Server and method for processing requests for searching for on-demand services

By adjusting the filtering of the distribution radius and service provider list based on the confirmation allocation rate requested by the user, and combining factors such as predicting delivery experience values, demand suppression and user experience problems in the existing technology are solved, and the visibility and quality of service providers are improved. .

CN118974754APending Publication Date: 2024-11-15GRABTAXI HOLDINGS PTE LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202380030128.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2022-04-01
Filing Date
2023-02-23
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

The prior art controls demand by narrowing the delivery radius and increasing delivery costs when handling peak-time on-demand service requests, which may suppress demand rather than shape demand, resulting in a decline in user experience and service provider quality not meeting user expectations.

Method used

The server processor determines the distribution radius based on the requests of multiple users, generates a list of service providers, and ranks and filters based on predetermined factor values ​​such as predicted delivery experience value, revenue, profit and search matching scores, and provides users with Improve visibility and quality of service providers.

Benefits of technology

Without suppressing demand, the user experience is improved, the quality and visibility of service providers are ensured, and user needs are met, avoiding users from turning to other platforms.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118974754A_ABST
    Figure CN118974754A_ABST
Patent Text Reader

Abstract

Aspects relate to a server configured to: receive a request for a search; determining a distance from the location of the computing device to generate a first list of service providers within the distance; ranking the first service provider list based on a plurality of predetermined first factor values, wherein the plurality of predetermined first factor values comprise a predicted delivery experience value of each service provider in the first service provider list; generating a second service provider list from the first service provider list based on the ranking of each service provider in the first service provider list; determining whether a portion of the second service provider list is filtered based on a size of the second service provider list; and filtering the portion of the second service provider list based on the predicted delivery experience value of each service provider in the second service provider list.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] Various embodiments are directed to a server and method for processing a request for searching for on-demand services. Background Art

[0002] Due to the development of information technology, users can use computing devices to request on-demand services. On-demand services can allow users to meet their needs by immediately accessing goods and / or services. Users can use a user interface presented on a computing device to request on-demand services, such as delivery services or transportation services.

[0003] As an example, a user may use a computing device to request a search for an on-demand service (e.g., a food delivery order). The server providing the on-demand service may then aggregate the locations of available service providers (e.g., restaurants), available service types (e.g., food), estimated costs, and other information, and provide the aggregated information to the computing device so that the user can make a selection on a user interface presented on the computing device. Once the user makes a selection on the user interface, the server may assign a service contractor (e.g., a driver) from among a plurality of service contractors to deliver the selected food from the selected restaurant to the user.

[0004] Although the demand for on-demand services fluctuates greatly from time to time, the number of multiple service contractors can tend to remain relatively constant. For example, during peak periods (e.g., lunch time and dinner time), demand may far exceed supply levels. Typically, as a solution, the on-demand service platform can batch multiple orders to meet as much demand as possible. As another solution, the on-demand service platform can use a combination of reducing the delivery radius and surging delivery costs. The delivery radius can be reduced based on the confirmed allocation rate of multiple orders received in the area where the user is located. If the confirmed allocation rate is low, the on-demand service platform can reduce the delivery radius to control the visibility of the service provider, thereby allowing only short-distance delivery. The surge in delivery costs may suppress excessive demand.

[0005] However, the above solutions may suppress demand rather than shape it. A surge in delivery costs can directly disperse demand. A shrinking delivery radius can reduce the visibility of service providers to users. Figure 1 An example of reducing the delivery radius for a request to search for on-demand services according to conventional techniques is shown. Figure 1 As shown, during peak hours, the delivery radius may be reduced from radius 2 (R2) to radius 1 (R1) to curb demand by showing fewer service providers to the user. Figure 1As shown in (A), when the delivery radius is radius 2 (R2), the first to seventh service providers (M1 to M7) are displayed to the user. However, when the delivery radius is reduced to radius 1 (R1), only some service providers (M1 and M2) are displayed to the user. Even though reducing the delivery radius can help the demand fulfillment rate to a certain level, it may affect the user experience because the quality of the service providers within the reduced delivery radius may not meet the user's expectations. For example, the service providers within the reduced delivery radius may not have a good historical delivery experience value. In addition, when the user wants to order food, the user may only observe very few service providers or even no service providers. For example, as Figure 1 As shown in (B), when the delivery radius is radius 2 (R2), the third to seventh service providers (M3 to M7) are displayed to the user. However, when the delivery radius is reduced to radius 1 (R1), the service provider is not displayed to the user. Therefore, the user may not order food from the current on-demand service platform and then switch to another on-demand service platform.

[0006] Therefore, there is a need to provide a solution for processing requests for searching on-demand services without suppressing the demand. Summary of the invention

[0007] According to various embodiments, there is a server for processing a request for searching for on-demand services. The server includes: a communication interface configured to receive a request for searching for on-demand services from a computing device; and a processor, which can be connected to the communication interface to receive the request for searching, and is configured to determine the distance from the location of the computing device based on the confirmed allocation rate of multiple requests for on-demand services received from multiple users to generate a first service provider list within the distance. The processor is configured to: rank the first service provider list based on multiple predetermined first factor values, the multiple predetermined first factor values ​​including the predicted delivery experience value of each service provider in the first service provider list; generate a second service provider list from the first service provider list based on the ranking of each service provider in the first service provider list, and determine whether to filter out a portion of the second service provider list to generate a third service provider list based on the size of the second service provider list. If the processor determines that the portion of the second service provider list is to be filtered out, the processor is configured to filter out a portion of the second service provider list based on the predicted delivery experience value of each service provider in the second service provider list to generate a third service provider list, and provide the third service provider list as a search result to the computing device through the communication interface.

