FR2842002A1 - Method of storage and access to data in a computer booking system for airline travel reservations, uses access criteria based on departure and arrival locations for journey rather than flight number - Google Patents

Abstract

The database has rules for each type of information to be stored, associating a set of criteria defining journeys to which the information applies, and content representing the value of the information to be applied. The set of criteria of a rule comprises a pair of markets: one relating to the geographical zone of the departure, and the other related to the geographical zone of the arrival.

Description

The present invention relates to a method for storing and accessing

  data in computer travel reservation system databases. It also relates to a device capable of implementing this method. The invention will particularly find its application for systems

  air transport reservation computer.

  At present, the companies responsible for the management of such reservation systems by computer need a large number of people to ensure the maintenance of the data necessary for the sale of the products of 1o reservation. These data are each linked to reservation schedules which must be available for publication from the IT systems adhering to the global system. This can include both internal systems and

  systems external to these companies' systems.

  Data maintenance necessary for the reservation system by

  computer, as it is currently operated, poses several drawbacks.

  First, the data is applied to the different flight schedules, schedule by schedule. However, the organization of computer reservation systems has a logic currently oriented towards a representation of

  markets in geographic terms and not in planning.

  Another current drawback is that the data is organized by flight number and not in terms of the origin and destination of the flight. Another disadvantage of current techniques is the time required to generate the changes in the databases. A person is indeed necessary for the update of each individual flight but the changes assigned to a flight can imply multiple modifications, notably in different flight periods. A single flight change can result in several hundred line changes in schedules or schedules.

  reservation inventory records.

  Another disadvantage is that the modifications currently carried out involve strong human intervention, which increases the risk of error and

  inconsistency issues between flight schedules and inventories.

  The present invention overcomes the drawbacks of the techniques known so far and, to do this, it introduces new logic for database management in order to generate faster changes with

  greater precision and better suited to customer needs.

  In this context, the present invention provides a method of storing and accessing data which uses a mode and selection criteria making it possible to update or search for content corresponding to a type of information more efficiently and appropriate in terms of market corresponding to areas

distinct geographic areas.

  Other objects and advantages will appear during the description which follows

  of a preferred embodiment of the invention which is not, however, limiting.

  The present invention relates to a method of storing and accessing data in databases of a computer travel reservation system. According to the invention: - for each type of information to be stored, rules are created associating: ò a set of criteria defining the trips to which the information applies, ò content representing the value of the information to be applied; - the set of criteria of a rule comprises a pair of markets with a market of origin which is defined by the geographical area of departure of the trip and a market of destination which is defined by the geographical area of arrival of the trip, - we store the markets defined in a database,

  - we store the rules created in a database.

  According to preferred embodiments, this method is such that: - each market is defined by a geographic area chosen from the following types of geographic areas: airport, city, state and country, country, region

  geographic, world or a combination of these types of geographic areas.

  - we create a table in which we assign to each type of geographic area a rank of decreasing priority according to the precision of the geographic area. - the priority of each market is determined according to the priority rank of the type of geographical area which defines it or of the combination of the priority ranks of

  types of geographic areas that defines it.

  - the priority of each pair of markets is determined by summing the priority

  of the origin market and the priority of the destination market.

  - additional criteria are used in the set of criteria, - a weight is assigned to each additional criterion defining its degree of importance. - we access the stored rule corresponding to a travel search by: * entering the place of origin and the place of destination of the trip, * looking for the pairs of markets that agree with said places of origin and destination, * calculating the priority of each pair of matching markets for said places of origin and destination, * returning the content of the rule having the pair of markets whose

  priority has the lowest value.

  - if two pairs of matching markets have the same priority value, the content of the rule whose original market has the lowest priority value is returned. Preferably, for access to the rules: 10. you enter additional travel search criteria, 2. for each rule whose market pairs agree with the places of origin and destination, we calculate the weight of the rule corresponding to the sum of the weights of the additional criteria of the rule, 30. we select the rule which has the weight the higher, 40. if the additional travel search criteria do not match the additional criteria of the selected rule, the rule with the next lower weight is selected, 50. if the additional travel search criteria match the criteria additional to the rule selected last, the content of the rule is stored, 60. as long as the stored content only partially meets the additional criteria for searching for a trip, steps 30 to 50 are repeated,

  70. the content of the rules which has been stored is returned.

