Order scheduling method and computer storage medium

Abstract

This application provides an order scheduling method and a computer storage medium. The method includes receiving travel order information, obtaining service provider information and historical travel service information of at least two candidate service providers corresponding to the travel order information; determining expected revenue based on the travel order information, service provider information, and historical travel service information; and allocating orders to at least two candidate service providers based on the expected revenue to obtain an order scheduling result. Expected revenue represents the overall revenue, which includes not only the short-term revenue corresponding to the travel order but also the probability of candidate service providers offering travel services in the future, reflecting medium-term and long-term revenue. From the perspective of overall revenue, this increases the number of service providers offering travel services in the future predetermined time period, improving the travel experience.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to an order scheduling method and a computer storage medium. Background Technology

[0002] With the widespread adoption of smart devices and the development of mobile internet technology, ride-hailing services are becoming increasingly popular. Service requesters (e.g., passengers) can submit vehicle requests through online ride-hailing platforms when traveling. These platforms then allocate these requests to service providers (e.g., drivers) according to a specific strategy.

[0003] Different order dispatch methods will affect the service providers' subsequent service provision, impacting the number of available vehicles on online ride-hailing platforms. The number of available vehicles directly affects passenger travel experience and the platform's revenue. Order dispatch methods not only affect the overall dispatch efficiency of the platform but also its sustainable development. Therefore, there is an urgent need to provide an order dispatch method that enables online ride-hailing platforms to better manage order dispatch. Summary of the Invention

[0004] In view of this, embodiments of this application provide an order scheduling scheme to at least partially solve the above-mentioned problems.

[0005] According to a first aspect of the embodiments of this application, an order scheduling method is provided, comprising: receiving travel order information; obtaining service provider information and historical travel service information of at least two candidate service providers corresponding to the travel order information; determining expected revenue based on the travel order information, the service provider information and the historical travel service information; and allocating orders to at least two of the candidate service providers based on the expected revenue to obtain an order scheduling result.

[0006] According to a second aspect of the embodiments of this application, another order scheduling method is provided, comprising: receiving travel order information sent by a service requester through an order scheduling platform, wherein the order scheduling platform is an aggregation platform that aggregates multiple travel platforms; sending the travel order information to the multiple travel platforms and receiving feedback information from the multiple travel platforms, wherein the feedback information carries service provider information and historical travel service information of at least two candidate service providers corresponding to the travel order information; determining expected revenue based on the travel order information, the service provider information, and the historical travel service information; and allocating orders to at least two of the candidate service providers based on the expected revenue to obtain an order scheduling result.

[0007] According to a third aspect of the embodiments of this application, another order scheduling method is provided, comprising: sending current location information of a service provider terminal and current service status information of the service provider terminal to a travel platform, such that the travel platform determines at least two candidate service providers based on multiple current location information, multiple current service status information, and travel order information received from the order scheduling platform, generates feedback information, and sends the feedback information to the order scheduling platform; the order scheduling platform determines expected revenue based on the travel order information, service provider information of the at least two candidate service providers carried in the feedback information, and historical travel service information, and allocates orders to the at least two candidate service providers based on the expected revenue, thereby obtaining an order scheduling result, wherein the order scheduling result represents the service provider matching each of the travel order information; and receiving the matched travel order information sent by the order scheduling platform through the travel platform corresponding to the matched service provider, and the expected revenue corresponding to the matched travel order information.

[0008] According to a fourth aspect of the embodiments of this application, another order scheduling method is provided, comprising: receiving a travel request initiated by a service requester, and generating travel order information based on the travel request; sending the travel order information to an order scheduling platform, such that the order scheduling platform sends the travel order information to multiple travel platforms, and determining expected revenue based on service provider information and historical travel service information of at least two candidate service providers fed back by the multiple travel platforms, combined with the travel order information, allocating orders to at least two of the candidate service providers according to the expected revenue, thereby obtaining an order scheduling result; and receiving the service provider information allocated in the order scheduling result sent by the order scheduling platform.

[0009] According to a fifth aspect of the present application, an electronic device is provided, comprising: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other via the communication bus; the memory is used to store at least one executable instruction, wherein the executable instruction causes the processor to perform an operation corresponding to any of the order scheduling methods described in the first to fourth aspects.

[0010] According to a sixth aspect of the embodiments of this application, a computer storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the order scheduling method as described in any one of the first to fourth aspects.

[0011] According to the order scheduling scheme provided in this application embodiment, travel order information is received, and service provider information and historical travel service information of at least two candidate service providers corresponding to the travel order information are obtained. Based on the travel order information, service provider information, and historical travel service information, expected revenue is determined. Orders are allocated to at least two candidate service providers based on the expected revenue, resulting in an order scheduling result. In allocating orders, this application embodiment considers not only the single information corresponding to the travel order information but also the service provider information and historical travel service information of at least two candidate service providers capable of providing travel services for the travel order. Therefore, the determined expected revenue can characterize the comprehensive revenue situation. The comprehensive revenue includes not only the short-term revenue corresponding to the travel order but also the medium-term and long-term revenue reflected by the probability of candidate service providers providing travel services in the future. From the perspective of comprehensive revenue, this increases the number of service providers providing travel services in the future predetermined time period, ensures the number of available vehicles on the order scheduling platform, achieves global optimization of the travel scheduling service, improves the sustainable development of the order scheduling platform, and enhances the travel experience. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings.

[0013] Figure 1 A flowchart illustrating the steps of an order scheduling method provided in this application embodiment;

[0014] Figure 2 This application provides a schematic diagram of the architecture of an order scheduling system.

[0015] Figure 3 This is a schematic diagram illustrating an application scenario of an order scheduling method provided in an embodiment of this application;

[0016] Figure 4 A structural block diagram of an order scheduling platform and a travel platform provided in an embodiment of this application;

[0017] Figure 5 A flowchart illustrating the steps of another order scheduling method provided in this application embodiment;

[0018] Figure 6 A flowchart illustrating the steps of another order scheduling method provided in this application embodiment;

[0019] Figure 7A flowchart illustrating the steps of another order scheduling method provided in this application embodiment;

[0020] Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0021] To enable those skilled in the art to better understand the technical solutions in the embodiments of this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art should fall within the protection scope of the embodiments of this application.

