Unit cost or < +S= ?Unit revenue 8 Unit /ost@ or < +S= ?+verage load !actor > ;ield 8 Unit /ost@ Unit Cost Units Sold/ Load Factor

Average Yield/

Revenue Manageme nt

Unit -ig.7 Relating terms o! RM Revenue *#e follo+ing examples explain RM,    %he ability to optimally match supply to demand at the best price points %he ability to proactively a!!ect this price point. Aetting the most revenue possible !rom your assets ) particularly i! they are perishable. $! it is not sold today you loose the ability to attain that opportunity revenue.  %he e!!ort to gather all possible data to try and reach this mar&et &nowledge so you can be proactive with your product or purchase and not be reactive to the mar&et. $t is your product or your money why be reactive rather than proactive with the mar&etB  %he ability to use this in!ormation to target di!!erent mar&et segments through their uni#ue discreet distribution channels and there!ore divide the mar&et into segments you can target and reach at the right time the right price and the right product without diluting your main mar&et base.  %he ability to !ul!ill the idea in the C#uotesC section !rom C%he +rt o! 5arC



+ proven way to increase revenues !rom your e'isting mar&et share without battling to ta&e mar&et share !rom competitors. %his can be !ar more cost e!!ective than !ighting to increase revenues and pro!its !rom gaining mar&et share !rom your competitors.



+ multimillion dollar a year industry embraced by the largest travel energy and broadcasting companies. +merican +irlines has stated that Revenue Management increases its revenues by about DE per year which many years has been the di!!erence in pro!its or losses. %he 5all Street Fournal has called RM the most power!ul new business tool o! the ne't century.

-roblem and "#allenges
%he ob*ective in revenue management is to ma'imi9e pro!its2 however airline short)term costs are largely !i'ed and variable costs per passenger are small2 thus in most situations it is su!!icient to see& boo&ing policies that ma'imi9e revenues. +lso although there is lower ris& in accepting a current boo&ing re#uest than in waiting !or later possible boo&ings boo&ing decisions are repeated millions o! times per year2 there!ore a ris&) neutral approach is *usti!ied. /onsider the arrival o! a boo&ing re#uest that re#uires seats in an itineraryGone or more !lights departing and arriving at speci!ied times within a speci!ic booking class at a given !are. %he !undamental revenue management decision is whether or not to accept or re*ect this boo&ing. + large computer reservations systems must handle !ive thousand such transactions per second at pea& times thus the decision must be reached within milliseconds o! the re#uest’s arrival. 3ot surprisingly no current revenue management system attempts !ull assessment o! each boo&ing re#uest in real time. $nstead precomputed aggregate control limits are set that will close the system !or !urther boo&ings o! speci!ic types while leaving it open !or others. %he reservations system can #uic&ly determine the open or closed status o! a boo&ing category and report bac& to an agent or customer without actually evaluating the re#uest. %he accept8re*ect decision can be restated as a #uestion o! valuationH 5hat is the e'pected displacement cost o! closing the incremental seats in the re#uested itineraryB %o ma'imi9e e'pected revenues the re#uest should be satis!ied only i! the !are value o! the re#uested itinerary e#uals or e'ceeds the e'pected displacement cost. %he apparent simplicity o! this

!

valuation problem is deceptiveGa complete assessment must allow !or all possible !uture reali9ations o! the reservations process that could be in!luenced by the availability o! any o! the seats on any o! the legs in the boo&ing. -ully traced this in!luence propagates across the entire airline networ& because a boo&ing can displace potential boo&ings that will have subse#uent impacts o! their own. %his in!luence also propagates !orward in time because many a!!ected itineraries will terminate later than the boo&ing being considered. +lso a boo&ing will normally have a return component at a later date with its own set o! concurrent and downstream e!!ects. Many other !actors increase the comple'ity o! the evaluation process. %able $ lists some o! them. +s can be seen in %able $ the practical comple'ities o! revenue management are dauntingGwe do not have space here to discuss all o! them. +s is always true modeling theoretical analyses and implementation rely on assuming away many o! these complicating !actors and appro'imating others. $t is important to remember that such appro'imations have yielded enormous revenue bene!its !or airlines and other enterprises. %he per!ormance o! a given revenue management system depends in large part on the !re#uency and accuracy o! updates to control limits and the number o! distinct boo&ing classes that can be controlled. %able 1 Demand Revenue .a&tors are values !ncertainty of fare value re"uent flyer redemptions Company vouchers Cancellation penalties or restrictions /ariable "ost .a&tors #arginal costs per passenger Denied boarding penalties $oodwill costs .are -rodu&ts %umber of products or travel agent special

"ustomer .ore&asting

Be#avior

and

Demand volatility Seasonality, day-of-week variation Special events Sensitivity to pricing actions Demand dependencies between booking classes Return itineraries Batch bookings Cancellations Censorship of historical demand data Defections from delayed flights

"

Diversions $o-Shows $roup bookings &nterspersed arrivals %o-shows Recapture !pgrades "ontrol System Booking lead time 'often ()) days or more* %umber of controllable booking classes +eg-based, segment-based, or full ,D control Distinct buckets, parallel nesting, or full nesting Reservations systems connectivity re"uency of control updates ,verbooking

ences 'restrictions* -roblem S&ale +arge airline or airline alliance- e.g., !nited.+ufthansa.S/S ,R&,% System0 1,))) flights and (2),))) passenger itineraries.day 3see $/R456 '7889*, B,6D '788:*; -roblem Interfa&es #arket strategy Code-sharing alliances Routing $ate ac"uisition and schedule planning leet assignment

SourceH Report on IRevenue ManagementH Research .verview and :rospects’ by F4--R4; $. M/A$(( and A+RR4%% F. J+3 R;K$3

Revenue Management 'b(e&tives
 :roduce a demand !orecast based on mar&et data commercial ob*ectives and calendar)related events  +d*ust demand !orecasts based on environmental changes guidelines and per!ormance metrics  Set up and manage an allocation strategy using the revenue management system tools in a systematic and e!!icient manner  :rioriti9e administrative tas&s management

#

"#apter$2 Resear&# Met#odology
%his research is +nalytical with #ualitative approach. $n +nalytical research the researcher has to use !acts or in!ormation already available and analy9e these to ma&e a critical evaluation o! the material. So here in this pro*ect $ have collected data mainly !rom the secondary sources. $n primary source $ have tal&ed to one Revenue Management e'pert Mr. Aary :ar&er :resident o! Revenue Management %raining group /anada. "ased on the in!ormation !rom Mr. Aary and various secondary sources $ go through the revenue management scenario worldwide and able to analy9e its current and !uturistic trends. +lso !ind new Revenue Management Strategies and challenges and to some e'tent try to predict !uture o! the Revenue Management based on the industry e'perts views.

0iterature Revie+
Davis M %!!12 Iyield management techni#ues are reportedly #uite valuable. .n an estimate +merican airline made an e'tra LD00 million per year based on its yield management techni#ues’. +ccording to an estimate the pricing system used by +merican airlines change the prices more then hal! a million per day. Dene&)ere and M&Afee %!!32 IRevenue management techni#ues provide tools to use consumer surplus through dynamic pricing’. 3evertheless there is little doubt that dynamic price discrimination is economically important. %he pricing system by most ma*or airlines is opa#ue to the customer. %he only thigh which customer came to &now about the airline is the #uality o! service. "y applying the dynamic pricing techni#ues and by providing best services to the customer airlines can increase the revenue and also the loyalty o! the customer to the brand. 4erag#ty2 5evin 2 12 IRevenue management o!!ered us a way to capture revenues that

were being le!t on the table. Revenue management implements the basic principle o! supply and demand economics in a tactical way to generate incremental revenues.’ +irlines monitor through the use o! speciali9ed so!tware how seats are being reserved and

1$

react accordingly !or e'ample by o!!ering discounts when it appears as i! seats will otherwise be vacant. 6uring pea& season airline charge the passenger as much as they can and o!!er products to those who are willing to pay more while in lean season various mar&eting techni#ues li&e !ree companion scheme discounted one way !ares returned !ares e'cursion !ares and bas&et scheme etc. are being o!!ered by airline according to their mar&eting model. Andy Boyd22 6 I.ne)way airlines segregate customers is by imposing advance

purchase re#uirements and Saturday night stays !or cheap tic&ets. %hese restrictions act as MMgatesNN that separate price)sensitive leisure travelers !rom time)sensitive business travelers. %he challenge !or other industries is to !ind the right gates’. %he sensitivity o! demand can be controlled to some e'tent i! an airline has a good no o! loyal customers and an ability to attract new customers even at the time o! low demand. Sengupta A22 3 I+irlines that use revenue management periodically review

transactions !or services already supplied and !or services that are to be supplied in !uture. %hey may review in!ormation such as past statistics up coming events li&e sports holidays !estival or une'pected past events such as terrorist attac&s and other in!ormation SU/O as competitive in!ormation 0including prices1 seasonal patterns and other pertinent !actors that a!!ect sales.’ %he success o! an airline depends upon the capability o! the air line to !orecast the !uture and deploy the ma'imum capacity on the routes with comparatively high demand but by ta&ing the cost o! operation into consideration. Belobaba 2 62 I+irline who is busy in maintaining its load !actors by decreasing its !are

with analy9ing that it can harm its !uture policies. %hese airlines are more susceptible to go out o! the picture.P Revenue management system provides enables airlines to satis!y customer and made pro!its to the greater possible e'tent without any lose o! image.