[0008] In some implementations, the processor is configured to filter out a portion of the first service provider list based on a ranking of each service provider in the first service provider list to generate the second service provider list.

[0009] In some implementations, the processor is configured to rank the second service provider list based on a plurality of predetermined second factor values, and generate a third service provider list from the second service provider list based on the ranking of each service provider in the second service provider list.

[0010] In some implementations, if the size of the second service provider list is less than a predetermined size, the processor is configured to determine not to filter out the portion of the second service provider list to generate the third service provider list.

[0011] In some embodiments, if the size of the second service provider list is the same as or greater than a predetermined size, the processor is configured to determine to filter out the portion of the second service provider list based on a ranking of each service provider in the second service provider list to generate a third service provider list.

[0012] In some implementations, if the size of the second service provider list is the same as or greater than a predetermined size, the processor is configured to filter out at least one service provider whose predicted delivery experience value is lower than a predetermined minimum threshold of the predicted delivery experience value.

[0013] In some implementations, if the size of the second service provider list is the same as or greater than a predetermined size, the processor is configured not to filter out at least one service provider whose predicted delivery experience value is higher than a predetermined maximum threshold of the predicted delivery experience value.

[0014] In some implementations, the plurality of predetermined first factor values ​​further includes at least one of revenue, profit, type, and search matching score for each service provider in the first list of service providers.

[0015] In some implementations, the plurality of predetermined second factor values ​​include at least one of surge, estimated time of arrival, revenue, profit, and type of each service provider in the second service provider list.

[0016] According to various embodiments, there is a method for processing a request for searching for on-demand services, the method comprising: receiving a request for searching for on-demand services from a computing device; determining a distance from a location of the computing device based on a confirmed allocation rate of a plurality of requests for on-demand services received from a plurality of users to generate a first list of service providers within the distance; ranking the first list of service providers based on a plurality of predetermined first factor values, the plurality of predetermined first factor values ​​comprising a predicted delivery experience value of each service provider in the first list of service providers; generating a second list of service providers from the first list of service providers based on the ranking of each service provider in the first list of service providers; determining whether to filter out a portion of the second list of service providers to generate a third list of service providers based on a size of the second list of service providers; and if it is determined to filter out the portion of the second list of service providers, filtering out the portion of the second list of service providers based on the predicted delivery experience value of each service provider in the second list of service providers to generate a third list of service providers; and providing the third list of service providers as a search result to the computing device.

[0017] In some embodiments, the method further includes filtering out a portion of the first service provider list based on a ranking of each service provider in the first service provider list to generate a second service provider list.

[0018] In some embodiments, the method further includes: ranking the second service provider list based on a plurality of predetermined second factor values; and generating a third service provider list from the second service provider list based on the ranking of each service provider in the second service provider list.

[0019] In some embodiments, the method further includes: if the size of the second service provider list is smaller than a predetermined size, determining not to filter out the portion of the second service provider list to generate the third service provider list.

[0020] In some embodiments, the method further includes: if the size of the second service provider list is the same as or greater than a predetermined size, determining based on a ranking of each service provider in the second service provider list to filter out the portion of the second service provider list to generate a third service provider list.

[0021] In some implementations, if the size of the second service provider list is the same as or larger than a predetermined size, at least one service provider whose predicted delivery experience value is lower than a predetermined minimum threshold of the predicted delivery experience value is filtered out.

[0022] In some implementations, if the size of the second service provider list is the same as or greater than a predetermined size, at least one service provider whose predicted delivery experience value is greater than a predetermined maximum threshold of the predicted delivery experience value is not filtered out.

[0023] In some implementations, the plurality of predetermined first factor values ​​further includes at least one of revenue, profit, type, and search matching score for each service provider in the first list of service providers.

[0024] In some implementations, the plurality of predetermined second factor values ​​include at least one of surge, estimated time of arrival, revenue, profit, and type of each service provider in the second service provider list.

[0025] According to various embodiments, a data processing device is provided, wherein the data processing device is configured to execute the method in any one of the above embodiments.

[0026] According to various embodiments, a computer program element is provided. The computer program element includes program instructions. When the program instructions are executed by one or more processors, the one or more processors are caused to perform the method in any one of the above embodiments.

[0027] According to various embodiments, a computer readable medium is provided, the computer readable medium including program instructions, which, when executed by one or more processors, cause the one or more processors to perform the method in any of the above embodiments. The computer readable medium may include a non-transitory computer readable medium. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The invention will be better understood when referring to the detailed description, taken in conjunction with the non-limiting examples and accompanying drawings, in which:

[0029] - Figure 1 An example of reducing the delivery radius for a request to search for on-demand services according to conventional techniques is shown.

[0030] - Figure 2 An infrastructure of a system including a server for processing requests for searching on-demand services is shown according to various embodiments.

[0031] - Figure 3 A block diagram of a server for processing a request to search for on-demand services is shown in accordance with various embodiments.

[0032] - Figure 4 A flow chart of a method for processing a request to search for on-demand services is shown in accordance with various embodiments.

[0033] - Figure 5A block diagram of a server for processing a request to search for on-demand services is shown in accordance with various embodiments.