  The invention also relates to a device for storing and accessing data in databases of a computer travel reservation system capable of implementing the method described above. According to the invention, this device: - comprises means for storing the markets of origin defined by the geographical areas from which the trips depart and the destination markets defined by the geographic areas from which the trips arrive, - for each type of information to be stored, it includes means for entering rules associating: * a set of criteria defining the trips to which the information applies, ò content representing the value of the information to be applied

  - It includes means for storing the rules entered.

  The attached drawings are given by way of examples and are not limitative of the invention. They represent only one embodiment of the invention and

  will make it easy to understand.

  Figure 1 is a general schematic view of certain components of a

  computer reservation system.

  Figure 2 presents a diagram illustrating the addition of a pair of markets.

  Figure 3 is another diagram illustrating the addition of additional criteria

in a rule.

  Figure 4 shows an example of finding a rule applicable in the

part of a travel search.

  With reference to FIG. 1, a user can implement the method of the invention via a graphical user interface 6 which allows

  access other components of a computer reservation system 1.

  Among these constituents, a central server 2 ensuring the management of the rules of the system is present and connected to a central database 3. The database

  central 3 is able to store the rules which will be defined below.

  Furthermore, a geographic walking definition server 4 is also present and includes a walking database 5 able to store

  a list of markets which will also be defined below.

  According to the invention, the storage of information of new information is managed

different types.

  The types of information relate to the data assigned to the different

  trips. In the following description, an example corresponding to a

  preferred application to air transport where journeys correspond to neighbors. By flight is meant a transport of passengers between a point of origin (place

  of origin) and a point of destination (place of destination).

  In this example, different types of information are necessary and in particular: - boarding terminal numbers, - information concerning traffic restrictions, - rules relating to the distribution of tickets over the Internet, - rules relating to the supply of meals, - the types of reservation classes that are eligible To manage the content assigned to each type of information for each flight element contained in the databases of the reservation system, the

  The present invention uses rule creation.

  These rules include a first section corresponding to the criteria used

  for the selection of trips to which the type of information applies.

  Another part of the rule consists of the content corresponding to the

  description of the value assigned to the types of travel information

corresponding to the rule.

  To this end, a set of criteria for defining each rule is used.

  First, one criterion used is the creation of market pairs. In each pair of markets, there is a source market and a destination market relating respectively to the geographic area of departure and the area

  geographic arrival of trips.

  Each market can be an airport, a city, a state, a country, a region or even the whole world. We thus assign to the rules a geographic definition,

by market ,.

  Markets may not be homogeneous, that is, they may consist of a combination of different types of geographic areas. Of course, the definition of the markets must be consistent. For example, the NICE geographic area cannot be included in the geographic area at the same time

  French and in the Italian geographic area.

  The different markets created are listed in a table in which they are assigned a priority value. The value of the priority of a market is a function of the priority rank of the geographical area which defines it or of the combination of the priority ranks of the geographical areas which defines it (if the market is not homogeneous). By assigning a priority rank to the different types of geographic areas,

  the precision or the wide extent of the corresponding geographical area is reflected.

  We give in the table below an example of assignment of ranks of

  priority to different types of geographic areas.

  Type of geographic area Priority rank Airport i City 2 Country and state 3 Country 4 Geographic region 5 World 6 Using this table, we can associate, at server level 4, the priority value with each market corresponding to a parameter of Entrance. For example, usage considers the following non-homogeneous market definition: Ml = London + France, if the entry made by a user entering is: Charles de Gaulle airport, the market Ml matches this search criterion with a value of priority of 4 because Charles de Gaulle airport agrees with the definition of country: France whose priority rank for this type of area

geographic is 4.

  If we consider a user input during a search operation corresponding to Heathrow Airport, market 1 agrees with a priority value of 2 because Heathrow Airport is included in the geographical area of

  London which is a type of geographic area with a priority rank of 2.