11

"#apter$6 5ey Areas of Revenue Management
%) .ore&asting 2) -ri&ing 6) Seat Inventory "ontrol 1) 'verboo)ing

%) .ore&asting
-orecasting is an important component o! planning in any enterprise2 but it is particularly critical in airline revenue management because o! the direct in!luence !orecasts have on the boo&ing limits that determine airline pro!its. -orecasting !or airline revenue management system is inherently di!!icult. /ompetitive actions seasonal !actors the economic environment and the constant !are changes are a !ew o! the hurdles that must be overcome. $n addition the !act that most o! the historical data is censored !urther complicates the problem. %he number o! seats an airline can sell on a !light is determined by the boo&ing limits set by the revenue management system. +n airline continuous to accept reservation in a !are class until the boo&ing limit is reached. +t that point the airline stops selling seats in that !are class. $t also stops collecting valuable data. 6emand !or travel in that !are class may e'ceed the boo&ing limits but the data does not re!lect this. So the data is censored or constrained at the boo&ing limit. 5hile some models e'ist that produce unbiased !orecasts !rom censored data it is pre!erable to unconstrain the censored observations so that they represent true demand. %hen the !orecasting model may be chosen based on the structure o! the problem rather than the nature o! the data.

2) -ri&ing

12

:ricing has been around as long as people have traded. :ricing o! product or services represents the &nowledge and understanding o! their product and ultimately determines the growth prospectus o! the organi9ation. $! product and Qor service is not e'clusive or not providing the desired satis!action to the customer or unable to give value !or money will not succeed. .n the other hand i! product andQor service has high level o! customer satis!action and high manu!acturing Qoperating costs it will lead to increase in the time period o! reaching its brea& even point and loss o! revenue. %hus price is to be determined in such a way that !irm can reap the pro!it o! increase in demand and let customers to ta&e the advantage o! availability o! product andQor service. Stati& and Dynami& -ri&ingH :ricing is ma*or area o! airline revenue management. :ricing is generally categori9ed into to parts static and dynamic. $n the static pricing the price is set at being o! the boo&ing period. $n the dynamic pricing the price changes through out the boo&ing period. +irline pricing in $ndia is almost opa#ue and the end users o! airline services are not aware o! any o! these comple'ities o! the prices determination. :rice and #uantity are determined by the interaction o! demand and supply in the mar&et. Oowever given the large number o! buyers !irms can decide prices at which they will sell tic&ets. $n !act in the airlines sector !irms go in !or third degree price discrimination and segment the mar&et charging a higher price to the mar&et with a relatively inelastic demand 0such as !ares between business and economy class travelers or between emergency travel and leisure travel by providing ape' !ares1. %he low cost airlines !ollow this di!!erent pricing strategy. /ustomers boo&ing early with carriers such as +ir 6eccan will normally !ind much lower prices i! they are prepared to commit themselves to a !light by boo&ing early on the *usti!ication that consumer’s demand !or a particular !light becomes more inelastic the nearer to the time o! the service. Differential pri&ing in airline industry,

13

+irlines generally adopt di!!erential pricing $n order to utili9e all the available seats in a particular time limit to generate ma'imi9e their revenue. + particular !light is divided into various classes o! service 0!irst business and economy class1. 6i!!erent !are products are o!!ered in economyQcoach cabin class with di!!erent restrictions at di!!erent at di!!erent prices. Jirtually every airline in the world o!!ers multiple price points 0even low !are carriers with simpli!ied !are structures1. 4conomic trade o!! in pricing decision o! an airline constitutes o! stimulation o! new demand and diversion o! e'isting demand to lower !ares. Recent pricing di!!iculties o! networ& airlines due in part to greater diversion o! revenue than stimulation o! demand because o! increasing no o! competitors. 7ffe&tive pri&ing of airline inventory, %raditional ways o! determining tic&et price are not in pace with airline competitive pricing practices. Still it is very di!!icult to estimate e'actly the price elasticity willingness to pay potential !or stimulation and diversion. More over there is no practical tool !or airline to determine RoptimalP prices. Some air lines are now implementing R:ricing decision support systemP which wor& on the basis o! historical demand and prices. $t is primarily a monitoring o! price change. %he most dominant practice is to match low !are to !ill planes and retain mar&et share.

:rice elasticity

5illingness to pay

:otential !or stimulation and diversion

:ricing o! :roduct

:rice elasticity

-ig S Re#uirements !or e!!ective pricing in +irline

14

6) Seat Inventory &ontrol
$nventory control system basically means the procedure by which the product or the !inished commodity is being controlled in the mar&et. %he system depends upon the commodity that is being o!!ered by the producer i! the commodity is perishable the inventory system is di!!erent !rom that o! !or durable goods. "ut when we tal& about service industry we don’t thin& o! a commodity rather we thin& o! a service !or e.g. li&e in +viation industry. %hey don’t o!!er a tangible commodity rather they o!!er a product which can be e'perience but canNt be touched or !elt. +s we tal& about -M/A mar&et we get a tangible commodity in hand but when we compare it with +viation sector we get an intangible commodity in terms o! e'perience. $n the /ommodity industry inventory system has di!!erent procedures li&e !irst)in)!irst) out and last)in)!irst)out. %he mar&et demand !ashion tastes and pre!erences o! customers durability o! the commodity etc a!!ect this but in +viation industry things are bit di!!erent !rom the commodity mar&et. $t has number o! seats as the product in a passenger airline and space to carry load in cargo airline. $n this scenario the rate or the price o! the commodity is the only thing that regulates the inventory. $n the aviation industry the companies ta&e care o! the inventory in two di!!erent !orms nested and non)nested. %he low cost carrier is incorporating non)nested !orm while the !ull service carriers are ta&ing the nested approach. 3on)nested approach says that there would no di!!erence between the !ares o! the seats as they are all the same while the nested system o! airline inventory control system says that even i! the seats are o! the same class still they have a scope to earn more revenue as compared to other seats in the class. $n the non)nested system the prices o! the tic&ets increases as the time o! the !lights decreases while in the nested !orm o! inventory management the seats in !orm groups when a group is sold out then the tic&ets are sold in the second group. %he system states

15

i! the seats are sold in say R=P class that has a lower price band then the no. o! seats in the mother class i.e. say R;P class will decrease. $t simply means that i! a seat is sold in the sub)class so it will a!!ect the mother class i.e. number o! seats will decrease. 3owadays the low cost carriers consider the non)nested approach as they don’t have any class di!!erentiation so it would be easier to maintain the inventory system as compared to the nested approach that is being considered by !ull service carrier as they have the di!!erent class considerations !or e.g. !irst class business class and economy class with spot !ares. %he /omputer Reservation System 0/RS1 and Alobal 6istribution System 0A6S1 are managing the nested and non)nested inventory control system through di!!erent service providers in $ndia 0li&e ;atra.com Ma&emytrip.com etc1 and all over the world 0li&e Aalileo +madeus etc1. $n the low cost carrier reservation system seats are boo&ed on a general basis irrespective o! the classes as there are no classes while in the !ull cost carriers the seats are boo&ed considering the classes with !ares. Since there is only one class in the model so we ta&e it as non)nested approach as it doesn’t ma&e any di!!erence but in case o! nested approach since the classes are di!!erent. %he -//’s seat boo&ing system is such that i! a seat is boo&ed in the mother class it wonNt e!!ect the number o! seats in sub)classes but i! the seat is boo&ed in any o! the sub) classes then it will e!!ect the number o! seats in mother class. +s !ar as (//’s is concerned these seats boo&ed never a!!ect any mother class or sub)class as they don’t have any class in di!!erentiation.

Seat Inventory &ontrol stru&tures
*#e 8ested Approa&# $n order to aid the airlineNs seat inventory control and overboo&ing policy airline o!!icials employ the revenueQyield management concept. $n a nested structure re#uests !or higher !are classes can be accommodated i! seats are available in lower !are classes.

1

%he availability !or each discount class is controlled through nesting. 3esting ma&es subsets o! the seats available to various levels o! discount !ares. Smaller subsets are available to lower)valued discount classes than to !ull !are or moderate)discount classes. $! the !are classes are controlled independently it would be possible to sell a low)revenue reservation and simultaneously turn away a high)revenue passenger. 0%his situation would happen when all the seats allocated to the high)revenue class are sold while some seats were still available to a lower)revenue class.1 3esting simpli!ies the maintenance o! discount !ares by automatically ensuring that a low)value seat is never available when a higher)valued !are is closed to additional sales. %he use o! nesting means that !orecasting too much demand in a higher)revenue class causes poor discount allocation only !or lower)revenue classes. $! deep)discount seats are available !or sale so are the moderate and !ull)!are seats. $! no moderate discount seats are available then neither is deep) discount seats. 5hen !ull)!are seats are not available no seats are available 0the !light has reached its overboo&ing level1. %he nested structure can be !urther divided into twoH the static nested structure and the dynamic nested structure. $n the static nested structure boo&ing limits are set at the beginning o! the boo&ing period. $n the dynamic nested structure boo&ing limits are updated during the boo&ing period depending on the actual boo&ing status.