[0022] It should be noted that the "first" and "second" in this application are only for distinguishing names and do not represent a sequential relationship. They should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated, such as "first benefit" and "second benefit".

[0023] The specific implementation of the embodiments of this application will be further described below with reference to the accompanying drawings.

[0024] Example 1

[0025] The order scheduling method provided in Embodiment 1 of this application, such as Figure 1 As shown, Figure 1 This is a flowchart of an order scheduling method provided in an embodiment of this application.

[0026] To facilitate the explanation of the order scheduling scheme provided in the embodiments of this application, the system architecture and application scenarios to which this scheme is applicable will be described by way of example below.

[0027] like Figure 2 As shown, Figure 2 This is a schematic diagram of the architecture of an order scheduling system provided in an embodiment of this application. Figure 2In this framework, the order dispatch platform acts as an aggregation platform, connecting to multiple ride-hailing platforms (represented as ride-hailing platforms 1, 2...N in the diagram). Each ride-hailing platform can be considered a sub-platform of the order dispatch platform. Each ride-hailing platform may connect to one or more (two or more) ride-hailing service providers (three are used as an example in the diagram). Each ride-hailing service provider has a certain number of contracted service providers (drivers) (represented as n in the diagram). Thus, the order dispatch platform can aggregate more service providers, providing faster and better services to service requesters (ride users) and offering more ride-hailing opportunities to service providers (drivers). Furthermore, it effectively reduces the implementation cost of ride-hailing platforms. These platforms do not need complex functions such as predicting expected revenue, predicting service provider retention rates, predicting revenue per order, predicting order revenue within a predetermined time period, or predicting order revenue within a future predetermined time period to provide quality-assured ride-hailing services.

[0028] The order scheduling method involves the interaction between the order scheduling platform, the travel platform, the service request terminal, and the service provider terminal. When a service requester needs to travel, they can send a travel request through the interface (such as the application interface) provided by the order scheduling platform via the service request terminal. The service request terminal generates travel order information based on the travel request. After receiving the travel order information, the order scheduling platform will send the travel order information to the multiple travel platforms it aggregates.

[0029] Each travel platform receives current location and service status information from its contracted travel service providers. Based on multiple current location information, multiple current service status information, and travel order information, each travel platform identifies at least two candidate service providers and generates feedback information. The feedback information carries the service provider information of at least two candidate service providers corresponding to the travel order information and the historical travel service information of at least two candidate service providers, and sends the feedback information to the order scheduling platform.

[0030] Finally, the order scheduling platform selects the service provider that matches each travel order message from at least two candidate service providers, obtains the order scheduling result, and provides feedback to the service requester.

[0031] It should be noted that in practical applications, the aforementioned order scheduling platform can be deployed on a server or server cluster. In this case, the server or server cluster can also be considered the order scheduling platform. However, it is not limited to this; the order scheduling platform can also be deployed in the cloud. In this case, the cloud-based hardware and software setup with the functions implemented by the aforementioned order scheduling platform can also be considered the order scheduling platform. Similarly, each travel platform can also be deployed on a server, server cluster, or in the cloud, and the corresponding hardware and software setup can be considered the travel platform. The following describes the travel order processing scheme based on the above architecture through several embodiments.

[0032] For example, the server deployed by the ride-hailing platform can be used to process information and / or data related to ride-hailing orders. The server stores historical travel service information from multiple service providers. In a real-world travel scenario, after receiving travel order information from the order dispatch platform, the ride-hailing platform obtains service provider information from at least two candidate service providers within a preset range of the pick-up point, whose distance from the pick-up point is less than a preset distance (e.g., 3km, 8km), and / or whose estimated pick-up time from the pick-up point is less than a preset time (e.g., 5 minutes, 10 minutes). Then, based on the service provider information of the at least two candidate service providers, the platform filters out the historical travel service information of at least two candidate service providers from the historical travel service information pre-stored in the server.

[0033] The aforementioned preset range, preset distance, and preset time can be appropriately set by those skilled in the art according to actual needs, or determined through analysis of a large number of range thresholds, distance thresholds, and time thresholds used in the processing of determining at least two candidate service providers based on a large number of orders.

[0034] In this embodiment, the number of orders is not limited; it can be one, two, or more. Taking multiple orders as an example, these multiple orders can be orders received by the order dispatch platform within a certain time period (e.g., 5 seconds, 10 seconds, 20 seconds). For instance, these multiple orders can be orders located in a certain area, and the starting point of these multiple orders is located in a certain area. This area can be a pre-defined regular or irregular shape such as a rectangle, hexagon, or circle; this embodiment does not impose any restrictions on this. In this embodiment, the number of service providers is at least two. These service providers can be multiple drivers waiting to accept orders located in a certain area. Taking multiple travel orders as an example, the multiple drivers waiting to accept orders can be in the same area as the multiple travel orders.

[0035] like Figure 3 As shown, Figure 3This is a schematic diagram illustrating an application scenario of an order scheduling method provided in an embodiment of this application. The application scenario includes: a service request terminal, a service provider terminal, and an order scheduling platform. Figure 3 The example shows two service request terminals and four service provider terminals. Through the order scheduling method provided in this application embodiment, the order scheduling platform assigns order A to service provider (4) and order B to service provider (2).

[0036] Based on the above Figure 2 China's order scheduling system and Figure 3 The application scenarios of the order scheduling method include the following steps: Figure 1 As shown.

[0037] Step S101: Receive travel order information and obtain service provider information and historical travel service information of at least two candidate service providers corresponding to the travel order information.

[0038] In this embodiment of the application, the travel order can be an order initiated by a service requester (e.g., a passenger) to an order dispatch platform. The travel order information is used to characterize the relevant information of the service requester's travel needs, including the service requester's identifier, the travel (ride) origin, the travel distance between the travel (ride) origin and the travel (ride) destination, and the travel time corresponding to the order (e.g., whether it is a weekend, whether it is at night, whether it is during rush hour).