[0034] - Figure 6 A block diagram of a server for processing a request to search for on-demand services is shown in accordance with various embodiments. DETAILED DESCRIPTION

[0035] The following detailed description refers to the accompanying drawings, in which specific details and embodiments are shown by way of example, in which the present disclosure may be implemented. The description of these embodiments is detailed enough to enable those skilled in the art to implement the present disclosure. Other embodiments may be utilized and structural and logical changes may be made without departing from the scope of the present disclosure. Various embodiments are not necessarily mutually exclusive, as some embodiments may be combined with one or more other embodiments to form new embodiments.

[0036] Embodiments described in the context of one system and method are also applicable to another server and method. Likewise, embodiments described in the context of a server are also applicable to the method, and vice versa.

[0037] Features described in the context of one embodiment may accordingly apply equally to the same or similar features in other embodiments. Features described in the context of one embodiment may accordingly apply equally to other embodiments, even if not explicitly described in these other embodiments. Furthermore, additions and / or combinations and / or substitutions as described for features in the context of an embodiment may accordingly apply equally to the same or similar features in other embodiments.

[0038] In the context of various embodiments, the articles “a,” “an,” and “the” as used with respect to features or elements include reference to one or more of these features or elements.

[0039] As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0040] Throughout this specification, the term "module" may be understood as an application specific integrated circuit (ASIC), an electronic circuit, a combinational logic circuit, a field programmable gate array (FPGA), a processor that executes code, other suitable hardware components that provide the described functionality, or any combination thereof. The term "module" may include a memory that stores code executed by a processor.

[0041] Hereinafter, embodiments will be described in detail.

[0042] Figure 2The infrastructure for a system 200 including a server 100 for processing requests for searching on-demand services is shown according to various embodiments.

[0043] like Figure 2 As shown, the system 200 may include, but is not limited to, a server 100 , a database system 140 , a computing device 161 , one or more external devices 170 , 180 , and a network 190 .

[0044] In some embodiments, an on-demand service may be a service that allows user 160 to satisfy user needs by immediately accessing goods and / or services. User 160 may use a user interface presented on computing device 161 to request a search for an on-demand service, such as a delivery service or a transportation service.

[0045] In some embodiments, the network 190 may include, but is not limited to, a local area network (LAN), a wide area network (WAN), a global area network (GAN), or any combination thereof. The network 190 may provide wired communication, wireless communication, or a combination of wired and wireless communication between the server 100 and the computing device 161 and between the server 100 and one or more external devices 170, 180 (e.g., service provider device 170 and service contractor device 180).

[0046] In some embodiments, computing device 161 may be connected to server 100 via network 190. In some embodiments, computing device 161 may be configured to communicate data or signals with server 100 via network 190. In some embodiments, computing device 161 may include, but is not limited to, at least one of a mobile phone, a tablet computer, a laptop computer, a desktop computer, a head-mounted display, and a smartwatch. In some embodiments, computing device 161 may belong to user 160. Although not shown, in some embodiments, system 200 may also include multiple computing devices, each of which belongs to multiple users. In some embodiments, user 160 and multiple users may be customers requesting on-demand services (e.g., food delivery).

[0047] In some embodiments, computing device 161 may include a location sensor. In some embodiments, the location sensor may communicate with at least one of a global positioning satellite (GPS) server, a network server, and a Wi-Fi server to detect the location of computing device 161. In some embodiments, computing device 161 may generate information about the location of computing device 161.

[0048] In some embodiments, the server 100, for example implemented by a server computer, may include a communication interface 110, a processor 120, and a memory 130 (as will be seen in Figure 3 described).

[0049] In some implementations, server 100 may communicate with computing device 161 via network 190. In some implementations, computing device 161 may receive a request to search for on-demand services from user 160. Computing device 161 may send the request to server 100 via network 190. In some implementations, computing device 161 may send information about the location of computing device 161 to server 100 via network 190.

[0050] In some embodiments, server 100 may receive a request with information about the location of computing device 161. Server 100 may then determine the distance from the location of computing device 161 (also referred to as a "delivery radius") to generate a list of service providers within that distance. In some embodiments, server 100 may aggregate the locations of available service providers (e.g., restaurants), the types of available services (e.g., food), estimated costs, and other information, and provide the aggregated information to computing device 161 over network 190 so that user 160 can make a selection on a user interface presented on computing device 161.

[0051] In some embodiments, the system 200 may further include a database 150. In some embodiments, the database 150 may be part of a database system 140, wherein the database system may be external to the server 100. The server 100 may communicate with the database 150. In some other embodiments, although not shown, the database 150 may be implemented locally in the memory 130 of the server 100.

[0052] In some embodiments, the server 100 may communicate with an external device 170 (e.g., a service provider device 170) via a network 190. The service provider device 170 may belong to a service provider. Service providers may include, but are not limited to, restaurants, cafes, shops, and merchants that can provide goods and / or services (e.g., food). In some embodiments, the server 100 may receive a request with information about the location of the computing device 161, and then determine a delivery radius to generate a list of service providers within the delivery radius. In some embodiments, the server 100 may communicate with the service provider device 170 to check the availability of the service provider. In some embodiments, the server 100 may communicate with the service provider device 170 to aggregate information (the information includes, but is not limited to, the location of the service provider, the type of available service, the estimated cost, and other information) to generate a list of service providers within the delivery radius. In some embodiments, the computing device 161 may display a list of service providers with aggregated information on a user interface. In some embodiments, after the user 160 makes a selection on the user interface for a request for an on-demand service, for example, by selecting a service provider and food, the server 100 may communicate with the selected service provider (e.g., service provider device 170) to prepare the selected food.