1!

-ig.T %raditional leg and segment inventory control structures /onsider each o! these scenariosH  5ith non)nested controls the boo&ing classes are independent o! each other. %his implies that situations could arise when a boo&ing class with a lower !are value 0eg J1 is open !or sale while a higher !are class 0eg ;1 is closed !or sale. -or this reason non)nested controls are rarely used in practice. $t can be practical only i! demand !or each boo&ing class can be !orecast with absolute certainty which is impossible to do.  :arallel nesting controls are an improvement over non)nested controls because all the lower)valued boo&ing classes are nested into the higher class ; guaranteeing that ; has always the same or higher availability than the lower classes. +s the !are classes below ; in the hierarchy are independent o! each other however it is li&ely that a lower)valued class 0eg U1 is open !or sale while a higher)valued !are class 0eg "1 is closed !or sale causing revenue dilution.  Serial nesting controls by virtue o! their serial nested hierarchy based on !are class value guarantees that a lower)valued class will never be open when a

1"

higher)valued class is closed !or sale. $t is !re#uently used in con*unction with segment limits or segment close indicators.  Mi'ed nesting controls are !re#uently used by airlines who wish to control consolidator cruise line or promotional tra!!ic with a guaranteed allocation. $n this scenario this low)yielding tra!!ic is parallel nested into ; class.  Oybrid nesting is a variant where there are two or more independent nesting structures in a cabin which are &ept separate. $n the e'ample both ; and R !are classes are at the same level in the hierarchy !or a cabin. %his is sometimes used to manage demand !rom di!!erent channels separately. $t su!!ers !rom the problem that when a !are hierarchy is sold out seats cannot be borrowed !rom the second !are class hierarchy without manual intervention.

*ypes
a) Single$0eg Seat Inventory "ontrol ('ptimi9ation *e&#niques) Utili9ing demand !orecasts !or individual !light legs !are class yield management 0-/;M1 systems use optimi9ers which determine seat allocation !or the set o! !are classes on each leg within a networ&. %he most commonly used !are class mi' allocation is the idea o! serial RnestingP o! !are classes 8 a problem !irst solved by (ittlewood at ".+/ !or the case o! two !are classes. +s opposed to allocating seats in partitioned !are classes nesting instead protects seats in high !are classes by limiting the number o! seats sold in lower !are classes based on a !orecast o! demand !or each class as well as the e'pected seat revenue. "elobaba e'tended the nested seat allocation problem !rom (ittlewood’s two classes to multiple !are classes with the 4'pected Marginal Seat Revenue 04MSR1 heuristic in his :h.6. thesisT0. %his algorithm employs leg)based demand !orecasts by !are class to produce leg)based seat protection levels !or nested boo&ing classes. 4MSR determines

1#

boo&ing limits based upon the e'pected marginal revenue 8 the probability o! selling an additional seat in a given !are class multiplied by the revenue gained !rom selling that seat. +s the number o! seats protected in a particular !are class increases the probability o! selling that ne't seat decreases2 thus the boo&ing limit !or a !are class is determined when the 4MSR is e#ual to the !are o! the ne't lower class. "elobaba’s updated 4MSRb algorithm protects *oint upper classes !rom the ne't !are class *ust below and has become somewhat o! an industry standard !or establishing boo&ing limits on a !light leg basis. $n order to simpli!y calculations !or *oint classes the !are class demands are assumed normal and independent8 assumptions which are violated when the !are class restrictions are eased. %o calculate the optimal protection levelsH Define -i(Si ) : probability t#at ;i < Si2 where Si is the number o! seats made available to class i >i is the random demand !or class i /oncept o! protecting seats can be demonstrated !or 7 !are classes using a simple marginal revenue analysis approachH :rotect another seat !or !ull !are demand i! :r Vspill o! !ull !are demandW X R7QR1 where R7 < discount !are R1 < !ull !are and spill means we !ill up the plane with too many discount passengers and thus must re*ect 1Y !ull)!are pa'. %he e'pected marginal revenue o! ma&ing the St# seat available to class i isH 7MSRi(Si ) : Ri = -i(Si ) where Ri is the average revenue 0or !are1 !rom class $ %he optimal protection level π1 !or class 1 !rom class 7 satis!iesH

2$

7MSR%(π %) : R% = -%(π % ) : R2 .nce π1 is !ound set "(7 < /apacity ) π1 /onsider the !ollowing !light leg e'ampleH %able 7 -are /lass ; M " U +vg. 6emand 10 1D 70 S0 Std. 6ev. S D , 10 -are 10000 ,000 D000 SD00

%o !ind the protection !or the ; !are class we want to !ind the largest value o! π; !or which 4MSR;0π; 1 < R; Z :;0π; 1 X R" 4MSR;0π; 1 < 1000 Z :;0π; 1 X ,00 :;0π; 1 X 0.,0 where :; 0π; 1 < probability that >; X π; $! we assume demand in ; class is normally distributed with mean standard deviation given earlier then we can create a standardi9ed normal random variable as 0>; ) 101QS. 3e't we use 4'cel or go to the Standard 3ormal /umulative :robability %able !or di!!erent RguessesP !or π;. -or e'ample :rob ? 0>; )101QS X 0> ) 101QS @ < 0.8T1 :rob ? 0>; )101QS X 0? ) 101QS @ < 0.,T, :rob ? 0>; )101QS X 0! ) 101QS @ < 0.[S !or π; < , !or π; < 8 !or π; < 9

21

So we can see that π; < 8 is the largest integer value o! π; that gives a probability X 0., and there!ore we will protect 8 seats !or ; class Oow many seats to protect *ointly !or classes 1 and 7 !rom class SB %he !ollowing calculations are necessaryH

< 1 7 = <1 + < 7
7 \1 7 = σ \17 +σ \7 σ

R1 7 =

R1 Z < 1 + R7 Z < 7 <1 7

= 1 7 0 S 1 = :r ob 0 < 1 + < 7 > S 1

%o !ind the protection !or the ; and " !are classes !rom M we want to !ind the largest value o! π;" that ma&es 4MSR;"0π;" 1 <R;" Z :;"0π;" 1 X RM $ntermediate /alculationsH R;" < 010Z1000 Y 1D Z,001Q 010Y1D1 < 870
<6 \6 σ = < 6 + < B = 10 +1D = 7D
7 7 \6 \B = σ +σ = S7 + D7 = ST = D.8S

B B

%he protection level !or ;Y" classes satis!iesH 870 Z :;"0π;" 1 X D00 :;"0π;" 1 X .[098 +gain we can ma&e di!!erent RguessesP !or π;".

22

:rob ? 0>;" )7D1QD.8S X 02 ) 7D1QD.8S @ < 0.80D :rob ? 0>;" )7D1QD.8S X 022 ) 7D1QD.8S @ < 0.[9, :rob ? 0>;" )7D1QD.8S X 026 ) 7D1QD.8S @ < 0.[SS :rob ? 0>;" )7D1QD.8S X 021 ) 7D1QD.8S @ < 0.D[,D

!or π;" < 70 !or π;" < 77 !or π;" < 7S !or π;" < 7T

So we can see that π;" < 7S is the largest integer value o! π;" that gives a probability X 0.[098 and there!ore we will *ointly protect 7S seats !or ; and " class !rom class M] Suppose we had an aircra!t with authori9ed boo&ing capacity 80 seats our "oo&ing (imits would beH "(; < 80 "(" < 80 ) 8 < ,7 "(M < 80 ) 7S < D, Basi& modeling assumptions for serially nested &lasses, a1 demand !or each class is separate and independent o! demand in other classes. b1 demand !or each class is stochastic and can be represented by a probability distribution c1 lowest class boo&s !irst in its entirety !ollowed by the ne't lowest class etc. d1 boo&ing limits are only determined once 0i.e. static optimi9ation model1 b) 'rigin@Destination "ontrol .^6 controls with virtual nesting Jirtual nesting was !irst developed 0Smith 198[1 to control inventory by .^6. Jinod ". 0199D1 I.rigin and 6estination ;ield Management’ Massachusetts institute o! technology described that the use o! Rvirtual buc&etsP to compare networ& value o! local and connecting !are classes is one approach to networ& .6 control. 6+J3 couples .6 !orecasting with leg)based seat inventory control and uses a deterministic linear program

23

0(:1 with an ob*ective o! networ& revenue ma'imi9ation to calculate a Rpseudo !areP !or each !are class in the networ&2 this pseudo !are corrects the regular !are !or networ& displacement e!!ects. "y grouping each leg’s pseudo !ares into similar sets or buc&ets and then optimi9ing boo&ing limits !or those buc&ets the airline has a mechanism !or ma'imi9ing revenue while accounting !or displacement costs over its networ&. %he ob*ective o! a nested inventory control structure is to ensure that a lower)valued service 0.^6 one)way itinerary1 class is not open !or sale when a higher)valued service class is closed !or sale. %his statement is not completely true since service classes that are mapped to the same virtual buc&et have the same availability even though there may be !are di!!erences between them.