  By calculating the priority of each market, we can obtain the value of the priority of a pair of markets by simple sum of the priority values of each

constituting market.

  As an additional example on this subject, if we consider the following pair of markets that has been created: FRANCE - NICE + GREAT BRITAIN -i NEW ZEALAND + SYDNEY +

MELBOURNE,

  The two markets that make up the pair are heterogeneous. Their priority is made with reference to a pair of places of origin and destination returned by

  the user, for example during a search operation.

  Thus, the user entering the place of origin AEROPORT CHARLES DE GAULLE, will get a priority value of the original market of 4 (rank of the geographical area FRANCE). Furthermore, if it enters SYDNEY airport as the destination, the destination market will have a priority value of 1 (corresponding to

  priority rank of the SYDNEY geographical area).

  Consequently, the priority value of the pair of markets thus formed

io is equal to 4 + 1 = 5.

  It is easy to understand that the definition of such pairs of markets and the allocation of content of types of information is not always sufficient for the

definition of precise rules.

  To refine the selection thus made by the market pairs, iS advantageously uses additional criteria which are an integral part of the whole.

  criteria defining the trips to which the information applies in the rule.

  In addition, we assign a weight to each additional criterion in order to define its

degree of importance.

  An example of additional criteria is given in the table below.

  usable with their respective weight.

  Additional criteria Weight Flight group 64 Type of shared use 32 Flight period 16 Day of the week 8 Type of aircraft 4 Departure time 2 Flight duration I A set of criteria associated with a rule may include a

  combination of additional criteria as presented above.

  In this context, the total weight of the additional criteria assigned to a rule

  is the sum of the individual weights of these criteria.

  For example, for a rule whose only additional criterion is the flight group, the total weight is 64. This weight is therefore greater than that of a rule whose additional criteria are flight period + day of the week + time starting from

  vol (whose weight is 16 + 8 + 2 = 26).

  The combination of these additional criteria and the definition of market pairs ensures an unambiguous creation of rules making it possible to assign a

  content of a type of information to travel data.

  Of course, we check the integrity and consistency of the data thus contained in each rule. In particular, if additional criteria are of the same type and have the same weight to define different rules, the system will check that there is no overlap between the values assigned to these different criteria. Referring successively to FIGS. 2 to 4, different figures are given below

  examples of implementation of the method, for different functionalities.

  Figure 2 illustrates a possibility of adding or updating rules.

  First, the process starts with a check that the market pair entered does not already exist for the type of information considered. If such

  is the case, an error information is returned by a conflict indicator.

  If this pair does not yet exist for the type of information considered, it is checked that it does not imply an overlap with other pairs of markets. To this end, we seek the origins and common destinations in other

market elements.

  If such an overlap exists, conflict information is returned by a

  conflict indicator. Otherwise, we add the rule thus constituted by default.

  Then update the history and end the operation.

  To add additional criteria to a rule so predefined, we

  performs the process shown in Figure 3.

  In this context, the predefined rule obtained by the steps of Figure 2 is open. If it is not found in the system, error information is

returned.

  If it is found, the priority of the additional criterion entered at input is calculated. We then search if additional criteria having the same priority

already exist.

  We then check if there is a possible overlap between the value of

  criteria in such a case additional criteria with the same priority.

  If there is no overlap, the additional criterion can be added to the

  rule. If there is an overlap, we return "conflicting rules" information.

  A third possible functionality of the method of the invention is given below, according to which a rule applicable to data is sought.

  Entrance. This example is illustrated in Figure 4.

  As input, the user enters a pair of places corresponding to the place of origin and the place of destination desired for his search. Optionally, other additional search criteria can be entered such as the desired flight period. Generally, the places of origin and destination consist of

airport names.

  At this stage, we ask server 2 for the list of market pairs

  corresponding to the criteria entered as well as their priority value.

  We open the corresponding rules to these pairs of markets which

  agree for the type of information sought.

  The weight of the additional criteria of each rule is then determined for

  define the total weight of the rule.

