CN-121986348-A - Oversubscription modeling based on machine learning

Abstract

Embodiments optimize hotel room reservations for hotels. For a first day of a plurality of future days, embodiments automatically determine an oversubscription limit for each category of hotel rooms for a hotel based on an objective function, wherein the hotel includes a plurality of different room categories. An embodiment receives a first reservation request for a first category of rooms for a first day. When the determined oversubscription limit for the first category room is not reached, the embodiment accepts the first reservation request. When the accepted first reservation request is in the business of a check-in hotel on the first day, the embodiment automatically determines to reject the first reservation request, accept the first reservation request, or upgrade the first reservation request to a higher category room based on the objective function.

Inventors

• A. Vahtinskay
• ZHOU ZIJIE

Assignees

• 甲骨文国际公司

Dates

Publication Date

20260505

Application Date

20240813

Priority Date

20230927

Claims (20)

1. 1. A method of optimizing hotel room reservations for hotels, the method comprising: For a first day of a plurality of future days, automatically determining an oversubscription limit for each category of hotel rooms for the hotel based on the objective function, wherein the hotel includes a plurality of different room categories; Receiving a first reservation request for a first category room for a first day; Accepting the first reservation request when the determined oversubscription limit for the first category room is not reached; When the accepted first reservation request is transacting a check-in hotel on the first day, a check-in decision is automatically determined based on the objective function to reject the first reservation request, accept the first reservation request, or upgrade the first reservation request to a higher category room.
2. 2. The method of claim 1, further comprising determining an oversubscription limit for the sequence of future days.
3. 3. The method of claim 1, further comprising generating a prediction of the new reservation for the future days using a first machine learning model.
4. 4. The method of claim 1, further comprising generating a prediction of cancellation of an existing appointment using a second machine learning model.
5. 5. The method of claim 1, further comprising: Generating a mixed integer optimization problem (MILP) modeling oversubscription limits and modeling check-in decisions, and The MILP is converted to a linear optimization problem using a set of polyhedral uncertainties for the total number of customers arriving at the hotel.
6. 6. The method of claim 5, wherein the MILP comprises a plurality of random variables, the set of polyhedral uncertainties eliminating the random variables.
7. 7. The method of claim 6, wherein the linear optimization problem comprises a plurality of constraints including uncertainty, further comprising formulating a robust counterpart without random variables for each of the plurality of constraints.
8. 8. The method of claim 1, further comprising: Responsive to transacting the check-in decision, the corresponding hotel card is automatically encoded.
9. 9. A computer-readable medium having instructions stored thereon, which when executed by one or more processors, cause the processors to optimize hotel room reservations for hotels, the optimizing comprising: For a first day of a plurality of future days, automatically determining an oversubscription limit for each category of hotel rooms for the hotel based on the objective function, wherein the hotel includes a plurality of different room categories; Receiving a first reservation request for a first category room for a first day; Accepting the first reservation request when the determined oversubscription limit for the first category room is not reached; When the accepted first reservation request is transacting a check-in hotel on the first day, a check-in decision is automatically determined based on the objective function to reject the first reservation request, accept the first reservation request, or upgrade the first reservation request to a higher category room.
10. 10. The computer-readable medium of claim 9, the optimizing further comprising determining an oversubscription limit for the sequence of future days.
11. 11. The computer-readable medium of claim 9, the optimizing further comprising generating predictions of new reservations for the future days using a first machine learning model.
12. 12. The computer readable medium of claim 9, the optimizing further comprising generating a prediction of cancellation of an existing appointment using a second machine learning model.
13. 13. The computer-readable medium of claim 9, the optimizing further comprising: Generating a mixed integer optimization problem (MILP) modeling oversubscription limits and modeling check-in decisions, and The MILP is converted to a linear optimization problem using a set of polyhedral uncertainties for the total number of customers arriving at the hotel.
14. 14. The computer readable medium of claim 13, wherein an MILP comprises a plurality of random variables, the set of polyhedral uncertainties eliminating the random variables.
15. 15. The computer-readable medium of claim 14, wherein the linear optimization problem comprises a plurality of constraints including uncertainty, further comprising formulating a robust counterpart without random variables for each of the plurality of constraints.
16. 16. The computer-readable medium of claim 9, the optimizing further comprising: Responsive to transacting the check-in decision, the corresponding hotel card is automatically encoded.
17. 17. A cloud-based hotel reservation system that optimizes hotel room reservations for hotels, the system comprising: one or more processors adapted to: For a first day of a plurality of future days, automatically determining an oversubscription limit for each category of hotel rooms for the hotel based on the objective function, wherein the hotel includes a plurality of different room categories; Receiving a first reservation request for a first category room for a first day; Accepting the first reservation request when the determined oversubscription limit for the first category room is not reached; When the accepted first reservation request is transacting a check-in hotel on the first day, a check-in decision is automatically determined based on the objective function to reject the first reservation request, accept the first reservation request, or upgrade the first reservation request to a higher category room.
18. 18. The system of claim 17, the processor further determining an oversubscription limit for the sequence of future days.
19. 19. The system of claim 17, further comprising a first trained machine learning model and generating predictions of new reservations for the future days using the first trained machine learning model.
20. 20. The system of claim 17, further comprising a second trained machine learning model and generating a prediction of cancellation of an existing appointment using the second trained machine learning model.