24

.^6 controls with &ontinuous nesting + growing number o! airlines are using continuous nesting controls to manage seat inventory. /ontinuous nesting controls 0Jinod 199D1 are called bid price controls. %he bid price can be de!ined as the opportunity cost o! not having an incremental seat on a !light leg in the networ&. +lternatively the bid price can be de!ined as the incremental total revenue that can be derived i! one e'cess unit o! capacity is available. %he e'cess revenue would be obtained by accepting a di!!erent mi' o! passengers by service 0origin destination routing1 and class in the networ&. %he bid price is also !re#uently interpreted as the minimum acceptable !are on a !light leg. "id price controls with a bid price and gradient provide the !le'ibility o! detailed control by service class without introducing a signi!icant overhead to reservations processing. %he total bid price !or a service is simply the sum o! the individual bid prices o! the !light legs that ma&e up the service. $! the total bid price is less than the !are the re#uest is accepted. .therwise it is re*ected. %here!ore !inancial availability is determined by the net contribution given by

-aresc is the !are !or service s and class c _ * is the bid price !or !light leg *2 s ϵ S where S is the collection o! services in the networ&. $! the net contribution is positive the reservation re#uest is accepted. .therwise it is re*ected. $n addition the incremental bid price or gradient is used to ad*ust the bid price when boo&ings and cancellations occur. %he gradient can be interpreted as the change in bid price !or a unit change in available seats. 5ith each new boo&ing the bid price is increased by the bid price gradient. 5ith each cancellation the bid price is decreased by the bid price gradient. %here!ore i! a re#uest is accepted the bid price is updated by adding the gradient as !ollows

25

where _ * is the current bid price o! leg * _Y * is the new bid price on leg * `* is the gradient on leg * and Seats"oo&ed* represents the number in party !or the boo&ing re#uest. Similarly i! a cancellation occurs the bid price is subtracted by the gradient as !ollows

3ote that this assumes that the relationship o! the gradient to the change in capacity is linear. %his however is an appro'imation that can only be applied !or small changes in available seats. + better alternative is to generate a bid price vector !rom the networ& optimisation model based on the assumption that primal !easibility e'ists as boo&ings are accepted. "y storing the bid price curve on the !light inventory record the linear assumption can be rela'ed. +n argument can be made that !re#uent reoptimisation o! the networ& obviates the problem and the bid priceQgradient approach is comparable in per!ormance to the bid price curve. %here will always be latency however between networ& reoptimisations !rom revenue management which ma&es the bid price curve more appealing. /ontinuous nesting provides a level o! granularity beyond virtual nesting controls. 4very service class availability re#uest is considered uni#ue unli&e virtual nesting where availability was the same !or all service classes mapped into a virtual buc&et.

1) 'verboo)ing
%he travel industries in particular are plagued by the problem o! Rno)showsP 8 people who boo& inventory and then do not show up to use it 0or pay !or it1. %he attachments o! cancellation penalties to airline discount !ares are attempts to mitigate this problem which have met with some success. %o compensate !or no)shows travel !irms Roverboo&P their capacity trading o!! the possibility o! empty units i! they don’t

2

overboo& enough against the ill will and out)o!)poc&et compensation to customers that occurs when customers are RbumpedP 0airlines1. %his tradeo!! should be considered in the !ollowing manner. -irst the probabilities o! incurring various 3o)Show rates must be !orecasted in much the same manner that demand is !orecasted. $n the %ravel industry no)show rates o!ten vary by rate classQmar&et segment and time period. %he RcostP o! !ailing to honor a customer’s boo&ing including both out o! poc&et costs such as cash compensation to RbumpedP airline passengers and a consideration o! the potential loss o! !uture revenue !rom the disgruntled customers must also be calculated. 0+irlines attempt to minimi9e customer ill will by compensating passengers who voluntarily relin#uish their reservations with !ree tic&ets.1 5ith this in!ormation the e'pected oversale cost 0probable number o! oversales times the total cost per oversale1 can be calculated !or any level o! overboo&ing above the actual number o! inventory units available !or sale. %he RcorrectP level o! overboo&ing is where the e'pected cost o! an oversale !or the ne't unit to be sold is e#ual to the 4MR value !or the ne't unit to be sold. +s long as the 4MR value is higher than the marginal e'pected cost o! an oversale it will pay to allow another unit to be sold !or at least the 4MR value.

*raditional /s Anrestri&ted Revenue Management
%raditional revenue management assumes mar&et segmentation and !ares which have rules and restrictions. 4ven without RM there would be a mi' o! passengers owing to !are rules. %raditional RM per!ormance measures are typically revenue opportunity spoilage and dilution. .n o!!)pea& !lights all classes could be le!t open as the !ares rules ensured some mi' o! passengers. .n busier !lights RM controlled classes according to historical !are mi'. RM o! unrestricted !ares is a di!!erent issue. 5ith no rules and restrictions di!!erent classes are no longer individual entities. %he passenger will always buy the lowest !are in the cabin)there is no incentive to up sell as there is no distinction between !are types. .n lower load !actor !lights demand will always shi!t to the lower class)there!ore the emphasis is now on controlling all !lights at the detailed level whether traditionally pea&

2!

or o!!)pea&. %he dilution ris& on o!!)pea& !lights must be assessed against the potential !or volume growth on o!!)pea& !lights. %he role o! RM has been re)de!ined2 the #uestion becomes when to close the current available classQ!are)a!ter ' number o! boo&ings or y days be!ore departureB %he traditional RM system becomes a vehicle to gather data optimi9e overboo&ing levels and cabin split. User overrides are re#uired to ensure that classes are closed on the approach to departure to assist with up sell. .verrides are also re#uired to cap seat limits by class to protect against dilution !rom higher class. +chieving the balance o! how many seats and how long to &eep a class open is the &ey to a one)way RM approach. /alculating the number o! seats to sell and when to close classes should re!lect the level o! price elasticity e#uation however are such that it is di!!icult to model)passenger choice depends on many !actors including schedule competitive price !are etc. %he main issue now present is that o! boo&ing !rames !or business and leisure travelers. 5hen these time !rames start to overlap the #uestion o! which class will generate the highest revenue is crucial. -or e'ample business passengers usually boo& within the last two wee&s on short)haul routes. .n many o! these routes however leisure passengers are also loo&ing to buy within a similar time !rame. %he trade 8o!! between higher !areQlower volume vs lower !areQhigher volume must be considered. %o help ascertain the potential level o! business dilution i! lower classes are le!t open :3R chec&s are use!ul. %hese can be manual or via a :3R database. "y reviewing the volume o! business itineraries boo&ing in the lower leisure classes it is straight!orward to establish the li&ely level o! dilution on a !light. 0%he number o! business itineraries can be based on certain assumptions lin&ed to length o! stay source o! boo&ing advance boo&ing period etc1 Until trials are made to close classes however it is di!!icult to establish what will happen !ollowing a change in strategy.

2"

"#apter$1 0atest trends in RM
>Rising fuel prices are leading to opportunity to e?plore intelligent R# Strategies@ 5ith !uel prices s&yroc&eting and customers resisting higher !ares increasing revenue is a primary challenge !or all airlines. >%etwork ,ptimiAation in a #i?ed 5nvironment of Restricted and !nrestricted Structures@ %he development o! the optimi9ation methodology in response to the changed business environment with intensi!ied competition !rom (ow /ost /arriers with unrestricted !are structures. %he approach e'tends the .^6 networ& optimi9ation principles !or optimi9ing revenue in a restricted !are structure to an arbitrary mi' o! !are structures consisting o! both restricted !ares and unrestricted !ares on a path basis. > enceless /nalyst@ 5ith the introduction o! !enceless !are)structures traditional RM systems have to be adapted in order to prevent spiral down. Many airlines actually use manual overrides to achieve this. +s the optimi9ation and !orecasting capability o! the systems evolves to ta&e into account buy)down and even competitive scenarios the role o! the analysts will change again. Aame theory and simulations used to suggest that that role should not be reduced to ad*usting the !orecasts o! the new)style systems while leaving the optimi9ation to the computer. .n the contrary the responsibility o! the analysts must grow as they need to de!ine the strategy on how to deal with the competition. >Different =ricing and R# Strategies on #arkets with +ow are Competitors@ 5ith recent trends toward !are simpli!ication airlines have been struggling to adapt their RM models to cope with reduced demand segmentation and Cspiral downC caused by less restricted !ares. are-Structures, Revenue #anagement Systems, and the %ew Role of the are

2#

>Coordinating Revenue #anagement in /irline /lliance@ /ode sharing agreements in airline alliances allow alliance members to virtually e'tend the reach o! their networ&s. Such agreements however create a di!!icult coordination problemH each member ma&es revenue management decisions to ma'imi9e its own revenue and the resulting behavior may produce sub)optimal revenue !or the alliance as a whole. + variety o! mechanisms have been proposed to coordinate alliance sales and share revenues including !ree sale and so!t bloc& arrangements. %he #uestions to e'amine are 0i1 how revenue management decisions o! the alliance partners are in!luenced by the revenue)sharing arrangement 0ii1 the e!!ect on alliance)wide pro!its and 0iii1 how revenues are split among alliance partners. >Bow to Cake /dvantage of ,n-+ine Sales &nformation in the R# 5"uation@ 5ith the growth o! on)line sales 5eb +nalytics used in tandem with revenue management can provide a power!ul tool !or revenue optimi9ation. $t is now possible to trac& e'actly who uses website and precisely how they do it what products are selling 0and which arenNt1 activity trends types o! visitors where they come !rom number o! visitors per page the page they were on when they le!t your site the percentage o! customers who created a shopping cart and chec&ed out and many other !actors. $n short on)line sales provide vital in!ormation that can guide airlines in improving their product o!!ering their customers’ e'periences and ma'imi9ing revenue. >Supply Side Revenue #anagement@ %he ability to match demand 0or yield1 with capacity is central to the operational success o! an airline. Jariants o! the -leet +ssignment Model 0-+M1 have proven invaluable during the schedule development and planning processes at many airlines. (i&ewise much has been discussed about demand)driven dispatches. Oowever near)term !leeting ad*ustments are o!ten met with operational hurdles. >Revenue #anaging Consumers through Competitive =ricing@ >Che 5volution of Revenue #anagement- &ntegrating your CR# and Revenue #anagement Systems@