  We can thus classify the processing order of the different rules that have been

returned.

  To do this, we start by selecting the rule with the highest total weight. If it turns out that several rules have the same weight, we start by selecting the one that has the most precise priority for the pair of markets. In addition, if it also turns out that the rules have the same priority level of market pair, we first select the one that has the priority level of the original market.

the most accurate.

  Once the rule is thus selected, it is checked whether the additional criteria match the search criteria entered by the user. If it turns out that these criteria match, the content associated with the rule is stored. Otherwise, we repeat the step of selecting another rule, starting with the rule of

immediately lower weight.

  When the rule content has been stored capable of informing the user for

  respond to its request, we return the final content thus stored.

  It is possible that the content of a rule only partially informs the user in his search. For example, the content of a rule can only relate to a particular flight period that does not fully cover the flight period sought by the user. In this context, once the content of the selected rule has been stored, the rule selection operations are again repeated in order to search in the successive rules for the additional content necessary for the response. To further illustrate the operation of the present invention, two examples detailed in the rule search corresponding to a

io user request.

  In a first example, 4 markets are defined as follows: Ml = Great Britain + France - Charles de Gaulle M2 = Great Britain + Bangkok + Singapore M3 = Europe M4 = Bangkok Under a traffic restriction type rule, the The following rules are defined: Criteria Identification of the rule Content of the rule Weight Pair of markets: M3 - M4 R5 K 64 + 16 = 80 Flight group BAl-10 Flight period January 20 Pair of markets: M1l- M2 R4 R , L 64 + 16 = 80 Flight group: International Flight period: January 1 to 15 Market pair: M1l- M2 R3 W 64 + 8 = 72 Flight group: Indoor Day of week: 7 Market pair: M3 - M4 R2 L 16 Flight period: January 10 to 20 Pair of markets: Ml - M2 Rl K 0 The flight period entered is:

  LONDON HEATHROW - BANGKOK - BA9 - 1 OJAN-31 JAN - 1234567

  Two rules have the same priority (80) and the entry matches all the criteria. The process determines which rule will be selected by selecting the most precise market pair: it M3 matches London Heathrow with a priority of 5 M4 matches Bangkok with a priority of 1 Ml matches London Heathrow with a priority of 4 M2 makes match Bangkok with a priority of 5 Therefore the pair of markets M3-M4 has a priority of 5 + 1 = 6 and the pair of

  markets M1-M2 has a priority of 4 + 1 = 5.

  The process will therefore select the market pair Ml-M2 (rule R4) which

  is the most precise market pair.

  Rule R4 applies because BA9 is part of an international flight group.

  1o On the other hand, as there is a flight period, rule R4 only applies from 10 to

January.

  The process must therefore find another rule. It selects the second rule which has the same priority but whose market pair is less precise (at

  know rule R5) which only applies on January 20.

  As the flight is not completely defined, the method will search for the rule which has the best priority, namely rule 3 which has a priority equal to 72. On the other hand, as one of the criteria of this rule is internal and that the entry is international (BA9 is part of an international flight group), this rule R3 does not

not apply.

  The procedure then looks at rule R2 which has a priority of 16 and notes that it applies from January 16 to 19. The Rl rule, which has no priority, applies from 21

as of January 31.

  Therefore, the rules that will be returned are: - R5 which applies only on January 20 - R4 which applies from January 10 to 15 - R2 which applies from January 16 to 19 - RI which applies from January 21 as of January 31 In a second example, the 4 markets are defined as follows: Ml = Japan + France M2 = Australia M3 = World M4 = Asia Within the framework of a traffic restriction type rule, the following rules are defined: Criteria Identification of the rule Content of the rule Weight Pair of markets: M3 - * M4 Ri K 64 + 16 = 80 Flight group: BAl-10 Flight period: January 10 to 15 Pair of markets: M3- * M4 R2 R, L 16 Flight period: January 10 to 20 Market pair: M2-> MI R3 W 16 Flight period: January 10 to 20 Market pair: M3 - * M3 R4 L 0 The flight period entered is:

  MELBOURNE -> TOKYO - BAI - 1 OJAN-20JAN - 1234567

  The method will select the RI rule having a priority of 80. However, as the rule has a flight period which does not cover all the parameters of

  the entry, the user will have to look for another rule.