[0039] Among them, the origin of the trip is related to the available vehicles nearby, which is used to filter candidate service providers; the travel distance is related to the revenue of the order; the travel time of the order is closely related to the availability of vehicles and has a great influence on the revenue of the order. For example, during rush hour, the number of available vehicles is less than during normal time periods, and the travel time for the same distance will be longer than during normal time periods. Correspondingly, both travel time and vehicle availability will affect the revenue of the order.

[0040] Understandably, in addition to the key information mentioned above, travel order information may also include at least one of the following: travel (ride) destination, travel (ride) origin and / or whether the travel (ride) destination is a business district, travel (ride) origin and / or travel (ride) destination is a transportation hub, order origin city, vehicle type, number of passengers, estimated order price, estimated travel time, estimated travel distance, estimated arrival time, weather conditions, and traffic conditions (e.g., whether tolls are incurred, whether the route passes through congested areas).

[0041] In this embodiment, the service provider (e.g., driver, operator) receives orders through an order dispatch platform. Service provider information characterizes the service capacity of a service provider within a preset range of the pick-up point when providing the service for this trip. This includes vehicle status information (e.g., whether it is idle, about to become idle, or carrying passengers but not full) and current location information. The vehicle status information reflects the vehicle's availability to provide service in the coming period, and the current location information reflects the area where the vehicle is located. Candidate service providers can be filtered based on the current location information and the trip order information.

[0042] Understandably, in addition to the key information mentioned above, service provider information may also include at least one of the following: vehicle information (e.g., model, class, license plate number, number of seats), current driving speed, distance from the order's pick-up point, estimated time of arrival at the order's pick-up point, and driver information (e.g., name, driving experience, service rating, occupation).

[0043] Historical travel service information is used to characterize the service provider's service capabilities when providing travel services in the past. This includes the average number of historical rejected orders within a scheduled time period (e.g., daily, weekly, monthly, quarterly, yearly), the average duration of historical travel services provided within a scheduled time period, the time period corresponding to the historical travel services provided (e.g., morning, afternoon, evening, night, weekday, weekend), the region corresponding to the historical services provided (e.g., drivers tend to prefer familiar places), and the average historical order revenue within a scheduled time period.

[0044] Among these, rejected orders, service duration, service area, and service time period reflect the service provider's past service habits. Based on these habits, the number of orders accepted and the service duration can be estimated for a given period. The average historical order revenue within the scheduled time period can be used to predict order revenue within that period. It is understood that, in addition to the aforementioned key information, historical travel service information may also include at least one of the following: the ratio of historical rejected orders to historical accepted orders, the average number of historical completed orders within the scheduled time period, and the areas where the service provider stopped accepting orders during historical scheduled time periods (e.g., drivers often tend to stop accepting orders closer to home).

[0045] Step S102: Determine the expected revenue based on travel order information, service provider information, and historical travel service information.

[0046] In this embodiment, when allocating orders, the expected revenue is determined based on the travel order information, the service provider information corresponding to at least two candidate service providers capable of providing travel services for the travel order, and their historical travel service information. The expected revenue includes the comprehensive revenue of each candidate service provider for each travel order. The comprehensive revenue includes not only the revenue generated by the travel order but also the probability that the candidate service provider will provide travel services in the future. The revenue generated by the current travel order can be understood as short-term revenue. The probability that the candidate service provider will provide travel services in the future can be reflected by medium-term and long-term revenue. Both medium-term and long-term revenue represent whether the travel order is allocated to a particular candidate service provider, which will affect whether the candidate service provider will provide travel services in the future, thereby affecting the revenue generated by the candidate service provider in the future.

[0047] Optionally, in one embodiment of this application, step S102 may include steps S1021-S1023.

[0048] Step S1021: Predict the service provider retention rate based on travel order information, service provider information, and historical travel service information.

[0049] Among them, the service provider retention rate is used to characterize the probability that a candidate service provider will provide travel services within a future scheduled time period.

[0050] The time corresponding to the order is used as the current time, and the time for future bookings is later than the current time. For example, the time corresponding to the order is the same day, and the future booking time could be the next day.

[0051] In this example, for a driver who historically receives an average of 8 orders per day and whose historical ride-hailing services were provided during the daytime, if the driver only receives two orders before 5:00 PM on a given day, the driver may not provide ride-hailing services the following day, which can be interpreted as the driver's lack of motivation.

[0052] In this embodiment, the service provider retention rate of each service provider is predicted based on travel order information, service provider information, and historical travel service information. For example, if there are M orders and N service providers, the retention rates of M×N service providers are predicted. This example uses travel order information, service provider information, and historical travel service information to predict service provider retention rates, so that subsequent order allocation to service providers can be determined based on these retention rates, thereby increasing service provider motivation and ultimately improving the order revenue of the order scheduling platform.

[0053] Step S1022: Based on historical travel service information and service provider retention rates, predict order revenue for the future booking period.

[0054] In this example, the historical travel service information includes the average historical order revenue within the scheduled time period, the average historical number of completed orders within the scheduled time period, and the average duration of travel services provided within the scheduled time period. By combining historical travel service information with service provider retention rates, order revenue for future scheduled time periods can be estimated. For example, taking a scheduled time period of one day, the order revenue for the next day can be estimated based on the average number of completed orders and the service provider retention rate within that day. Alternatively, the order revenue for the next day can be estimated based on the average historical order revenue and the service provider retention rate within that day. Or, the order revenue for the next day can be estimated based on the average duration of travel services within that day and the service provider retention rate. In this embodiment, order revenue for future scheduled time periods can also be understood as long-term revenue.

[0055] Step S1023: Based on the order revenue within the future scheduled time period, and combining the pre-determined single order revenue corresponding to each travel order information and the order revenue within the scheduled time period, determine the expected revenue.

[0056] The time of the scheduled time period is earlier than the time of the future scheduled time period. The time of the scheduled time period includes the order time corresponding to the travel order information. The future scheduled time period and the scheduled time period are adjacent in time.

[0057] In this embodiment, the pre-determined revenue per order and revenue within a predetermined time period corresponding to each travel order can be determined by the order scheduling platform based on the travel order information, service provider information, and historical travel service information, or it can be obtained from other relevant servers. This embodiment does not impose any restrictions on this.

