Sunday, July 13, 2014

7 Steps for airline route forecasting

1) Current demand - Collect origin and destination data on relevant markets.  The data is best collected at a passenger per day each way (PDEW) aggregation.  If possible the data should be broken into premium and economy with average fares.

2)  Current supply - Assess all the potential ways that relevant O&D markets can travel today.  The most common way to assess this information is with a Quality of Service (QSI) path build model.  These models will take into account nonstop flights, single stop flights, online connecting flights, code-share connecting flights, and interline connecting flights.  The QSI is a simple way to assign a value to each unique path to travel from O to D.  For example;

  • nonstop flights receive a value of 1.0
  • single stop flights (same aircraft/flight number) receive a value of .30
  • online connecting flights receive a value of .10
  • code-share connecting flights receive a value of .05
  • interline connecting flights receive a value of .02  
Calibration to historical data will give the precise values to use for your airline.


3) Build your flight schedule - Create the flights you want to forecast with a operationally desirable fleet type and block times.

4) New demand - Consider how demand will be changed by your new service.  The topics to consider are;

  • normal growth in the market since the last demand was captured.  This is mainly driven by population changes, income changes, and propensity to travel.
  • service stimulation due to the difficulty to travel in the market previous to this service
  • price elasticity if lower fares will be offered in the market
  • market substitution if this market has similar characteristics as other markets particularly if the market is a tourist destination
5) Market shares - The QSI model for relevant O&Ds is the primary determinant.  Other factors to be considered are;
  • reputation of airline
  • airline product preference (i.e. seats, IFEC, food and beverage)
  • connecting airport quality for connecting passengers
  • aircraft type
  • demand variation for time of year, time of day, and day of week,
  • elapsed time of journey
  • seat factors on competing services (if other services are full they can not carry all of the demand)
  • pricing
  • classes of service offered 
6) Cost allocation - Calculate variable and fixed costs.  Notable costs include fuel, aircraft ownership, crews, ground handling, airport fees, maintenance, and overheads.

7) Allocate revenue and calculate profit - Revenue on point to point travelers are easy to assign.  Connecting travelers revenue needs to be prorated across the flights used.  This is usually done on a distance basis but with more revenue being allocated to shorter flights.  Ancillary revenue also needs to be allocated.  The profitability calculations need to be done on a segment and network basis on both variable and total costs.  It is also important to consider how much of the revenue is incremental to the airline. 

Once you have completed these 7 steps use the correct profitability calculation to make your decision to operate, not operate, or recreate a better schedule in step 3.  The correct profit metric can vary based on the time period being assessed.  For example making a decision to acquire an incremental aircraft is different to the decision to allocate existing aircraft.