3$

"#aper$B Revenue Management Strategies
RM strategies have been changing with time due to changing business environment. Jarious !actors de!ining this new environment are discussed below. A ne+ business environment  /ontinuous growth o! the (ow /ost /arriers 0(//1 − .ne)way !are structures − −    4'tremely low bottom prices 3on)transparent pricing structure

-ull service networ& carriers reduce !are conditions /apacity is continuously e'panding ;ield and tra!!ic volume under pressure − Resulting in declining revenues

 

Rapidly changing environment $ncreasing costs − -uel − +irport !ees

Rea&ting to Mar)et "#anges

-ig.D Reacting to mar&et changes strategy chart

31

5hile reacting to mar&et changes using revenue management strategies it is important to cover concepts li&e demand and supply customer segmentation pricing boo&ing class assignment and inventory nesting. 5ith a good understanding o! mar&eting economic theory understanding customer behavior and mar&et dynamics is easier. Mar&et &nowledge is important when analy9ing and evaluating how demand will be impacted by changes in the mar&et place. %here can be many &inds o! changes both e'ternally and internally that can challenge a Revenue +nalyst to maintain optimal seat allocations. 5e call these changes triggers. %he RM system has many levers and allows the revenue analyst to set up initial seat allocations based on current strategies. 5hen a change in the mar&et triggers a re#uired ad*ustment the analyst needs to understand what needs to be done what system changes are re#uired and what are the best levers to modi!y. $n order to set up an initial allocation strategy and to be alerted to changes or trends a good reporting system is needed. $! you &now what to loo& !or and you have reports that give good indications o! per!ormance and also direct you to the appropriate levers to ma&e ad*ustments you will reali9e the bene!its o! the RM system !aster. Scheduled reports that loo& at =:$ on a daily or wee&ly basis allow the Revenue +nalyst to maintain the RM system and to optimi9e revenue potential. %hey also allow !or the necessary !eedbac& to management on the per!ormance o! their strategy. /oncepts and models such as this provide a re!erence base !or how wor& !lows and people interact and what are best demonstrated practices. %hey ensure a consistent introduction and training is delivered to new employees as well as those !ollowing recurrent training. %he entire model is documented !or re!erence at any time.

32

Mar)et "#ange *riggers  Demand Management :rice o :roactive pricing o Reactive pricing  Mar&et growth Qdecline 0share1 o Mar&et demand is growing or declining o /ompetitive schedule change  Aroup management o Aroup acceptance evaluation  4vent management o $denti!ying the event in the RM system o Managing events 0moving dates1   S&#edule Management  + 3ew !light departure o /reating a new !orecast o /opying history o Setting overboo&ing 4valuating schedule change impact on demand /apacity planning boo&ing class

%able S -actors o! Mar&et /hanges

33

/alue -ositioning .or Mar)et Segmentation
+ 3 0

P e r c e iv e dP r o d u c tQ u a lity

+ 2 0

Good Value Premium Brands

Value line

1 0 0

+ 1 0

Mid-Market

-1 0

Economy Brands -2 0 Poor Value

-3 0

-30

-20

-10

100

+10

+20

+30

Relative Price

-ig.[ :erceived product #uality and Relative price relationship :ricing has to be designed to !it into a mar&et positioning strategy. $n the above -ig. :rices that position a brand below the diagonal will be perceived as poor value by consumers because their price is too high !or the #uality they are seen as o!!ering. :ricing above the diagonal o!!ers outstanding value with customers perceiving high #uality brands at relatively low prices. $n a strategic approach to pricing Management !irst de!ines its desired mar&et position relative to its target customers and competing brands2 this positioning de!ines the acceptable price which in turn determines the acceptable level o! costs. %he product has then to be designed to meet cost constraints. 5hile mar&ets have traditionally been divided into three or !our segments the evidence is that the number is increasing. +t the limit every customer can be an individual segment !or which a separate product is designed and price set. %his is re!erred to as one)to)one mar&eting. Several !actors shape this trend. .ne is rising customer e'pectations !or individually tailored solutions. %hese e'pectations are !ed by greater a!!luence and more competition

34

among suppliers. %he second !actor is in!ormation technology which !acilitates one)to) one communication between buyers and sellers. -inally new production methods !rom !le'ible manu!acturing to new networ&ing relationships among supply chain participants increasingly !acilitate customi9ed solutions. "ompetitive .lig#t 'ptimisation

-ig., SourceH (u!thansa Systems :ro!it(ine ;ield Rembrandt. "ompetition Mat&# Rule$ Based "ontrol  Scan the internet !or the lowest available price on selected routes   6e!ine the correlation between each !light departure and the competition 4stablish business rules !or matching lowest available price on a !light departure basis

"ompetitive -ri&ing "#allenges 5e consistently !ind e'amples where an airline could improve boo&ings and revenue by ma&ing dynamic pricing changes or better communicating their value proposition. $t is di!!icult !or airlines to price per!ectly becauseH %hey are !orced into a matchQdon’t match game %here is no single right price !or a population %hey don’t have enough real)time customer in!ormation

35

%hey don’t &now what else is displayed on the screen at the time o! sale Redu&e Internal -ro&essing "ost Strategies and -oli&ies

-ig.8 SourceH 5.O.%. "lom J: :ricing ^ Revenue Management 4urope =(M "arcelona , ^ 8 March 700D Reducing internal processing costs involves e'amining strategies and policies to improve business processes and to set guidelines and policies about how and when to ad*ust strategies in response to the competitive mar&et place. -light per!ormance categories are very use!ul when applying a systematic approach to seat allocation strategies. 5e can see nine possibilities in a two dimensional perspective. +nother dimension might include !light treatment in a competitive or non)competitive environment.

3

.lig#t Management Strategies

-ig.9 SourceH Ferry -oran "ritish +irways $+%+ RM ^ :ricing .ct. 700T %his table illustrates how !light categories can be used to manage !lights on a route by day)o!)wee& and time)o!)day. +n on)going ad*ustment to the categories is ta&ing place to account !or special events as well as patterns and trends over time.

8e+ revenue management &#allenges
       $ncreased competition :assenger %ra!!ic Arowth +verage airline yields decline Restriction !ree pricing + new pricing model :rice simpli!ication Jalue positioning !or mar&et segmentation

3!

"#apter$3 .uture of Revenue Management
5hat will drive the !uture o! RMB     :ricing transparency /omputing power and database manipulation Understanding o! consumer behavior 0especially web analytics1 /onsumer tolerance !or being IRM’d’.

:ricing transparency has grown #uic&ly over the past several years. -rom the days when pricing was transparent only to the travel agent to the advent o! online agencies li&e 4'pedia %ravelocity and .rbit9 to the creation o! in!ormation conduits li&e =aya& and Side&ic&. Such transparency has made the other airline’s selling price much more relevant to the determination o! my optimal selling price such that RM systems will need to get much better at &nowing the competition’s price and incorporating it into the calculation o! !uture demand and the elasticity o! demand. $ e'pect that airlines will !igure this out gradually over the ne't ten years. /omputing power will ma&e the incorporation o! competitive prices viable. $t will also enable more tailored o!!erings 0such as genuine one)to)one pricing1 as airlines learn how to use what they &now about a customer to alter the o!!ering. "ut in order to customise the o!!ering they will need to &now a lot more about their customers than they do today. %hey will need to learn how to use some o! the power!ul data that can be gleaned !rom the clic& streams o! their website visitors. +nd to combine other data sources with their vast databases o! !re#uent !lier travel. -inally all o! this Iscience’ will be !or naught unless it provides a buying environment that customers li&e 0or at least will tolerate1 and good value. +irlines can *usti!y changing prices constantly charging !ive times more !or the last seat than the !irst one or charging more on the -riday be!ore /hristmas than on /hristmas 6ay all we want but the travelling public will