[0058] In this example, the revenue per order corresponding to each travel order can be understood as short-term revenue. For instance, the inherent value of an order can also be understood as its price. The revenue per order corresponding to each travel order can represent the inherent value of each order. The price of the order can be referenced to the estimated order price; however, the final transaction price of the order will differ from the estimated order price during the actual trip. In one feasible approach, the price of the order can be determined based on at least one of the following factors: origin, destination, order initiation time, estimated arrival time, estimated travel distance, weather conditions, traffic conditions, and the supply and demand relationship between the driver and the order. It is understood that the revenue per order can also include other order-related revenue, such as passenger tips and trip subsidies for returning empty. In a travel scenario, the price of the order can be the fare paid by the service requester for the order.

[0059] In this example, taking a predetermined time period of one day as an example, the order revenue within the predetermined time period can represent the order revenue within one day. Whether a current order is assigned to a service provider has a significant impact on the order revenue generated by the service provider subsequently. For example, if the origin of the current order is located in an area with high service demand (e.g., a city center), and the destination of the order is located in an area with low service demand (e.g., a suburb), then the driver may not be able to receive any orders for a long time after completing the order, or the driver may need to return empty to an area with high service demand to receive orders. Even considering the inherent value of the order, the driver's total order revenue in the future (e.g., one hour, several hours, one day, etc.) may be reduced due to accepting the order.

[0060] In calculating expected revenue, this application embodiment considers not only the current single order revenue and the order revenue within the scheduled time period, but also the order revenue within the future scheduled time period. That is, it considers the short-term (single order revenue), medium-term (order revenue within the scheduled time period), and long-term (order revenue within the future scheduled time period) perspectives respectively. This ensures that when allocating orders based on expected revenue in the future, it increases the number of travel services provided by the service provider within the future scheduled time period, guarantees order revenue, and improves the travel experience.

[0061] Optionally, in one embodiment of this application, the revenue per order and the revenue per order within a predetermined time period in S1023 above can be obtained in the following way: Obtain current travel road information; perform traffic analysis on the current travel road information to obtain traffic analysis results; based on the traffic analysis results, combined with the current location information in the service provider information, and the travel origin location information and travel destination location information in the travel order information, predict the revenue per order and the revenue per order within a predetermined time period respectively.

[0062] Current travel route information includes, but is not limited to, current road conditions, the geographical location of the current road, and current vehicle availability. This information affects the completion time of the current order and the number of orders available for service within the next preset time period. In other words, current travel route information affects not only the revenue from the current order but also the revenue from orders available for service within the preset time period. By analyzing traffic conditions, service provider information, travel order information, and historical travel service information, the revenue per order and the revenue from orders within the preset time period can be estimated.

[0063] In this embodiment, by analyzing current road conditions and combining the service provider's current location information, the origin and destination locations, the travel distance, and the travel time corresponding to the travel order, as well as the information on unserved services, service duration, service time period, and service area in historical travel service information, the revenue generated by the travel order is predicted, thus improving the accuracy of single-order revenue prediction. When predicting order revenue within a predetermined time period, the number of travel orders corresponding to the travel time point is also considered to estimate the number of services and service duration provided by the service provider within the next predetermined time period, thereby predicting order revenue within the predetermined time period and improving the accuracy of order revenue prediction within the predetermined time period.

[0064] Step S103: Assign orders to at least two candidate service providers based on expected revenue to obtain order scheduling results.

[0065] In this embodiment, the expected revenue includes the combined revenue of each candidate service provider for each travel order. This can be understood as matching each candidate service provider with each travel order to form an order matching pair, and allocating travel orders by comprehensively considering the combined revenue of each order matching pair. Since the preset revenue comprehensively considers both short-term and future revenue, allocating orders based on the preset revenue improves the accuracy of order scheduling results and ensures the retention rate of service providers over the future period, i.e., the number of travel services provided.

[0066] The order scheduling scheme provided in this application not only considers the single information corresponding to the travel order information, but also the service provider information corresponding to at least two candidate service providers that can provide travel services for the travel order, as well as their historical travel service information. The determined expected revenue thus represents the comprehensive revenue situation. The comprehensive revenue includes not only the short-term revenue corresponding to the travel order, but also the medium-term and long-term revenue reflected by the probability of the candidate service providers providing travel services in the future. From the perspective of comprehensive revenue, this increases the number of service providers offering travel services in the future predetermined time period, ensures the number of available vehicles on the order scheduling platform, achieves global optimization of the travel scheduling service, improves the sustainable development of the order scheduling platform, and enhances the travel experience.

[0067] The order scheduling method of this application embodiment can be executed by any suitable electronic device with data processing capabilities, including but not limited to: servers, mobile terminals (such as mobile phones, PADs, etc.) and PCs.

[0068] Example 2

[0069] In one feasible order scheduling method, the order scheduling platform matches travel orders with candidate service providers. Based on an order scheduling model, it matches service providers for each travel order. This model can include a cancellation rate model and a spatiotemporal value model. Order scheduling is achieved by selecting to optimize either a short-term goal (e.g., using the cancellation rate model to optimize the current completion rate) or a medium-term goal (e.g., using the spatiotemporal value model to optimize the daily completion volume). However, this approach has the following problems: Concentrating orders on active top service providers (e.g., drivers) increases the workload of top drivers and hinders the development of work habits among inactive drivers, making it difficult to improve the retention and service duration of these inactive drivers.

[0070] Ride-hailing platforms provide ride-hailing services. Different types of drivers provide services on different days per week, for different durations per day, and during different time slots. The number of available vehicles aggregated by the order dispatch platform directly affects whether passengers can hail a ride and how quickly they can do so—a crucial aspect of the travel experience. This application provides an order dispatch method that can improve service provider retention, thereby increasing the number of days drivers provide ride-hailing services and the number of available vehicles on the order dispatch platform, ultimately improving the travel experience. Details are as follows.

