Unobscuring algorithm

Abstract

The system uses an unobscuring algorithm to determine dilution values and stimulation values. The system includes analysis methods and tools suitable for use in connection with yield management systems, inventory control systems, revenue management systems, and the like.

I claim: 
     
         1 . A method, comprising:
 converting, by a processor, booking information for a flight into a top-down format;   determining, by the processor in an iterative fashion, an unobscured demand for a current fare class using an expectation maximization algorithm and the booking information, in response to the current fare class being obscured; and   repeating, by the processor, the determining for each parent fare class until the unobscured demand for a highest fare class is obtained.   
     
     
         2 . The method of  claim 1 , wherein the expectation maximization algorithm in the current fare class is based on an expectation of bookings, an upper bound of unobscured bookings, a lower bound of unobscured bookings, a mean, a standard deviation, a cumulative distribution function of the upper bound, a cumulative distribution function of the lower bound, a probability density function of the upper bound, and a probability density function of the lower bound. 
     
     
         3 . The method of  claim 1 , wherein the unobscured demand for the current fare class is used as an upper bound for an unobscured demand for a parent class. 
     
     
         4 . The method of  claim 1 , wherein a value for the unobscured demand is converted to a discrete value. 
     
     
         5 . The method of  claim 1 , wherein the expected demand for the higher fare class is approximately 0. 
     
     
         6 . The method of  claim 1 , wherein expectation maximization algorithm uses a cohort comprising a predetermined number of flights and historical information from a time period. 
     
     
         7 . The method of  claim 1 , further comprising converting, by the processor, values of an expected unobscured demand out of the top-down format and into a discrete format, to reduce double counting. 
     
     
         8 . The method of  claim 1 , further comprising allocating, by the processor, each booking in an expected unobscured demand to the highest fare class that can be claimed. 
     
     
         9 . The method of  claim 1 , further comprising at least one of weighting or ignoring, by the processor, any portion of the booking information. 
     
     
         10 . The method of  claim 1 , wherein the determining the unobscured demand includes combining at least one of the unobscured demand from previous calculations or passenger information. 
     
     
         11 . The method of  claim 1 , further comprising obtaining, by the processor and using an unconstraining module, unconstrained demand information associated with a flight, wherein the booking information includes the unconstrained demand information. 
     
     
         12 . The method of  claim 11 , wherein the unconstraining module generates predicted booking information that differs from actual observed bookings. 
     
     
         13 . The method of  claim 1 , further comprising:
 receiving, by the processor, an unobscured booking table comprising the unobscured demand for the flight;   identifying, by the processor, constrained fare classes; and   converting, by the processor, the unobscured demand for each of the constrained fare classes to unconstrained demand information based on statistical information about a cohort of flights containing the flight.   
     
     
         14 . The method of  claim 13 , wherein the statistical information from the cohort comprises a comparison of a current constrained fare class to overall statistical information for the current fare class in the cohort. 
     
     
         15 . The method of  claim 13 , further comprising converting, by the processor, the unconstrained demand information into a discrete value. 
     
     
         16 . The method of  claim 13 , further comprising at least one of weighting or ignoring, by the processor, any portion of the unobscured demand. 
     
     
         17 . The method of  claim 1 , further comprising determining, by the processor, that opening a new lower fare class is a revenue positive decision. 
     
     
         18 . The method of  claim 1 , further comprising:
 determining, by the processor, a monetary stimulation value of opening a closed class, based on seat bookings and forecasted demand;   determining, by the processor and using the seat bookings and the forecasted demand, a monetary dilution penalty of opening the closed class; and   comparing, by the processor, the monetary stimulation value to the monetary dilution penalty to generate booking instructions,   wherein the determining the monetary dilution penalty comprises:
 determining, by the processor, a proportion of the seat bookings that are price oriented; and 
 multiplying, by the processor, the proportion of the seat bookings that are price oriented by a loss associated with a seat booking in the closed class, and 
   wherein the determining the proportion of the seat bookings that are price oriented comprises:
 identifying, by the processor, any of the seat bookings in the bookings table file that have been modified as a result of an unobscuring algorithm; and 
 identifying, by the processor, any of the seat bookings in the bookings table file that are associated with a price oriented customer. 
   
     
     
         19 . The method of  claim 18 , wherein, in response to the monetary stimulation value exceeding the monetary dilution penalty, the booking instructions comprise instructions to open the closed class for bookings. 
     
     
         20 . A method, comprising:
 receiving, by a processor, an unobscured booking table comprising the unobscured demand for the flight;   identifying, by the processor, constrained fare classes; and   converting, by the processor, the unobscured demand for each of the constrained fare classes to unconstrained demand information based on statistical information about a cohort of flights containing the flight.