3"

decide i! they buy it. $t remains to be seen i! there is su!!icient value in !oregoing the e'tremes o! RM in order to provide a more palatable pricing structure or G on the contrary G i! the public demands greater access to last seat availability and is willing to pay !or the privilege. So how will it play outB %o borrow a trite phrase I%he /ustomers will decide’. %hey will determine such critical !actors asH • • %he importance o! loyalty and loyalty programs %olerance !or non)re!undability G paying a he!ty price !or holding inventory versus the hotelQrental car model • %rans!erability G shouldQwill a customer that bought a non)re!undable be able to trans!er usage • :redictability G how important will it be to customers to I&now’ that a !are is !air that it won’t go down later that it isn’t materially higher than !rom a di!!erent channel or a competitive serviceB 1. (oyalty programmes will get better and even more important than they are today. +irlines will ma&e it even more worth your while to &eep using them to the e'clusion o! competitive options through the use o! escalating bene!its. +nd their use o! personal data about customers will enhance these o!!erings. 7. +irlines will have to !ind a better way to balance the inventory holding costs and the customer’s need !or some !le'ibility. %hey will use inventory controls and bid prices to govern which changes are !ree or ine'pensive and which ones are not. S. Security concerns will continue to limit the buyer’s rights to trans!er tic&ets but airlines will !igure out how to allow buyers to trans!er the value o! the tic&ets and the rights to a speci!ic seat on a speci!ic !light to a new customer. T. +irlines will never agree on the need !or !are predictability. Some will mar&et themselves as Ievery day low !ares’ and some will mar&et their last seat availability !or

3#

their best customers who pay the Ibest’ 0ie the highest1 !ares. Fust li&e in other retail business. SourceH "y Scott 6. 3ason Jice :resident GRevenue Management at +merican +irlines Fournal o! Revenue and :ricing Management

*#e &ontinuing evolution, "ustomer$&entri& revenue management
$n an e!!ort to get closer to the consumer airlines are investing in data mining business intelligence and advanced data analytics to understand customer traits behaviors and pre!erences in order to improve customer retention ac#uire new customers and ma'imi9e the revenue)generation potential !rom the customer base. %he renewed !ocus on customer loyalty and the customer e'perience are &ey areas o! investment !or airlines to di!!erentiate themselves. /ustomer relationship management 0/RM1 applications automate the customer)!acing interactions between an enterprise and a customer based on an ac&nowledged !act that it costs three times as much to ac#uire a new customer as it ta&es to retain an e'isting customer. %raditional business processes that are the !ocus o! /RM within an enterprise are mar&eting sales and service.

5ey 7nablers of "ustomer &entri& Revenue Management
/ustomer)centric revenue management is a /RM enabler to increase an airline’s pro!itability based on customer insight. %raditionally it has been the role o! mar&eting in an airline to ac#uire new customers in the most cost)e!!ective manner. $n today’s environment however it re#uires a combination o! mar&eting revenue management and real)time inventory control to !acilitate one)to)one targeted responses to manage the customer li!ecycle across all customer touch points. -igure 10 illustrates the &ey enablers o! customer)centric revenue management. 5hile these initiatives can be sometimes viewed as independent initiatives they need to come together in a cohesive !ramewor& !or the practice o! e!!ective revenue management in a changing landscape.

4$

-ig. 10 =ey 4nablers o! //RM .ore&asting based on &onsumer preferen&es   + !orecasting approach that !ollows the consumer demand process + top down approach that provides insight into mar&et demand and consumer pre!erences • 6isplacement time elapsed time competitor schedulesQ!ares aircra!t type restrictions etc. • • -irst choice demand $mproved estimates o! recapture up sell and price elasticity

41

-orecasting conditional demand based on the selling !are    Aiven R:rice)6emand Relationship calibrate the R:rice /urveP 4stimation o! :rice)6emand Relationship + causal model !or !orecasting conditional demand a a Oistoric price points observed boo&ings ^ inventory controls 4stimation o! buy)up buy)down behavior using pro'imity o! !ares by days prior to departure

-ig.11 :rice sensitivity graph by relating it with demand and days o! departure Significant improvements can be achieved in following the demand process D augmented with a consumer choice model

42

.are Simplifi&ation Che pricing revolution and impacts on revenue management %raditional -are Structure   4ach product is independent and governed by a set o! restrictions Segments passenger demand by creating !ences. $ndependent uni#ue products by !are class  %raditional revenue management !orecasts demand by !are class and determines optimal allocations against available capacity -are Simpli!icationH .riginal (// Model      :roducts 0!are classes1 are not independent. (ower !are di!!erential Multiple !ares are !iled but with the same identical restrictions :romotes 100E sell down to the open !are class due to the absence o! !ences Revenue management should !orecast dependent demand 0based on current !are class that is open1  +ctive monitoring and closure o! selling !are at the right time is re#uired to promote sell)up to a higher !are %he pricing revolution and impacts on revenue management  Oybrid !are structure  :roducts 0!are classes1 with identical restrictions are not independent.  Multiple !ares are !iled with identical restrictions

43

 :romotes less than 100E sell down since multiple classes with di!!erent restrictions may be open  Restriction !ree tari!!s can be on local or connecting mar&ets  Revenue management should !orecast dependent demand  +ctive monitoring and closure o! selling !are at the right time is re#uired to promote sell)up to a higher !are

44

Alternate segmentation Aetting closer to the customer re#uires an understanding o! the data and an investment in a data warehouse. 5ith an investment in the storage and analysis o! passenger name record and tic&et data airlines are interested in segmentation o! customers beyond the traditional boo&ing class that is used !or inventory control and distribution o! availability through the Alobal 6istribution Systems. -igure 17 illustrates a typical enterprise data management in!rastructure. /reation o! a data management in!rastructure supports a deeper understanding o! the customer base and prevents customers !rom leaving through the revolving door.

Key metrics and continuous feedback Propensities, attitudes, behaviors… Age, income, location… Name, address, email,… PNR, Ticket data, …

KPI’ s Preferenc es

Attribute Management

Demographic Data

Customer Profile Operational Data (transactional)

-ig.17 +n +irline’s 4nterprise 6ata Management $n!rastructure "randed products are another e'ample o! alternate segmentation beyond the boo&ing class. $n an e!!ort to overcome the perception o! an airline seat as a commodity a &ey initiative in the airline community is to !ocus on the brand describe the uni#ueness o! the products o!!ered !or sale and communicate the product o!!ering to the customer. %his is based on the !undamental premise that the airline seat is not a pure commodity as it is currently perceived. %able T illustrates e'amples o! branded products adopted by some airlines.

45

Airline +ir /anada

Branded -rodu&ts %ango %ango :lus (atitude 4'ecutive %iny 4conomy :remium 4conomy "usiness :romo 0:romotional1 4cono 0%ourist1 -le'i 0-le'ible1 :lena 0-ull Rate1 4*ecutiva 0"usiness1

bmi

+vianca

%able TH 4'amples o! branded products 4ach branded product has uni#ue traits that are essentially so!t #uali!iers bundled into the product de!inition such as access to pre)reserved seats !re#uent !lyer miles credit percentage lounge access baggage count allowed at no charge etc. %he standard segmentation o! customers !or revenue management is based on the boo&ing class. -rodu&t Anbundling, ‘Distribution with Differentiated Content@ $mpacts o! customer centric revenue management on distribution 5ith the growing emphasis on ancillary products as a revenue stream that can augment the bottom line airlines re#uire the capability to sell distribute and settle ancillary services across all channels o! distribution. %his implies that a capability is re#uired to set the prices !or ancillary services distribute products with di!!erentiated content and conduct !inancial settlement across all channels. %his has signi!icant impacts on the capabilities o! current airline reservations systems Alobal 6istribution Systems and revenue accounting. :romoting ancillary services also has an impact on revenue management. $! certain customer segments are more li&ely than others to consume ancillary services this should be !actored into the decision)ma&ing process when discount allocation controls are

4

established on an airline’s reservations inventory system. Oence the average passenger revenue !or a boo&ing class can be augmented with the ancillary revenue !orecast be!ore allocations are determined to ensure that seats protected !or boo&ing classes with an ancillary revenue upside receive additional seats. %his re#uires a !orecast o! e'pected ancillary revenues by customer segment based on historic consumption which can then be added to the average !are value o! the boo&ing class to get a true representation o! contribution when the networ& is optimised. /urrent revenue accounting systems do not aggregate ancillary services consumed by a !light segment. Oence the challenge is to enhance the e'isting revenue accounting systems to trac& the usage o! ancillary revenues by !light segment. +ncillary Services 0+ir and %ravel 4'tras1    $n 700T +%:/. estimated the opportunity value !or a global solution !or ancillary services to e'ceed L9 billion in revenue Unbundling o! airline products is becoming a reality − Some airlines are e'perimenting in the distribution o! a variety o! in)!light products and services 0i.e. pre)paid seats entertainment etc1 based on customer insight − .ther airlines !ocus on selling optional !le'ibility with the use o! their !ares 0upsell Q rule buster1  +n independent survey conducted by (e!lein +ssoc. in Fanuary 700[ showed that many travelers would pay !or e'tra per&s such as more overhead bin space in) !light internet access etc.  Ryanair reported that its ancillary revenue rose S1E in the #uarter ended Fune S0th 700[ outpacing its 70E increase in tra!!ic -ri&ing of Attributes  :ricing o! attributes based on customer willingness to pay baggage chec& meals

4!



Jaries by mar&et

4"

Differentiated "ontent  /apability to e'plain to a customer the products o!!ered by an airline.