[0071] like Figure 4 As shown, Figure 4 This is a structural block diagram of the order scheduling platform and travel platform provided in the embodiments of this application. The travel platform includes a historical travel service information storage section 402 and a historical travel service information query section 403; the order scheduling platform includes a supply and demand information input section 401, a service provider retention prediction section 404, a short-term revenue calculation section 405, a medium-term revenue calculation section 406, a long-term revenue calculation section 407, a matching problem solving section 408, and an order scheduling result output section 409. The order scheduling platform aggregates multiple travel platforms, and data and information can be exchanged between the order scheduling platform and the travel platforms.

[0072] Combination Figure 4 In the various parts of this application, Embodiment 2 is based on the solution of Embodiment 1. Optionally, Embodiment 2 of this application may include the following steps S201-S208.

[0073] Step S201: Obtain travel order information, service provider information of at least two candidate service providers corresponding to the travel order information, and historical travel service information.

[0074] The supply and demand information input section 401 is used to receive travel order information sent by service provider terminals. Based on the travel order information sent by the supply and demand information input section 401, the travel platform determines at least two candidate service providers. Based on the service provider information of the at least two candidate service providers, the historical travel service information query section 403 is used to query the historical travel service information storage section 402 for the historical travel service information of the at least two candidate service providers. The supply and demand information input section 401 is also used to receive service provider information and historical travel service information of the at least two candidate service providers sent by the travel platform.

[0075] Step S202: Input the travel order information, service provider information, and historical travel service information into the pre-trained retention rate prediction model to obtain the service provider retention rate.

[0076] The service provider retention prediction section 404 is used to predict the service provider retention rate after assigning an order to the service provider, and the service provider retention rate after not assigning an order to the service provider, based on travel order information, service provider information, and historical travel service information retrieved.

[0077] The retention rate prediction model in this example is used to predict the retention probability of the service provider. The retention rate prediction model can be any suitable neural network (NN) model capable of predicting retention rates based on travel order information, service provider information, and historical travel service information. This application does not limit the specific structure of the retention rate prediction model used; for example, it can be a convolutional recurrent neural network (CRNN) or a convolutional neural network (CNN).

[0078] In this example, the retention rate prediction model can be pre-trained as follows: Training samples are obtained, including travel order sample information, service provider sample information, and historical travel service sample information. These training samples are input into the initial retention rate prediction model to obtain the retention prediction probability. Based on the difference between the retention prediction probability and the retention label probability of the training samples, the initial retention rate prediction model is trained to obtain the trained retention rate prediction model. Here, the retention label probability corresponds to the training sample. When training the initial retention rate prediction model using a single sample, the model parameters are continuously adjusted and iteratively updated until a training termination condition is met to obtain the trained retention rate prediction model. The training termination condition may be, for example, reaching a preset number of training iterations, or the prediction result of the retention rate prediction model being within a preset deviation range.

[0079] It should be noted that the retention rate prediction model in this application is not limited to the pre-trained retention rate prediction model in this application. This application does not restrict the specific structure of the retention rate prediction model; that is, existing models trained by others, as long as they have the function of predicting retention rates, can be adapted to the solution of this application. The retention rate prediction model with the above-mentioned retention rate prediction function provided in this application is an optional improvement solution, not a mandatory solution.

[0080] In this example, a pre-trained retention rate prediction model is used to predict retention rates based on travel order information, service provider information, and historical travel service information, thereby improving the accuracy of service provider retention rates.

[0081] Optionally, in one embodiment of this application, the service provider retention rate includes an allocation retention rate, which characterizes the order allocation to the service provider, and an unallocated retention rate, which characterizes the order not allocated to the service provider.

[0082] This example uses a one-day booking period as an illustration. Whether or not an order is assigned to a driver on the same day directly affects the driver's motivation and has a significant impact on the probability that the driver will provide travel services the next day. In this embodiment, the service provider retention rate includes two parts: one part is the probability that the service provider will provide travel services the next day when an order is assigned to the service provider, and the other part is the probability that the service provider will provide travel services the next day when an order is not assigned to the service provider.

[0083] Step S203: Based on historical travel service information and service provider retention rates, predict order revenue for the future booking period.

[0084] The long-term revenue calculation section 407 is used to calculate the long-term revenue after allocating orders to service providers based on the historical travel service information and service provider retention rates of at least two candidate service providers.

[0085] Step S203 is the same as step S1022 in Embodiment 1, and will not be repeated here.

[0086] Optionally, in one embodiment of this application, when predicting order revenue within a future predetermined time period in step S203, it can be specifically implemented through the following two examples.

[0087] Example 1: Determine the first revenue based on the average number of completed orders within the historical travel service period in the historical travel service information and the allocated retention rate; determine the second revenue based on the average number of completed orders within the historical travel service information within the historical travel service period and the unallocated retention rate; determine the order revenue for future booking periods based on the difference between the first revenue and the second revenue.

[0088] This example uses a one-day booking period as an example. The first revenue is determined based on the average number of completed orders per day in historical travel service information and the allocation retention rate of orders assigned to service providers. Optionally, the product of the two is used as the first revenue. The second revenue is determined based on the average number of completed orders per day in historical travel service information and the unallocated retention rate of orders not assigned to service providers. Optionally, the product of the two is used as the second revenue. The difference between the first and second revenues is then determined as the order revenue for the second day.

[0089] Example 2: Determine the order revenue for future booking periods based on the difference between the allocated retention rate and the unallocated retention rate, and the average number of historical completed orders within the booking period in the historical travel service information.

[0090] Optionally, the product of the difference between the allocated retention rate and the unallocated retention rate and the average number of historical completed orders within the scheduled time period in the historical travel service information can be used to determine the order revenue for the future scheduled time period.

[0091] Example 2 and Example 1 illustrate two different methods for calculating order revenue. Specifically, Example 2 uses the average number of completed orders within a given booking period from historical travel service information, along with the allocated and unallocated retention rates, to determine order revenue for future booking periods, thus improving the accuracy of order revenue prediction.

[0092] Step S204: Input the travel order information, service provider information, and historical travel service information into the pre-trained completion rate prediction model to obtain the revenue per order.

[0093] The short-term revenue calculation section 405 is used to calculate the short-term revenue after allocating orders to service providers based on the received travel order information, the service provision information of at least two candidate service providers, and the historical travel service information of at least two candidate service providers.