[0053] In some embodiments, the server 100 may communicate with an external device 180 (e.g., a service contractor device 180) via a network 190. The service contractor device 180 may belong to a service contractor. The service contractor may include, but is not limited to, a driver capable of providing a delivery service from a first location (e.g., the location of the selected service provider) to a second location (e.g., the location of the user 160). In some embodiments, after the user 160 makes a selection for a request for an on-demand service, such as by selecting a service provider and food using a computing device 161, the server 100 may assign a service contractor from among a plurality of service contractors to deliver the selected food from the selected service provider to the user 160.

[0054] Figure 3 A block diagram of a server 100 for processing a request for searching for on-demand services is shown according to various embodiments.

[0055] like Figure 3 As shown, the server 100 (which is implemented by a server computer, for example) may include a communication interface 110 , a processor 120 , and a memory 130 .

[0056] In some embodiments, the memory 130 (also referred to as a "database") can temporarily or permanently store input data and / or output data. In some embodiments, the memory 130 can store data that allows the server 100 to perform the method 300 (as will be seen in Figure 4 In some embodiments, the program code may be embedded in a software development kit (SDK). The memory 130 may include an internal memory and / or an external memory of the server 100. The external memory may include, but is not limited to, an external storage medium, such as a memory card, a flash drive, and a network storage.

[0057] In some embodiments, the communication interface 110 may allow one or more computing devices (including the computing device 161) to communicate with the processor 120 of the server 100 via the network 190, such as Figure 2 In some embodiments, Figure 2 As shown, computing device 161 may belong to user 160 who wants to request an on-demand service. In some implementations, communication interface 110 may transmit signals to computing device 161 and / or receive signals from computing device 161 via network 190.

[0058] In some embodiments, the communication interface 110 may allow one or more external devices 170, 180 (eg, a service provider device 170 and a service contractor device 180) to communicate with the processor 120 of the server 100 via the network 190, such as Figure 2 In some implementations, the communication interface 110 may send signals to and / or receive signals from the external devices 170 and 180 via the network 190 .

[0059] In some implementations, the communication interface 110 may receive a request to search for on-demand services from the computing device 161 through the network 190. The communication interface 110 may then send the request to search for on-demand services to the processor 120.

[0060] In some implementations, communication interface 110 may also receive information about the location of computing device 161 from computing device 161 through network 190. Communication interface 110 may then send information about the location of computing device 161 to processor 120.

[0061] In some implementations, communication interface 110 may simultaneously receive the request for search and information about the location of computing device 161 from computing device 161 over the network. In some other implementations, communication interface 110 may first receive the request for search over network 190, and then receive information about the location of computing device 161 from computing device 161.

[0062] In some embodiments, the communication interface 110 may communicate with a service provider device 170 of a service provider to check the availability of the service provider and collect information that may be required to generate a list of service providers. In some embodiments, the communication interface 110 may communicate with the service provider device 170 to provide search results according to a user's request for searching for on-demand services. In some embodiments, the communication interface 110 may communicate with the service provider device 170 to provide on-demand services according to a subsequent request by the user for a certain on-demand service. In some embodiments, the communication interface 110 may communicate with the service contractor device 180 to deliver / transport on-demand services, such as selected food, from the selected service provider to the user 160.

[0063] The processor 120 may include, but is not limited to, a microprocessor, an analog circuit, a digital circuit, a mixed signal circuit, a logic circuit, an integrated circuit, a central processing unit (CPU), a graphics processing unit (GPU), a digital signal processor (DSP), a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), or any combination thereof. Any other kind of implementation of the various functions described in further detail below may also be understood as the processor 120.

[0064] In some embodiments, the processor 120 may be connected to the communication interface 110. In some embodiments, the processor 120 may be configured to communicate data or signals with the communication interface 110 to receive the request for the search and the information about the location of the computing device 161.

[0065] In some implementations, processor 120 may simultaneously receive the request for search and information about the location of computing device 161 from communication interface 110. In some other implementations, processor 120 may first receive the request for search from communication interface 110 and then receive information about the location of computing device 161.

[0066] In some embodiments, the processor 120 may determine the distance from the location of the computing device (also referred to as a "delivery radius") based on real-time signals about the market (e.g., a confirmed allocation rate (also referred to as a "CAR") of multiple requests for on-demand services (also referred to as "orders") received from multiple users in the area where the user 160 is located) to generate a first list of service providers within the delivery radius. In some embodiments, the processor 120 may calculate the confirmed allocation rate of multiple orders received from multiple users in the area where the user 160 is located during a certain period of time. In some embodiments, the processor 120 may obtain the number of allocated orders and the number of previously uncompleted orders from an external server (not shown). Then, the processor 120 may obtain the confirmed allocation rate by dividing the number of allocated orders by the sum of the number of allocated orders and the number of previously uncompleted orders. In some other embodiments, the processor 120 may obtain the confirmed allocation rate of multiple orders received from multiple users in the area where the user 160 is located during a certain period of time from an external server (not shown).

[0067] In some embodiments, the processor 120 may determine the delivery radius based on the confirmed allocation rate. For example, if the market has limited supply and high demand, the confirmed allocation rate may be reduced. If the confirmed allocation rate decreases, the processor 120 may reduce the delivery radius to reduce the delivery range. As another example, if the confirmed allocation rate increases, the processor 120 may increase the delivery radius to increase the delivery range.