Description

Oversubscription modeling based on machine learning Cross reference to related applications The present application claims priority from U.S. provisional patent application serial No. 63/585,735 filed on 9/27 of 2023, the disclosure of which is incorporated herein by reference. Technical Field One embodiment relates generally to a computer system, and in particular to a computer system implementing machine learning based oversubscription modeling (overbooking modeling). Background Revenue management is the process of dynamically adjusting the price of goods or services in response to changes in market conditions or changes in supply conditions. The revenue management process is first introduced by the passenger air industry and has been emulated by other industries such as freight airlines, hotels, car rentals, shippers, advertising brokers, and the like. A very popular application of revenue management involves service providers accepting reservations for "date constrained services". Date constrained services involve imposing transaction-specific restrictions on the date on which a purchaser can use the services that he or she purchased. Examples of such restrictions include specified arrival and departure dates for airline ticket reservations and specified entry and exit dates for hotel reservations. Time constraints make it particularly difficult to estimate demand and thus determine optimal pricing that maximizes revenue/profit for a date-constrained service, especially in the hotel industry. Hotel revenue management may be considered an extension of aviation revenue management. While the methods developed for hotels are often applicable to airlines, the reverse is not always possible. The main difference is the nature of hotel room reservations, which can span several days, allowing for reuse of the room. Thus, room availability varies every day, as some rooms may be occupied by guests staying for longer periods of time. In contrast, the airline seat stores exist consistently on each flight regardless of the class of the flight (e.g., first class, business class, or economy class). Thus, strategies and methods for hotel revenue management are inherently more sophisticated and demanding than those of aviation revenue management. Disclosure of Invention Embodiments optimize hotel room reservations for hotels. For a first day of a plurality of future days, embodiments automatically determine an oversubscription limit for each category of hotel rooms for a hotel based on an objective function, wherein the hotel includes a plurality of different room categories. An embodiment receives a first reservation request for a first category of rooms for a first day. When the determined oversubscription limit for the first category room is not reached, the embodiment accepts the first reservation request. When the accepted first reservation request is in the business of a check-in hotel on the first day, the embodiment automatically determines to reject the first reservation request, accept the first reservation request, or upgrade the first reservation request to a higher category room based on the objective function. Drawings The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate various systems, methods, and other embodiments of the disclosure. It will be appreciated that the boundaries (e.g., blocks, groups of blocks, or other shapes) of the elements shown in the figures represent one embodiment of the boundaries. In some embodiments, one element may be designed as a plurality of elements, or a plurality of elements may be designed as one element. In some embodiments, an element shown as an internal component of another element may be implemented as an external component, and vice versa. Additionally, elements may not be drawn to scale. Fig. 1 is an overview block diagram of a hotel reservation system in accordance with an embodiment of the invention. Fig. 2 is a block diagram of a computer server/system according to an embodiment of the invention. FIG. 3 is a flow diagram of the functionality of the system of FIG. 2 according to an embodiment when used as a hotel revenue management system providing oversubscription modeling and accepting reservations and facilitating the check-in process. FIG. 4 illustrates an example reservation quota decision across the whole week in accordance with an embodiment. Fig. 5 illustrates example decisions for a service period on day s according to an embodiment. Fig. 6-9 illustrate an example cloud infrastructure that may implement a hotel chain operation 104, which hotel chain operation 104 may include the oversubscription modeling module 16 of fig. 2, according to an embodiment. Detailed Description Embodiments generate and use a machine learning based model to determine reservation limits during hotel reservation periods, including determining room oversubscription limits for each hotel room category. Embodiments further use