[0094] Single-order revenue refers to the revenue generated by the current order, which can be understood as short-term revenue. In this example, the completion rate prediction model predicts the probability of order completion. This model can be understood as a cancellation rate model; in a travel scenario, orders can be cancelled by passengers or drivers. The cancellation rate model can be used to predict whether the current order will be completed and whether it will generate revenue. In this example, travel order information, service provider information, and historical travel service information are input into the cancellation rate model, and the output is single-order revenue.

[0095] It should be noted that the completion rate prediction model in this example is similar to the retention rate prediction model in step S202. Both can be understood as machine learning models, but the training samples are different and the functions implemented are different. The specific training process of the completion rate prediction model in this example can be referred to step S202, and will not be repeated here.

[0096] Step S205: Input the travel order information, service provider information, and historical travel service information into the pre-trained completion quantity prediction model to obtain the revenue of the allocated orders within a predetermined time period after the orders are allocated to the service provider and the revenue of the unallocated orders within a predetermined time period after the orders are not allocated to the service provider; determine the order revenue within the predetermined time period based on the difference between the revenue of the allocated orders within the predetermined time period and the revenue of the unallocated orders within the predetermined time period.

[0097] The interim revenue calculation section 406 is used to calculate the interim revenue after allocating orders to service providers based on the received travel order information, the service provider information of at least two candidate service providers, and the historical travel service information of at least two candidate service providers.

[0098] In this example, the completion quantity prediction model is used to predict the number of orders completed within a predetermined time period. The completion quantity prediction model can be understood as a spatiotemporal value model, which considers both time and space. Whether the current order is assigned to the service provider will affect the order revenue brought by the service provider in the future.

[0099] This example considers the spatiotemporal distribution of orders, which includes, but is not limited to, at least one of the following: the number of orders at the time of order initiation (e.g., a large number of orders initiated at the same time during the morning rush hour), the region of the origin, the region of the destination, the order initiation time, and the service provider's empty driving time. This example inputs travel order information, service provider information, and historical travel service information into the spatiotemporal value model to obtain revenue representing the allocated order revenue within a predetermined time period after an order is assigned to a service provider, and revenue representing the unallocated order revenue within a predetermined time period after an order is not assigned to a service provider. The difference between the allocated order revenue and the unallocated order revenue within the predetermined time period is determined as the order revenue within the predetermined time period, thus improving the accuracy of the order revenue within the predetermined time period.

[0100] Order revenue within a predetermined time period can be understood as interim revenue. Taking a predetermined time period of one day as an example, order revenue within a predetermined time period represents the order revenue within one day. In this example, the difference between the allocated order revenue within one day after the order is assigned to the service provider and the unassigned order revenue within one day after the order is not assigned to the service provider is determined as the order revenue within one day.

[0101] It should be noted that the completion quantity prediction model in this example is similar to the retention rate prediction model in step S202. Both can be understood as machine learning models, only the training samples and the functions they implement are different. The specific training process of the completion quantity prediction model in this example can be referred to step S202, and will not be repeated here. This application embodiment does not restrict the execution order of steps S203, S204, and S205, and steps S203, S204, and S205 can also be executed simultaneously.

[0102] Step S206: Based on the revenue focus, determine the three weights corresponding to the revenue per order, the revenue of orders within the scheduled time period, and the revenue of orders within the future scheduled time period.

[0103] Revenue focus includes the emphasis on revenue per order, revenue per order within a predetermined time period, and revenue per order within a future predetermined time period. For example, typically, the weights for all three are set to 1, meaning short-term, medium-term, and long-term revenue are considered comprehensively, without prioritizing any single revenue stream. Later, the weight values ​​can be adjusted to adjust the revenue ratio. For instance, if there is a greater focus on revenue per order within a future predetermined time period, the third weight is set higher, such as 1.5 or 1.2, meaning future revenue is considered more when allocating orders. Conversely, if the focus on revenue per order decreases, the first weight is set lower, such as 0.8 or 0.7.

[0104] In this example, the first weight for the revenue of a single order, the second weight for the revenue of orders within a predetermined time period, and the third weight for the revenue of orders within a future predetermined time period are determined based on the revenue focus. By adjusting the weights according to the revenue focus, the revenue ratio can be flexibly adjusted to improve revenue.

[0105] Step S207: Determine the revenue function to represent the expected revenue based on the revenue of each order, the revenue of orders within the predetermined time period, the revenue of orders within the future predetermined time period, and the three weights corresponding to the revenue of each order, the revenue of orders within the predetermined time period, and the revenue of orders within the future predetermined time period.

[0106] For example, the revenue function = revenue per order × first weight + revenue per order within a predetermined time period × second weight + revenue per order within a future predetermined time period × third weight. By comprehensively considering short-term, medium-term, and long-term revenue, this revenue function improves the accuracy of order scheduling results and enhances allocation effectiveness when subsequently allocating orders to at least two candidate service providers based on the revenue function.

[0107] Step S208: Assign orders to at least two candidate service providers according to the revenue function to obtain the order scheduling result.

[0108] The matching problem-solving section 408 is used to calculate and solve the matching problem between travel orders and service providers based on a benefit function determined by short-term, medium-term, and long-term benefits and their corresponding weights, and to find the optimal solution that maximizes the expected benefit. The order scheduling result output section 409 is used to output the calculated order scheduling results.

[0109] Since the revenue function in this example takes into account short-term, medium-term, and long-term revenue, it not only improves the allocation efficiency when scheduling orders for at least two candidate service providers, but also ensures the number of available vehicles on the order scheduling platform.

[0110] When allocating orders in step S208, the following two examples can be used to illustrate this.

[0111] Example 1: If there is only one order, the candidate service provider corresponding to the maximum function value of the revenue function will be selected as the service provider that matches the order.

[0112] If there is only one order, then we can choose one of at least two candidate service providers. In this example, the candidate service provider with the maximum function value of the revenue function is selected as the service provider that matches the order, which improves the allocation efficiency.