[0068] In some embodiments, the processor 120 may determine the delivery radius based on the confirmed allocation rate and at least one optimization parameter. In some embodiments, the processor 120 may use parameters to control the relationship between the delivery radius and the confirmed allocation rate. In other words, the processor 120 may determine the delivery radius based on the parameter and the confirmed allocation rate. In some embodiments, the processor 120 may select at least one parameter from these parameters and optimize the selected parameter based on historical performance. For example, the processor 120 may optimize on the minimum radius configuration within the automatic radius management system (ARM). The processor 120 may track performance metrics (e.g., FedX, conversion rate, revenue, etc.) at different minimum radii and apply Bayesian optimization to adjust the minimum radius based on the performance metric. Then, the processor 120 may use the optimization parameters to determine the delivery radius.

[0069] In some other embodiments, the processor 120 may determine the delivery radius based on the confirmed allocation rate and the scaler. The processor 120 may use the scaler to control the delivery radius. In some embodiments, the processor 120 may optimize the scaler and determine the delivery radius using the optimized scaler and other parameters in the lever (as described below). The scaler may be another type of control variable applied in the automatic radius management system. For example, after the automatic radius management system calculates the final radius, the scaler may be multiplied by the final radius to provide an enlarged radius or a reduced radius (based on the scaler value), which may be adjusted by Optimus, similar to the minimum radius described above.

[0070] In some embodiments, the processor 120 may rank the first service provider list based on a plurality of predetermined first factor values ​​(also referred to as a "default ranking" or "initial ranking"), wherein the plurality of predetermined first factor values ​​include a predicted delivery experience value (also referred to as "FedX") of each service provider in the first service provider list.

[0071] In some embodiments, the processor 120 may generate a second service provider list from the first service provider list based on the ranking of each service provider in the first service provider list. In some embodiments, the processor 120 may filter out a portion of the first service provider list based on the ranking of each service provider in the first service provider list to generate the second service provider list. For example, based on the ranking of each service provider in the first service provider list, the top N (e.g., N=100) service providers may be passed to the next round of ranking. In some embodiments, in the case where the user 160 may not be interested in all the top N service providers, service providers ranked between N and 2N may also be selected to be displayed to the user 160.

[0072] In some embodiments, the predicted delivery experience value of each service provider may be the percentage of unique requests that are completed at the first order and delivered within a predefined time threshold and comply with the shopping cart (basket) ETA (estimated arrival time) commitment. The shopping cart ETA may refer to the estimated time from when the order is created to when the order is completed and delivered to the user (customer). In some embodiments, the delivery experience value of each service provider may be predicted based on the historical delivery experience value, ETA performance, and real-time signals (such as confirmed allocation rate) of each service provider. In some embodiments, the processor 120 may use the FedX prediction model to predict the delivery experience value of each service provider in the first service provider list. In some embodiments, multiple predetermined first factor values ​​may also include, but are not limited to, at least one of the revenue, profit, type, and search matching score of each service provider in the first service provider list. For example, the type of each service provider may include a merchant type.

[0073] Since the predicted delivery experience value of each service provider is used as a feature in the default ranking process, service providers with high predicted delivery experience values ​​can be ranked higher and not filtered out. In addition, since multiple predetermined first factor values ​​are used in the default ranking process, service providers with low predicted delivery experience values ​​may also be included in the second service provider list if other feature values ​​of these service providers are high.

[0074] In some embodiments, after obtaining the second service provider list and the ranking and / or score obtained from the default ranking process, the processor 120 may rank the service providers based on more features. In some embodiments, the processor 120 may rank the second service provider list based on a plurality of predetermined second factor values ​​(also referred to as "final ranking" or "sustainable ranking"), and generate a third service provider list from the second service provider list. In some embodiments, the plurality of predetermined second factor values ​​include at least one of surge, estimated arrival time, revenue, profit, and type of each service provider in the second service provider list.

[0075] In some embodiments, the processor 120 may use the detailed FedX prediction model to filter the service providers based on the predicted delivery experience value of each service provider in the second service provider list, thereby generating a third service provider list. In some embodiments, in the default ranking process, the FedX prediction model may consider service provider (merchant) level features, and in the final ranking process, the FedX prediction model may consider service provider (merchant) level features and user (customer) level features (similar to the historical FedX in customer positions). The processor 120 is able to find service providers with high predicted delivery experience values ​​to display these service providers to the user 160. When the processor 120 filters out service providers with low predicted delivery experience values ​​to generate the third service provider list instead of lowering the ranking of these service providers, the service providers with high predicted delivery experience values ​​may be included in the third service provider list as search results. In this way, the user 160 can select a service provider with a high predicted delivery experience value, and may not visit a service provider with a low predicted delivery experience value, and thus the user experience may be improved.

[0076] In some embodiments, the processor 120 may determine whether to filter out the portion of the second service provider list based on the size of the second service provider list to generate a third service provider list as a search result. In some embodiments, if the processor 120 determines to filter out the portion of the second service provider list, the processor 120 may filter out the portion of the second service provider list based on the predicted delivery experience value of each service provider in the second service provider list to generate a third service provider list.

[0077] In some implementations, the processor 120 may rank the second service provider list based on a plurality of predetermined second factor values, and further generate a third service provider list from the second service provider list based on the ranking of each service provider in the second service provider list.

[0078] In some embodiments, if the size of the second service provider list is less than a predetermined size, the processor 120 may determine not to filter out the portion of the second service provider list to generate the third service provider list. For example, if the size of the third service provider list is too small, it may affect the conversion rate of the search, and there may not be enough options provided to the user 160.

[0079] In some embodiments, if the size of the second service provider list is equal to or greater than the predetermined size, the processor 120 may determine to filter out the portion of the second service provider list based on the ranking of each service provider in the second service provider list to generate a third service provider list. For example, the processor 120 may filter out the last M percentile service providers (e.g., "M" is a parameter to be optimized) so that the service providers in the third service provider list have a higher predicted delivery experience value, and the conversion rate of the search will not be affected too much.

