Agifors 2002 RMTutorial Stefan Poelt

Revenue Management Tutorial: 

Revenue Management Tutorial Stefan Pölt Network Management Lufthansa German Airlines stefan.poelt@dlh.de AGIFORS Reservations & Yield Management Study Group Berlin, 16-19 April 2002

What is Revenue Management ?: 

What is Revenue Management ? forecasting inventory control optimization

Revenue Management Definitions: 

‘Selling the right seats to the right customers at the right prices and the right time’ (American Airlines 1987) Revenue Management Definitions (Squeezing as many dollars as possible out of the customers) ‘Integrated control and management of price and capacity (availability) in a way that maximizes company profitability

Revenue Management History: 

RM was ‘invented’ by major US carriers after airline deregulation in the late 1970’s to compete with new low cost carriers Matching of low prices was not an alternative because of higher cost structure American Airline’s ‘super saver fares’ (1975) have been first capacity controlled discounted fares RM allowed the carriers to protect their high-yield sector while simultaneously competing with new airlines in the low-yield sector From art to science: By now, there are sophisticated RM tools and no airline can survive without some form of RM Other industries followed - hotel, car rental, cruise lines etc. Revenue Management History

Revenue Management Preconditions: 

Revenue management is most effective if the product is perishable and can be sold in advance the capacity is limited and can’t be increased easily the market/customers can be segmented the variable costs are low the demand varies and is unknown at time of decisions the products and prices can be adjusted to the market Revenue Management Preconditions

Revenue Management and Pricing: 

Revenue Management and Pricing Integrated RM and Pricing systems are not (yet) available Current practice is to exchange information

Revenue Management and Pricing: 

Revenue Management and Pricing Goal is to adjust the demand to the ‘fixed’ capacity Save seats for high-fare demand on full flights and channel low-fare demand to empty flights departure date demand capacity

Market Segmentation: 

Passengers are very heterogeneous in terms of their needs and willingness to pay A single product and price does not maximize revenue Market Segmentation price demand revenue = price • min {demand, capacity} capacity p1

Market Segmentation: 

Products (booking classes) are different in service (compartment) conditions (advanced purchase, Saturday night stay, non-refundability etc.) price Effective conditioning is essential for market segmentation (to prevent buy-down) RM is the last chance to mitigate the impacts of bad pricing decisions Market Segmentation

RM - Pricing - Scheduling: 

RM - Pricing - Scheduling - 3 Years - 1 Year - 6 Months - 3 Months Departure Flight Planning and Pricing Policy Planning the flight schedule Basic price structures (tariffs, conditions) Control parameters revenue management Tactical decisions Allocating and adjusting capacities Proactive and reactive pricing Strategic Decisions Fleet planning

RM - Pricing - Scheduling: 

Scheduling Pricing RM - Pricing - Scheduling

Revenue - Yield - Load Factor: 

Revenue - Yield - Load Factor Maximizing revenue is a balancing act between the contradictory goals of maximizing yield and maximizing seat load factor Upper management‘s motto alternates periodically between ‚increase load factor!‘ and ‚increase yield!‘ There are many combinations of load factor and yield which lead to the same revenue Since it is easier to monitor booked load factor than booked yield, management (and sales) often prefer a plane-filling strategy

Revenue Management Dilemma: 

Revenue Management Dilemma High-fare business passengers usually book later than low-fare leisure passengers Should I give a seat to the $300 passenger which wants to book now or should I wait for a potential $400 passenger? Most decisions in Revenue management are based on balancing risks, costs, or opportunities

Overbooking: 

Some (about 13% on average) booked passengers don’t show-up at departure due to double (fake) bookings missed connections etc. Overbooking to compensate for no-shows was one of the first Revenue Management functionalities (1970’s) Overbooking bkg 360 days prior departure time } no-shows cap

Overbooking: 

Sophisticated overbooking algorithms balance the expected costs of spoiled seats and over sales Typical revenue gains of 1-2% from more effective overbooking Overbooking booking limit capacity expected costs

Upgrading: 

Fixed cabin capacities do no suit demand in all cases Upgrading is a ‘virtual’ shift of capacity between cabins to allow more bookings in the lower cabin Upgrading Y C Y-demand C-demand

Forecasting: 

Most important forecasting items in RM are demand no-shows Forecasting is usually based on historical bookings The mass of things to forecast makes automation (computer systems) necessary Systems allow influences to react on changing conditions that are not reflected in stored booking history (fare changes, competitors, special events etc.) Forecasting

Forecasting: 

Forecasting The forecaster is a core module in RM systems revenue data current bookings historical bookings no-show data demand forecaster no-show forecaster fare-mix optimizer overbooking optimizer control parameters

Forecasting: 

Forecasting There are two possible consequences of bad demand forecasts: spoiled seats and bad fare mix (yield) As a rule of thumb, 10% improvement in forecast accuracy translates to 1-2% revenue increase If not covered by specific functionalities (sell-up, dynamic hedging, full fare/future protects) moderate over-forecasting increases revenue (especially at high-demand flights) There are two possible consequences of bad no-show forecasts: spoiled seats and denied boardings

Nesting: 

Nesting Almost all Reservation systems allow serial/linear nesting of booking classes Nesting prevents high-fare booking classes being sold out when lower-fare booking classes are still open

Optimization: 

Optimization Calculation of booking limits by booking class or Bid Prices EMSR robust and popular heuristic fare 400 300 200 cap protections seats

Leg - O&D: 

Leg - O&D Leg control can’t distinguish between local and connecting traffic leg 1 leg 2 low demand low demand leg control sufficient low demand high demand prefer connecting traffic by O&D high demand high demand prefer local traffic by O&D control O&D control can increase availability to long-haul passengers AND prevent long-haul passengers from displacing high-fare short-haul passengers

O&D: 

O&D Slogan in RM during the last years Preconditions structural (connecting traffic) technical (seamless link to CRS, O&D data base, etc.) organizational (market oriented RM organization) soft factors (management commitment, intense training) Obstacles complexity (complex algorithms, mass of data, source of errors) data quality (dirty PNR data etc.) costs (seamless availability, hardware, etc.) cheating (need of ‘married segments’ and ‘journey data’) But, around 2% increase of revenue!

... And Many More: 

... And Many More Group booking control Point of sale control Revision of RM decisions Data quality, outlier handling Reporting, monitoring, performance measurement Mathematics, algorithms, models etc.  Judy & Steve ...