4DS C Reservations Initiatives  :rovide the capability to sell distribute and settle air and travel ancillary services through the consumer direct and indirect channels o! distribution.  + &ey element o! merchandising is to di!!erentiate airline content provide capability !or upsell on inbound outbound and round trips. Intelligent -roa&tive .are Management, ‘Customer willingness to pay@ :ricing long !orgotten is a &ey enabler o! incremental revenues. %raditional airline pricing is limited to competitor !are monitoring and matching based on rules. %actical and strategic price leadership in a mar&et is increasingly being viewed as a competitive weapon. %actical pricing is the traditional !are management process o! responding to a !are action ta&en on a speci!ic !are in a mar&et by a competitor. Strategic pricing has a longer term view and is the process o! promoting an entirely new tari!! structure !or a mar&et. %he time dimension and !are management intelligence by %actical Js Strategic :ricing

4#

%able D *a&ti&al  4valuate and recommend a smart response based on a competitor’s speci!ic action 0!are speci!ic1  %ime window is 0 days 8 S months  Response to competitor actions based on #uality o! service and prevailing !ares  -ul!ils immediate ob*ectives with limited uncertainty  +ctive monitoring o! competitor !ares to e'ecute Rsmart) matchingP    Strategi& 4valuate and recommend a new tari!! structure 0all !ares1 !or the mar&et %ime window is S months8 17 months :rice leadership and implementation o! a sound pricing strategy  (ong term ob*ectives and greater uncertainty  :ro)active pricing based on customer willingness to pay and prevailing competitive mar&et conditions

%raditional airline pricing has relied on reactive !are matching to respond to competitor actions. Sometimes the reaction ripples through other mar&ets or di!!ers !rom the original change inducing a se#uence o! changes. %he ob*ective o! reactive !are changes is o!ten to match a competitor’s !are to preserve mar&et share. %raditional tactical !are matching can be replaced by determining the right response based on the #uality o! service o!!ered by the competitor that initiated the !are action. Oence the tactical !are response to a speci!ic !are action by a competitor can be an intelligent response as a !unction o! the #uality o! service. $! the #uality o! service o!!ered by a competitor that initiated the action is in!erior a !are match response may not be the desired alternative.

5$

5hile tactical pricing is !ocused on a response to a speci!ic !are action by a competitor strategic pricing is a pro)active approach to !iling a new tari!! structure !or a mar&et based on the desired pricing ob*ectives and business constraints. %he !irst step in the strategic pricing process is to de!ine the pricing strategy by mar&et or mar&et entity 0group o! mar&ets1. %he ob*ectives and goals almost always vary by mar&et or mar&et entity which represents a group o! mar&ets where the !ares are related. + strategy !or one mar&et or a group o! mar&ets may be very di!!erent !rom an alternate mar&et or mar&et entity. -or e'ample a speci!ic mar&et may have an overriding goal o! ma'imising mar&et share because there are dominant competitors and to be a player re#uires revenue volume. (i&ewise !or a di!!erent mar&et the airline may have the dominant li!t and hence the ob*ective o! ma'imising margins instead o! volume may be more appropriate. $n some situations a mar&et may be e'periencing very low boo&ings and hence improving the boo&ed load !actor on the pertinent segments may be a primary ob*ective. %he mar&et strategy ob*ective should satis!y the business constraints that need to be imposed be!ore a tari!! structure is generated !or the mar&et. %ypical e'amples o! business constraints are the price relationships that need to be maintained between the di!!erent tari!!s and competitor response to a !are action. "esides the potential increase in revenues with a !are action a secondary advantage o! using a strategic !are optimisation model is the introduction o! price consistency in a mar&et over time. %he decision variable used in the strategic !are optimisation process is the !are category which represent the broader customer segments where each product is de!ined based on a combination o! one or more o! the associated restrictions. %ypical attributes that ma&e up a !are category are • -are %ype 0+dult 4conomy G I4/.’ Military G IM$(’ Aovernment G IA.J’ /hild G I/O6’ Seniors G I+A4’ Student G IS%U’ "ereavement G I"RJ’1 • +dvance purchase restriction 0S+: G I0S’ ,+: G I0,’ 1T+: G I1T’ 71+: G I71’1

51



Minimum stay restriction 0Saturday 3ight Stay G IS’ M$3 S G I0S’ M$3 , G I0,’1



.ne)way versus round)trip !are 0I>’QIR’1

+ssociated with each !are category are multiple !are basis codes that can a!!ord subtle variations in restrictions and prices. .ne or more !are categories are associated with a !are class which is the level at which inventory is controlled. %he status o! a !are class 0available !or sale or not available !or sale1 is then determined by revenue management updates are established in the host /RS and then distributed to all the sales channels. -or pricing analysts to initiate an intelligent !are action is a multidimensional problem. -or e'ample setting prices too high may create an undesirable price image with respect to the competition and setting them too low may result in lost margins price wars and eventual brand erosion. %he ob*ective is to set and manage prices based on the strategic goals !or the mar&et or mar&et entity where !ares are related between mar&ets while simultaneously maintaining price consistency and price)image targets. +n important consideration is also the current mar&et share !or the speci!ic mar&et or mar&et entity and the assumptions made on competitor response which is the percentage o! other airline 0.+1 mar&et share that will match the pricing initiative. -igure 1S illustrates the !are action dilemma !aced by a pricing analyst. Aiven the historical data on average !ares and tra!!ic distribution by !are category !or a speci!ic period 0eg Summer 700,1 the pricing analyst has to determine price levels !or the target 0Summer 70081 period. Speci!ically 0i1 5hat is the initial price !or !ull ; 0economy !are with no restrictions1 !or the target periodB 0ii1 5hat are the discounts o!! !ull ; !are category that should be established to achieve a desired tra!!ic distribution across all !are categoriesB

52

-ig.1S %he pro)active !are action dilemma .nce the price discounts at the !are category level and the value o! !ull ; are &nown the pricing analyst can introduce multiple variations at the !are basis code level and determine the value o! the !are based on the historic tra!!ic !low !rom the calibration period. %he simple rule o! thumb is to ensure that the selling !are basis codes associated with a !are category average out to the recommended discount at the !are category level. A&&urate Availability, ‘Sense and respond inventory controls@ %he e'ponential growth in online boo&ings over the past decade has provided customers with instant access and visibility into competing schedules and !ares through the web supermar&ets li&e %ravelocity and 4'pedia. %his unparalleled transparency o! schedules and !ares over the internet has propagated a bargain hunting mentality among leisure online travelers resulting in a disproportionate growth in availability processing due to

53

the increased shopping activity. +s a result the need !or greater revenue and inventory control has not been greater. Revenue management is going real time to respond to mar&et conditions and shopping alerts is an integral part o! the decision ma&ing process to respond to real time alerts. %raditional RM does not consider competitor availability in the decision ma&ing process. %o be aware o! competitor there is a great emphasis on real time revenue management. Real time revenue management is the ability to sense and respond to prevailing mar&et conditions 0competitor schedules and availability1 in real time and provides a mechanism to override traditional revenue management controls. $n a nutshell /ustomer centric revenue management is all aboutb  -orecasting demand and no)show behavior based on consumer traits and pre!erences.  Simpli!ied pricing is here to stay with the emergence o! low cost carriers as a ma*or !orce worldwide.  $n an attempt to get closer to the customer airlines are contemplating alternative segmentation strategies to manage seat inventory.  +ccurate availability based on the competitive landscape and stated mar&eting ob*ectives is on the radar o! most leading airlines.  :rice leadership is increasingly being viewed as a competitive weapon.

't#er Important points in future of RM
8e+ "ustomer 'ptions %he ma*or US airlines have been surveying customers with the intention o! unbundling !are products. %hey are all loo&ing at this as a way to o!!er lower !ares at a lower cost in order to compete with the low)cost carriers and at the same time allowing customers to buy)up !or better service.

54

/ustomers purchasing airline tic&ets are o!ten interested in products such as !light insurance car rental or accommodations and may appreciate the convenience o! purchasing them at the same time. .ne option is to display icons that the customer can choose to clic& through to access these supplemental services a!ter they have !inished boo&ing and paying !or their !light. +nother option is to integrate the decision to purchase another product or service into the initial airline tic&et boo&ing process. + travel consumer is more li&ely to purchase an ancillary product such as !light insurance when its purchase is integrated into the airline tic&et Rboo&ing streamP. %his means that the customer is as&ed to ma&e this decision be!ore he or she has !inished boo&ing the !light. %he cost o! the e'tra product is added to the total bill be!ore !inal payment. 5hen Ryanair integrated an insurance product 0previously o!!ered on their site by clic&ing through a discrete icon a!ter purchasing the tic&et1 into the airline tic&et boo&ing stream the percentage o! customers purchasing the product rose !rom 7E to 10E. %o do this type o! dynamic product pac&aging airlines need computer so!tware that allows them to retain control over the boo&ing at a reasonable price. %his also means that you can only do this type o! mar&eting on your company website and not through a A6S. Some airlines pre!er the clic&)through option !or all ancillary products claiming that customers are smart enough to create their own pac&ages. 4ither option represents an opportunity to ma&e e'tra revenue with very little incremental cost to the airline. %he dreaded middle seat could be in high demand. %hatNs because it may soon cost more !or a coveted aisle seat or even a seat by the window. +irline revenue management is going a la carte. Some e'perts believe it is only a matter o! time be!ore passengers will pay according to seat location within the coach cabin. United !ired the !irst salvo with C4conomy :lus +ccess C in august o! 700,. -or a L799 annual !ee any passenger may re#uest an upgrade to the 4conomy :lus section o! the plane where the seat pitch can range !rom SC to DC more than a standard coach seat. %he 4conomy :lus section was !ormerly the province o! UnitedNs elite :remier !liers or others traveling on !ull !are tic&ets. 5ith +ccess United has assigned a premium price to the most sought)a!ter seats.