[0080] In some embodiments, if the size of the second service provider list is equal to or greater than a predetermined size, the processor 120 may filter out at least one service provider whose predicted delivery experience value is lower than a predetermined minimum threshold of the predicted delivery experience value. If most of the service providers in the second service provider list have low predicted delivery experience values ​​during low supply conditions (e.g., heavy rain, rush hour, etc.), there may be a situation where the service provider with the low predicted delivery experience value may still be in the third service provider list. When the processor 120 uses the predetermined minimum threshold of the predicted delivery experience value to avoid this situation, the service provider included in the third service provider list may not have a predicted delivery experience value lower than the predetermined minimum threshold of the predicted delivery experience value.

[0081] In some embodiments, if the size of the second service provider list is equal to or greater than a predetermined size, the processor 120 may not filter out at least one service provider whose predicted delivery experience value is higher than a predetermined maximum threshold of the predicted delivery experience value. For example, when most service providers have high predicted delivery experience values, the processor 120 may determine that these service providers do not need to be filtered out because these service providers may not have a negative impact on the user experience.

[0082] In some implementations, the processor 120 may use the following mathematical equation: filtering = relu (fedx score -threshold(percentile,N merchants ,min fedx ,max fedx )), where N merchants is the size of the second service provider list, min fedx and max fedx It is a configuration defined for each region (e.g., city) as a predetermined minimum threshold and a predetermined maximum threshold of the predicted delivery experience value. scoreis the delivery experience value (score) predicted based on the FedX prediction model, the percentile is the position of the delivery experience value (score) obtained in the second service provider list, and the threshold is the threshold of the predicted delivery experience value used for filtering. If the size of the second service provider list is higher than the predetermined maximum threshold of the predicted delivery experience value, the processor 120 may not apply filtering. The relu function may check the fedx score Is it higher than the threshold? If fedx score If fedx is above the threshold, "1" can be returned and the final ranking process can be operated based on other features. score If the value is equal to or lower than the threshold, the result with the value "0" may be filtered out, and the service provider may not be displayed to the user 160. An example of the algorithm of the relu function is as follows:

[0083] Check if len(second service provider list) > N merchants

[0084] If not, return 1

[0085] o If yes, get the FedX value at the percentile (threshold) in the second service provider list

[0086] Threshold = max(threshold,min(threshold,max fedx ),min fedx )

[0087] Return threshold

[0088] In some embodiments, surge (which is one of a plurality of predetermined second factor values) can control the delivery fee for on-demand service (order) requests based on real-time market conditions to shape demand. For example, if the supply of service contractors (drivers) is limited and the demand is high, the surge may increase to increase the delivery fee and reduce the user's demand. In some embodiments, the surge signal can be passed to the final ranking process and optimized along with other features (such as predicted delivery experience value, profit, revenue, etc.). Using the surge signal, the processor 120 can take the surge into account to rank the second service provider list. As a result, for example, service providers with low surge can be ranked higher, and this can lead to an increase in the conversion rate of the search.

[0089] In some embodiments, ETA padding can be another lever to consider to shape demand. Since the predicted ETA has some inaccuracy, users may prefer to have their orders delivered on time or slightly earlier than the predicted ETA. In some embodiments, the processor 120 can add a value (referred to as a "padding value") to the estimated ETA to adjust the predicted ETA to improve user satisfaction. The padding value added to the predicted ETA can be adjusted with other levers to provide improved demand shaping.

[0090] In some embodiments, the processor 120 may provide the third service provider list to the computing device 161 as a search result. In some embodiments, the third service provider list may be provided to the computing device 161 as a search result together with the top N service providers (ranked between N and 2N) that were not passed to the final ranking process based on the final ranking score. For example, a total of 200 service providers are listed to the user, and 190 service providers are passed from the default ranking process to the final ranking process and 10 service providers are filtered out during the final ranking process. If there are 180 service providers passed from the final ranking process, 20 service providers (not passed to the final ranking process) may be selected from the remaining service providers in the default ranking process. Therefore, the service providers displayed to the user may include 180 service providers ranked from 1 to 190 (10 service providers not filtered out) and 20 service providers ranked from 191 to 210.

[0091] As described above, in order to solve the above problems, the server 100 according to various embodiments can control the visibility of service providers based on more functions (such as revenue, profit, conversion rate of search, popularity, predicted delivery experience value, etc.), rather than using only radius management to control visibility based entirely on delivery distance and delivery cost surges. The server 100 according to various embodiments can consider the predicted delivery experience value and optimize multiple levers together. In this way, the server 100 according to various embodiments can control the visibility of service providers and thus shape demand in the best way.

[0092] Figure 4 A flow chart of a method 300 for processing a request for searching for on-demand services according to various embodiments is shown. According to various embodiments, a method 300 for processing a request for searching for on-demand services may be provided.

[0093] In some implementations, method 300 may include step 301: receiving a request to search for on-demand services from a computing device.

[0094] In some implementations, method 300 may include step 302 of determining a distance from a location of a computing device based on confirmed allocation rates of a plurality of requests for on-demand services received from a plurality of users to generate a first list of service providers within the distance.

[0095] In some implementations, the method 300 may include step 303: ranking the first service provider list based on a plurality of predetermined first factor values, the plurality of predetermined first factor values ​​including a predicted delivery experience value of each service provider in the first service provider list.