[0113] Example 2: If there are at least two orders, multiple order matching pairs are obtained based on the travel order information and service provider information. Each order matching pair includes one order and one candidate service provider. Different order matching pairs correspond to different revenue functions. Based on the multiple revenue functions corresponding to the multiple order matching pairs, orders are allocated to at least two candidate service providers to obtain the order scheduling result.

[0114] Optionally, if the number of orders is at least two, multiple order matching pairs are obtained based on the travel order information and service provider information. A bipartite graph matching algorithm is used to allocate orders to at least two candidate service providers according to the multiple revenue functions corresponding to the multiple order matching pairs, thereby obtaining the order scheduling result.

[0115] Bipartite graph matching algorithms can include maximum matching algorithms (e.g., Hungarian algorithm, Hopcroft-Karp algorithm, etc.) and matching algorithms (e.g., Kuhn-Munkres (KM) algorithm). Taking the KM matching algorithm as an example, each order matching pair corresponds one-to-one with the revenue function. The KM matching algorithm can find a better match with relatively good time complexity. It matches all travel orders and service providers, obtaining the service provider with the highest matching degree. The order dispatch platform then assigns orders to vehicles belonging to the service provider with the highest matching degree, improving the allocation efficiency.

[0116] In this embodiment of the application, by predicting the impact of order allocation on service provider retention, service provider retention is included as a long-term benefit in the optimization objective when allocating orders. This will make it more inclined to allocate orders to low-activity service providers in order to increase the service provider retention rate, that is, to increase the service time of these service providers, increase the number of available vehicles of the travel platform aggregated by the order scheduling platform, and improve the travel experience.

[0117] It should be noted that the embodiments in this application are illustrated by taking the cooperation and information exchange between the travel platform and the order dispatch platform as an example. It can be understood that multiple travel platforms and order dispatch platforms can be integrated into the online ride-hailing service platform. The online ride-hailing service platform stores the historical travel service information of multiple service providers. The steps of obtaining travel order information, determining at least two candidate service providers, querying historical travel service information, predicting retention rate and predicting revenue are all completed by the online ride-hailing service platform.

[0118] The order scheduling method of this application embodiment can be executed by any suitable electronic device with data processing capabilities, including but not limited to: servers, mobile terminals (such as mobile phones, PADs, etc.) and PCs.

[0119] Example 3

[0120] This application also provides an order scheduling method, such as... Figure 5 As shown, Figure 5 This is a flowchart illustrating another order scheduling method provided in this application embodiment. This embodiment describes the order scheduling method provided in this application embodiment from the perspective of an order scheduling platform. The order scheduling method in this embodiment includes the following steps.

[0121] Step S501: Receive travel order information sent by the service requester through the order scheduling platform. The order scheduling platform is an aggregation platform that aggregates multiple travel platforms.

[0122] Step S502: Send the travel order information to multiple travel platforms and receive feedback information from multiple travel platforms. The feedback information carries service provider information and historical travel service information of at least two candidate service providers corresponding to the travel order information.

[0123] Step S503: Determine the expected revenue based on travel order information, service provider information, and historical travel service information.

[0124] Optionally, in this embodiment, both the order scheduling platform and the travel platform can display the expected revenue. This allows for effective intervention in the order scheduling process based on the expected revenue.

[0125] Step S504: Assign orders to at least two candidate service providers based on expected revenue to obtain order scheduling results.

[0126] The specific implementation of the above-mentioned processing of the order scheduling platform and the technical effects it achieves can be found in the descriptions of the corresponding parts of Embodiment 1 and Embodiment 2 above, and will not be repeated here.

[0127] Example 4

[0128] This application also provides an order scheduling method, such as... Figure 6 As shown, Figure 6 This is a flowchart illustrating another order scheduling method provided in this application embodiment. This embodiment describes the order scheduling method provided in this application embodiment from the perspective of the service provider's service providing terminal. The order scheduling method in this embodiment includes the following steps.

[0129] Step S601: Send the current location information and current service status information of the service provider terminal to the travel platform, so that the travel platform sends feedback information to the order scheduling platform based on multiple current location information, multiple current service status information, and travel order information received from the order scheduling platform; the order scheduling platform determines the expected revenue based on the travel order information, the service provider information of at least two candidate service providers carried in the feedback information, and historical travel service information, and allocates orders to at least two candidate service providers according to the expected revenue, thereby obtaining the order scheduling result. The order scheduling result represents the service provider that matches each travel order information.

[0130] The service provider terminal sends its current location information and current service status information (e.g., whether it is idle, about to become idle, or has passengers but is not full) to its corresponding travel platform. Multiple travel platforms receive the current location information and current service status information from multiple service provider terminals. The specific implementation of the above processing by the order scheduling platform can be found in the descriptions of the corresponding parts of Embodiments 1 and 2 above, and will not be repeated here. Candidate service providers will be assigned matching travel orders, or they may not be assigned any travel orders.

[0131] Step S602: Receive travel order information sent by the order scheduling platform through the travel platform corresponding to the matched service provider, as well as the expected revenue corresponding to the matched travel order information.

[0132] Optionally, in this embodiment of the application, the service provider terminal (e.g., the terminal corresponding to the driver's terminal) can display the expected revenue, thereby enabling the service provider (e.g., the driver) to understand the potential revenue and increasing the service provider's motivation.

[0133] The ride-hailing order allocation process does not require the participation of service providers, who are unaware of the specific allocation strategy. Service providers assigned ride-hailing orders receive matching order information through the corresponding ride-hailing platform. If a service provider is identified as a matched provider, its terminal receives and displays the ride-hailing order information sent by the order dispatch platform through both the ride-hailing platform and the service provider. Simultaneously, the service provider's terminal also receives expected revenue, such as revenue per order, revenue for orders within a predetermined time period, and revenue for orders within future predetermined time periods. This allows service providers (such as drivers) to understand their potential revenue, increasing their awareness and motivation.

[0134] Example 5

[0135] This application also provides an order scheduling method, such as... Figure 7 As shown, Figure 7 This is a flowchart illustrating another order scheduling method provided in this application embodiment. This embodiment describes the order scheduling method provided in this application embodiment from the perspective of the service request terminal of the service requester. The order scheduling method of this embodiment includes the following steps.