55

+irline services are going a la carte however a la carte options are not a new concept in other industries and itNs !ast becoming the norm. %he good thing about it is that customers can decide i! they want it or i! they donNt want it. "ut this is only the beginning because the mar&etplace is there and the tools are there on everybodyNs 5eb sites to assign a price to almost any option. +irlines might soon give customers the option to use the special chec&)in lines as well as the e'press security lines at many airports !or a !ee. Many travelers will probably gladly pay to bypass the #ueue. /ustomers could accept anything that seems li&e a reasonable proposition. R$ndividual pricingP is yet another step in the evolution. :riceline.com allows travelers to name their own price though travel vendors can re*ect any bid. .rbit9 o!!ers a product called C6eal 6etectorC which also allows users to establish a threshold price. .rbit9 sends customers an email i! the price drops below their set threshold. 5ith individual pricing .rbit9 would pass the 6eal 6etector price to the airlines and they would decide i! they want to meet the threshold price on a case)by)case basis. 8on$*i&)et Revenues 5ith ancillaries accounting !or 18 percent o! total revenues at Ryanair and this number still growing it’s no wonder airlines around the world are ta&ing notice. %he pro!it margins on ancillaries are much higher than on the commodity based sale o! seats and gross pro!it margins could be as high as T0 percent with very low selling costs particularly i! sold online. Ryanair says they are only scratching the sur!ace today and in !act two years ago they didn’t have a manager who had sole responsibility !or this area. +ir /anada has developed a set o! !are !amilies that show Rthe trade)o!! between !are levels and the services o!!ered. +vailable products are displayed with a clear mention o! the services associated with them. %he customer can be in!luenced by such displays. $! the customer as&s !or the economy cabin the airline can display a special business !are and the price di!!erence is *usti!ied by the value delivered to the customer. %o compete with the (//s +ir /anada has adopted a display that shows one)way !ares in a combinable calendar !ormat !or competitive routes.

5

+ir /anada is an e'ample o! an airline that has adopted a value)based pricing model in which customers choose !are products based on the attributes associated with each product. $n some mar&ets all advance purchase re#uirements have been eliminated. %hey have created !our branded R!are !amiliesP %ango %ango plus (atitude and 4'ecutive /lass. 4ach !are !amily has speci!ic attributes that target a particular mar&et segment. -or e'ample 4'ecutive /lass passengers have priority chec&)in baggage handling and boarding advance seat selection and access to special airport lounges. Since we &now that !le'ibility is particularly important to the business segment that this brand targets these tic&ets are also !ully re!undable. -are products in the least e'pensive !are !amily %ango do not have any o! these attributes. %ango tic&ets cannot be cancelled or e'changed without !inancial penalty and do not even earn !re#uent !lyer points. Measuring per!ormance o! ancillary product sales is still an area that needs to be wor&ed on as well as where to assign revenues. Dybrid Inventory 8esting $n response to competition and low)!are airlines some airlines are redesigning their business models. %his has created a need !or other inventory control techni#ues. -or e'ample since the $nternet has created an alternative distribution channel !or airline tic&ets a hybrid nesting structure has emerged to help manage re#uests !or seats via the $nternet somewhat independently o! the seats sold through more traditional means. .lig#t "ategory *reatment %he !light category treatment is to set price points according to the demand and passenger mi' associated with each !light day o! wee& and time o! day pattern. 3ot only do customers have buy up opportunities but now they also have buy across with the introduction o! !are !amilies.

5!

-ig.1T SourceH www.aircanada.com .nce customers have chosen their !light and !are !amily they can begin the online boo&ing process. 6uring this process they can add or delete certain Ra la carteP attributes which will increase or decrease their tic&et price. -igure 1T shows two options being o!!ered to a passenger boo&ing a %ango :lus seat. $! you don’t want to chec& any baggage LD will be deducted !rom the !are price. $n /anada 10 percent o! +ir /anada’s customers choose to opt)out o! chec&ing baggage. $! you would li&e to buy an onboard /a!c voucher !or meal service LD will be added. 7D percent o! +ir /anada’s domestic passengers opt out o! !re#uent !lyer mileage points !or a reduction in price.

5"

EA$la$&arteF A&&ess to 0ounge

-ig 1D Ra)la)carteP option allows access to the airport lounge !or an additional L7D. "#ange in t#e Air   :aper tic&ets soon to be a thing o! the past %ravel agency commissions have all but disappeared in 3orth +merica and 4urope    .nline sales through airline websites continue to boom A6S and airlines have signed peace agreements A6S Ra)la)carteP des&top agency solution

Many A6S operators are reinventing themselves in order to stay relevant as a distribution partner. $n addition to e'panding and diversi!ying their products and services their !ocus

5#

is on reducing internal costs and !inding ways that costs are borne by users not the airlines. %hey are now o!!ering more !le'ible subscription terms. Aalileo a leading global distribution system 0A6S1 and subsidiary o! %ravelport and +ir /anada today announced in +ugust 700, a multi)year agreement !or a revolutionary graphical agency des&top solution that will provide Aalileo)connected /anadian travel agents access to the !ull range and attributes o! +ir /anada’s innovative d)la)carte !are products and -light :asses. +ir /anada will become the launch customer o! the des&top solution developed by Aalileo and powered by +ir /anada’s direct)connect application programming inter!ace 0+:$1 plat!orm called R+/7U.P Aalileo is the !irst and only A6S to o!!er a !ull merchandising solution capable o! supporting the !ull range o! +ir /anadaNs !are and -light :ass products. %he new des&top solution o!!ers Aalileo agencies !ull product descriptions the ease o! a graphical display and prompts the user when product options are available. +ll in!ormation is seamlessly integrated into the travel agent accounting and bac&)o!!ice systems. 4lobal Distribution +s long as airlines sell outside o! their home countries where they are not &nown presences they will need an intermediary such as the A6S 5ith new smaller *ets the low)cost carriers need to get higher yielding passengers to ma&e money. %hey needed the reach the A6Ss give into corporate mar&ets and clients where they could get higher yielding tra!!ic. %he low)cost carriers are being approached by the A6Ss with o!!ers o! a lot more !le'ibility both on what they give and on the economics.

$

"redit "ards

LS billion certainly has the attention o! airlines because that is what they cumulatively spend worldwide each year on credit card merchant !ees. /redit card !ees are trending higher. Some card companies had recently hi&ed rates to SE well above the historic averages o! 7.7E to 7.DE !or carriers. +irline o!!icials report there is no negotiating with credit card companies. Oaving already cut labor commissions and A6S costs the carriers now see credit card !ees as the ne't opportunity !or meaning!ul savings. %hat leaves two optionsH !inding lower)cost alternatives to credit card payments or !oisting o!! credit card !ees on someone else )) ultimately the customer which could violate contracts with credit card companies and might even be illegal. Southwest and 3orthwest have introduced the :ay:al service on their 5eb sites. %his is especially interesting to customers who do not want to use their credit cards

1

"#apter$> "on&lusion
5ithin the airline industry mar&et dynamics and RM are evolving. Oowever the basic theory o! RM remains unchanged. %he primary ob*ective o! RM is to ma'imise the pro!itability o! the company by applying &nowledge about the mar&et and the competition and by using RM systems and tools e!!ectively. +s the evolving trend suggests revenue management /RM and how products are distributed are converging with strong inter)dependencies that re#uire a holistic view to understand business impacts and how the various customer touch points need to be managed. %he continuing evolution o! pricing and revenue management is a winning proposition !or both the airline and the customer. +irlines who succeed in hiring people with the right s&ill set establishing robust RM processes and applying RM principles through use o! optimal systems and tools are bound to increase incremental revenues

2

Referen&es
Aary :ar&ar Revenue ManagementH+n 4volutionary Revelation Singapore +ug 71)77 700, "en Jinod +dvances in inventory control Fournal o! Revenue and :ricing Management Jolume T 3umber T ,th .ctober 700D "en Jinod /ontinuing 4volution o! Revenue ManagementH /ustomer /entric Revenue Management 3ovember 79 700[ 4. +ndrew "oyd +irline 6istribution and the :ractice o! Revenue Management 6r. :eter :. "elobaba Revenue and /ompetitive $mpacts o! .)6 /ontrolH Summary o! :.6S Results $3-.RMS Revenue Management Section Meeting 3ew ;or& Fune ,)8 7001 4. +ndrew "oyd Revenue Management and 6ynamic :ricingH :art $ Ste!an :elt Revenue Management %utorial +A$-.RS Reservations ^ ;ield Management Study Aroup "erlin 1[)19 +pril 7007

Sponsor Documents

Or use your account on DocShare.tips

Hide

Forgot your password?

Or register your new account on DocShare.tips

Hide

Lost your password? Please enter your email address. You will receive a link to create a new password.

Back to log-in

Close