[0096] In some implementations, the method 300 may include step 304 of generating a second service provider list from the first service provider list based on a ranking of each service provider in the first service provider list.

[0097] In some implementations, the method 300 may include step 305: determining whether to filter out a portion of the second service provider list to generate a third service provider list based on the size of the second service provider list.

[0098] In some embodiments, method 300 may include step 306: if it is determined to filter out the portion of the second service provider list, filtering out the portion of the second service provider list based on the predicted delivery experience value of each service provider in the second service provider list to generate a third service provider list.

[0099] In some implementations, the method 300 may include step 307 : providing the third service provider list as a search result to the computing device.

[0100] Figure 5 A block diagram of a server 400 for processing requests for searching on-demand services is shown according to various embodiments.

[0101] In some embodiments, the server 400 may include an optimization framework 400a and a general framework 400b. The optimization framework 400a may support multi-objective optimization to optimize multiple levers together within one optimization system and find appropriate parameters for them. The optimization framework 400a may be referred to as Optimus 400a, and the general framework 400b may be referred to as a separate system 400b.

[0102] In some embodiments, as described above, the multiple levers may include, but are not limited to, the following five levers: Automatic Radius Management (also referred to as "ARM"), Default Ranking, Final Ranking, Surge, and ETA Filling. In some embodiments, there may be unknown parameters that need to be optimized. For example, these parameters may include, but are not limited to: percentile and N in the final ranking process merchants, feature weights in the default ranking process and the final ranking process, delivery radius parameters in ARM, surge signals, and ETA filling parameters. In some embodiments, there may be parameters that exist in conventional systems, such as the weights of all features such as the type of each service provider, revenue, profit, matching score, etc. For example, the type of each service provider may include the merchant type.

[0103] like Figure 5 As shown, in some embodiments, the server 400 may include a model library 401 for score generation. The model may be an online service model for calculating ranking scores to rank service providers. The model may include, but is not limited to, FedX scores, ETA scores, etc. In some embodiments, the server 400 may include a metric evaluation module 402. The metric evaluation module 402 may be a service pipeline that calculates historical performance metrics based on stored historical data and configurations.

[0104] like Figure 5 As shown, in some embodiments, the optimization framework 400a may include, but is not limited to, a combiner 403 (also referred to as an "optimus combiner"), an optimization library 405 (also referred to as an "optimizer library"), and an adaptive rollout module 404. In some embodiments, the combiner 403 may combine necessary features and configurations to update the weights and parameters of the levers. In some embodiments, the optimization library 405 may have optimization logic, such as a Bayesian optimizer, etc. In some embodiments, the adaptive rollout module 404 may analyze the historical performance of each configuration based on historical data and configurations, and apply the optimization parameters to different layers through adaptive rollout. The performance results may then be saved for future optimization use.

[0105] like Figure 5 As shown, in some embodiments, the general framework 400b may include an online sustainable ranking module 406. The online sustainable ranking module 406 may be one of the services logged in to the Optimus platform (the platform that performs the final ranking process). The online sustainable ranking module 406 may rank service providers based on FedX performance, revenue, profit, popularity, etc.

[0106] In some embodiments, the general framework 400b may include a dynamic radius management module 407. The dynamic radius management module 407 may be another service logged on the Optimus platform. The dynamic radius management module 407 may be an automatic radius management service and is used to control the delivery radius.

[0107] In some implementations, the general framework 400b may include an ETA filling module 408. The ETA filling module 408 may be a service logged into the Optimus platform. The ETA filling module 408 may be used to control the estimated delivery time (ETA) displayed on the user's software application (app).

[0108] In some embodiments, the general framework 400b may include a two-sided pricing module 409. The two-sided pricing module 409 may be another service that is logged into the Optimus platform. The two-sided pricing module 409 may be used to control the delivery fees displayed on the user's software application and the delivery fees earned by the service provider, and to manage the delivery fees based on different supply-demand conditions.

[0109] In some embodiments, the universal framework 400b may include additional individual system modules 410 , 411 .

[0110] Figure 6 A block diagram of a server 500 for processing requests for searching on-demand services is shown in accordance with various embodiments.

[0111] In some embodiments, there is a Presto database 501, which is a Hive database that stores historical data, including but not limited to: user behavior data, browsing data, click data, checkout data, and order data. The order data may include whether the order is fulfilled, the time when the order is completed, etc.

[0112] In some implementations, there is an S3 storage system 502 that stores all configurations used in the system. Historical configurations can be retrieved from the S3 storage system 502. The configurations and historical data can be used to generate performance metrics.

[0113] In some embodiments, the optimization framework 400a (also known as "Optimus") can control the visibility of service providers. In some embodiments, the ARM (also known as "Automatic Radius Management") module 503 can calculate the delivery radius for the search based on external signals including CAR (also known as "Confirmed Allocation Rate"), and save the delivery radius data 504 to the database. In some embodiments, the ES (also known as "Elastic Search") module 505 can read the delivery radius and apply it to the search service. In some embodiments, after the search, the list of service providers is returned and passed to the default ranking module 506. The default ranking module 506 can rank the service providers based on the list of features including FedX, and only the top N (e.g., N=100) merchants can be passed to the final ranking module 507. In some embodiments, in the final ranking module 507, the filtering logic as described above can be applied, and service providers with high FedX performance are retained. External signals including ETA fill signals and surge signals can be applied to the ranking of service providers.