  Two rules have the same priority of 16 and match all the criteria. The process will determine which rule will be selected by selecting i0 the most precise market pair: M3 matches Melbourne with a priority of 6 M4 matches Tokyo with a priority of 5 M2 matches Melbourne with a priority of 4 MI matches Tokyo with a priority of 4 Therefore the pair of markets M3- M4 has a priority of 5 + 6 = 11 and the pair of

  MI-M2 markets has a priority of 4 + 4 = 8.

  The process will therefore select the corresponding M2-Ml market pair

  to rule 3 because it is the most accurate market pair.

  The flight is completely defined because all the parameters of the entry are covered by the criterion of the flight period. Therefore, the rules which will be returned are: - RI which applies from January 10 to 15 - R3 which applies from January 15 to 20 Of course, the process of the invention may include other

  functions for example linked to the deletion of previously created rules.

REFERENCES

  1. Computer reservation system 2. Central server 3. Central database 4. Point of sale server 5. Point of sale database 6. Graphical user interface

Claims (8)

  1. A method of storing and accessing data in databases of a computer travel reservation system, characterized in that - for each type of information to be stored, rules are created associating: * a set of criteria defining the trips to which the information applies, * content representing the value of the information to be applied; - the set of criteria of a rule comprises a pair of markets with a market of origin which is defined by the geographical area of departure of the trip and a market of destination which is defined by the geographical area of arrival of the trip, - we store the markets defined in a database,   - we store the rules created in a database.   2. Method according to claim 1 characterized in that each market is defined by a geographic area chosen from the following types of geographic areas: airport, city, state and country, country, region   geographic, world or a combination of these types of geographic areas.   3. Method according to claim 2 characterized in that one creates a table in which one assigns to each type of geographic area a rank of decreasing priority depending on the accuracy of the geographic area. 4. Method according to claim 3 characterized in that the priority of each market is determined according to the priority rank of the type of geographic area which defines it or of the combination of ranks of   priority of the types of geographic areas which defines it.   5. Method according to claim 4 characterized in that the priority of each pair of markets is determined by summing the   priority of the origin market and priority of the destination market.   6. Method according to any one of claims 1 to 5 characterized by   the fact - that additional criteria are used in the set of criteria, - that a weight is assigned to each additional criterion defining its degree of importance. 7. Method according to claim 5 characterized in that the stored rule corresponding to a travel search is accessed by: entering the place of origin and place of destination of the trip, * searching for the pairs of markets which agree with said places of origin and destination, * calculating the priority of each pair of matching markets for said places of origin and destination, * returning the content of the rule having the market pair whose   priority to the lowest value.   8. Method according to claim 7 characterized in that if two pairs of matching markets have the same priority value, we return the content of the rule whose original market has the lowest priority value. 9. Method according to claim 7 or claim 8 in combination with claim 6 characterized in that 10. one enters additional search criteria for travel, 20. for each rule whose market pairs match the places of origin and destination, we calculate the weight of the rule corresponding to the sum of the weights of the additional criteria of the rule, 30. we select the rule which has the highest weight, 4. if the additional travel search criteria do not match the additional criteria of the selected rule, the rule which has the immediately lower weight is selected, 50. if the additional travel search criteria match the additional criteria of the selected rule lastly, the content of the rule is stored, 60. as long as the stored content only partially meets the additional criteria for searching for a trip, steps 30 to 50 are repeated,   7. we return the content of the rules that have been stored.   10. Device for storing and accessing data in databases of a computer travel reservation system able to set up   implements the method according to any one of claims 1 to 9 characterized by the   fact - that it includes means of storing the markets of origin defined by the geographic areas from which the trips depart and the destination markets defined by the geographic areas from which the trips arrive, for each type of information to be stored, it includes means for entering rules associating: * a set of criteria defining the trips to which the information applies, ò content representing the value of the information to be applied   - that it includes means for storing the rules entered.