[0114] Although the present disclosure has been specifically shown and described with reference to specific embodiments, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the present invention as defined in the appended claims. Therefore, the scope of the present invention is indicated by the appended claims, and it is therefore intended to cover all changes within the meaning and range of equivalents of the claims.

Claims

1. A server for processing a request for searching for an on-demand service, the server comprising: a communication interface configured to receive the request to search for the on-demand service from a computing device; as well as a processor connectable to the communication interface to receive the request for the search, and configured to determine a distance from the location of the computing device based on a confirmed allocation rate of a plurality of requests for on-demand services received from a plurality of users to generate a first list of service providers within the distance, The processor is configured to rank the first service provider list based on a plurality of predetermined first factor values, generate a second service provider list from the first service provider list based on the ranking of each service provider in the first service provider list, and determine whether to filter out a portion of the second service provider list to generate a third service provider list based on the size of the second service provider list, wherein the plurality of predetermined first factor values ​​include a predicted delivery experience value of each service provider in the first service provider list, and If the processor determines to filter out the portion of the second service provider list, the processor is configured to filter out the portion of the second service provider list based on the predicted delivery experience value of each service provider in the second service provider list to generate the third service provider list, and provide the third service provider list as a search result to the computing device via the communication interface.

2. The server according to claim 1, wherein: The processor is configured to filter out a portion of the first service provider list based on a ranking of each service provider in the first service provider list to generate the second service provider list.

3. The server according to claim 1 or 2, wherein: The processor is configured to rank the second service provider list based on a plurality of predetermined second factor values, and generate the third service provider list from the second service provider list based on the ranking of each service provider in the second service provider list.

4. The server according to claim 3, wherein: If the size of the second service provider list is smaller than a predetermined size, the processor is configured to determine not to filter out the portion of the second service provider list to generate the third service provider list.

5. The server according to claim 3, wherein: If the size of the second service provider list is equal to or greater than the predetermined size, the processor is configured to determine to filter out the portion of the second service provider list based on a ranking of each service provider in the second service provider list to generate the third service provider list.

6. The server according to claim 5, wherein: If the size of the second service provider list is equal to or greater than the predetermined size, the processor is configured to filter out at least one service provider having a predicted delivery experience value lower than a predetermined minimum threshold of the predicted delivery experience value.

7. The server according to claim 5, wherein: If the size of the second service provider list is equal to or greater than the predetermined size, the processor is configured to not filter out at least one service provider having a predicted delivery experience value higher than a predetermined maximum threshold of the predicted delivery experience value.

8. The server according to any one of claims 1 to 7, wherein: The plurality of predetermined first factor values ​​further include at least one of revenue, profit, type, and search matching score of each service provider in the first service provider list.

9. The server according to any one of claims 3 to 7, wherein: The plurality of predetermined second factor values ​​include at least one of surge, estimated arrival time, revenue, profit, and type of each service provider in the second service provider list.

10. A method for processing a request for searching for an on-demand service, the method comprising: receiving, from a computing device, the request to search for the on-demand service; determining a distance from a location of the computing device based on confirmed allocation rates of a plurality of requests for on-demand services received from a plurality of users to generate a first list of service providers within the distance, ranking the first service provider list based on a plurality of predetermined first factor values, wherein the plurality of predetermined first factor values ​​include a predicted delivery experience value of each service provider in the first service provider list; generating a second service provider list from the first service provider list based on the ranking of each service provider in the first service provider list; determining whether to filter out a portion of the second service provider list to generate a third service provider list based on the size of the second service provider list; If it is determined to filter out the portion of the second service provider list, filtering out the portion of the second service provider list based on the predicted delivery experience value of each service provider in the second service provider list to generate the third service provider list; and The third service provider list is provided to the computing device as a search result.

11. The method according to claim 10, further comprising: A portion of the first service provider list is filtered out based on the ranking of each service provider in the first service provider list to generate the second service provider list.

12. The method according to claim 10 or 11, further comprising: ranking the second list of service providers based on a plurality of predetermined second factor values; and The third service provider list is generated from the second service provider list based on a ranking of each service provider in the second service provider list.

13. The method according to claim 12, further comprising: If the size of the second service provider list is smaller than a predetermined size, it is determined not to filter out the portion of the second service provider list to generate the third service provider list.

14. The method according to claim 12, further comprising: If the size of the second service provider list is equal to or greater than the predetermined size, filtering out the portion of the second service provider list to generate the third service provider list is determined based on the ranking of each service provider in the second service provider list.

15. The method according to claim 14, wherein: If the size of the second service provider list is equal to or greater than the predetermined size, at least one service provider having a predicted delivery experience value lower than a predetermined minimum threshold of the predicted delivery experience value is filtered out.

16. The method according to claim 14, wherein: If the size of the second service provider list is equal to or greater than the predetermined size, at least one service provider having a predicted delivery experience value higher than a predetermined maximum threshold of the predicted delivery experience value is not filtered out.

17. The method according to any one of claims 10 to 16, wherein: The plurality of predetermined first factor values ​​further include at least one of revenue, profit, type, and search matching score of each service provider in the first service provider list.

18. The method according to any one of claims 12 to 16, wherein: The plurality of predetermined second factor values ​​include at least one of surge, estimated arrival time, revenue, profit, and type of each service provider in the second service provider list.

19. A data processing apparatus configured to perform a method according to any one of claims 10 to 18.

20. A computer readable medium comprising program instructions which, when executed by one or more processors, cause the one or more processors to perform the method according to any one of claims 10 to 18.

Citation Information

Cited By

  • Server and method for guiding delivery service provider to move to location for facilitating provision of on-demand service

    WO2026213799A1