[0136] Step S701: Receive the travel request initiated by the service requester and generate travel order information based on the travel request.

[0137] The order dispatch platform or travel platform provides corresponding interfaces, such as corresponding applications, to the service request terminal of the service provider. Therefore, the service requester can initiate a travel request through the application in the service request terminal.

[0138] It should be noted that in practical applications, the entity contracted with the travel platform may be a travel service provider. Therefore, this step can also be considered as the travel platform receiving travel requests initiated by service requesters through the travel service provider. However, the aforementioned interface can still be provided by the order scheduling platform or the travel platform itself. Of course, it can also be provided by the travel service provider. Regardless of which party provides it, the key is to achieve timely interaction between the aforementioned parties.

[0139] Step S702: Send the travel order information to the order scheduling platform, so that the order scheduling platform sends the travel order information to multiple travel platforms. Based on the service provider information and historical travel service information of at least two candidate service providers fed back by multiple travel platforms, and combined with the travel order information, determine the expected revenue, and allocate orders to at least two candidate service providers according to the expected revenue to obtain the order scheduling result.

[0140] As mentioned earlier, the order scheduling platform aggregates multiple travel platforms. Upon receiving travel order information from a service requesting terminal, the platform distributes it to these platforms. By identifying candidate service providers for the current travel order through each platform, its contracted travel service providers, and the service providers contracted with those providers, the platform provides feedback up the chain of command to the order scheduling platform. Furthermore, the platform uses this feedback and the travel order information to predict revenue, thereby allocating the order to at least two candidate service providers.

[0141] The specific implementation of the order scheduling platform in determining the service provider that matches the travel order information and the technical effects it achieves can be found in the descriptions of the corresponding parts of Embodiment 1 and Embodiment 2 above, and will not be repeated here.

[0142] Step S703: Receive the service provider information of the service provider assigned in the order scheduling result sent by the order scheduling platform.

[0143] As mentioned earlier, the order dispatch platform identifies service providers that match the travel order information. Then, it sends the service provider information of the matched service provider back to the service requesting terminal. The service requesting terminal can then display this service provider information to the service requester.

[0144] Example 6

[0145] Based on any of the order scheduling methods described in Embodiments 1 to 5 above, this application provides an electronic device. It should be noted that the order scheduling method of this application can be executed by any suitable electronic device with order scheduling capabilities, including but not limited to: servers, mobile terminals (such as mobile phones, PADs, etc.), and PCs. Figure 8 As shown, Figure 8 This is a structural diagram of an electronic device provided in an embodiment of this application. The specific embodiments of this application do not limit the specific implementation of the electronic device. The electronic device 80 may include: a processor 802, a communications interface 804, a memory 806, and a communication bus 808.

[0146] The processor 802, communication interface 804, and memory 806 communicate with each other via communication bus 808.

[0147] Communication interface 804 is used to communicate with other electronic devices or servers.

[0148] The processor 802 is used to execute the computer program 810, specifically the relevant steps in the above-described order scheduling method embodiment.

[0149] Specifically, computer program 810 may include computer program code, which includes computer operation instructions.

[0150] The processor 802 may be a CPU, an Application Specific Integrated Circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application. The smart device includes one or more processors, which may be processors of the same type, such as one or more CPUs; or processors of different types, such as one or more CPUs and one or more ASICs.

[0151] Memory 806 is used to store computer program 810. Memory 806 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0152] The specific implementation of each step in program 810 can be found in the corresponding steps and units described in the above-mentioned order scheduling method embodiments, and will not be repeated here. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the devices and modules described above can be referred to the corresponding process descriptions in the aforementioned method embodiments, and will not be repeated here.

[0153] Example 7

[0154] Based on the order scheduling method described in Embodiments 1 to 5 above, this application provides a computer storage medium storing a computer program that, when executed by a processor, implements the order scheduling method as described in Embodiments 1 to 5.

[0155] It should be noted that, depending on the implementation needs, the various components / steps described in the embodiments of this application can be broken down into more components / steps, or two or more components / steps or parts of the operation of components / steps can be combined into new components / steps to achieve the purpose of the embodiments of this application.

[0156] The methods described in the embodiments of this application can be implemented in hardware, firmware, or as software or computer code that can be stored in a recording medium (such as a CD-ROM, RAM, floppy disk, hard disk, or magneto-optical disk), or as computer code downloaded over a network that is originally stored in a remote recording medium or a non-transitory machine-readable medium and will be stored in a local recording medium. Thus, the methods described herein can be processed by software stored on a recording medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware (such as an ASIC or FPGA). It is understood that the computer, processor, microprocessor controller, or programmable hardware includes storage components (e.g., RAM, ROM, flash memory, etc.) capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the order scheduling method described herein is implemented. Furthermore, when a general-purpose computer accesses the code used to implement the order scheduling method shown herein, the execution of the code transforms the general-purpose computer into a dedicated computer for executing the order scheduling method shown herein.

[0157] Those skilled in the art will recognize that the exemplary units and method steps described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of the embodiments of this application.

[0158] The above embodiments are only used to illustrate the embodiments of this application, and are not intended to limit the embodiments of this application. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the embodiments of this application. Therefore, all equivalent technical solutions also fall within the scope of the embodiments of this application, and the patent protection scope of the embodiments of this application should be defined by the claims.

Claims

1. An order scheduling method, comprising: Receive travel order information, and obtain service provider information of at least two candidate service providers and historical travel service information of at least two candidate service providers corresponding to the travel order information; Based on the travel order information, the service provider information, and the historical travel service information, the expected revenue is determined; wherein, the expected revenue includes short-term revenue, medium-term revenue, and long-term revenue. Short-term revenue is the revenue generated by the current travel order, while medium-term and long-term revenue both indicate whether the travel order is assigned to a candidate service provider and affect the revenue generated by that candidate service provider in the future. Based on the expected revenue, orders are allocated to at least two of the candidate service providers to obtain the order scheduling result; The expected revenue is determined based on the travel order information, the service provider information, and the historical travel service information, including: The service provider retention rate is predicted based on the travel order information, the service provider information, and the historical travel service information. The service provider retention rate is used to characterize the probability that a candidate service provider will provide travel services within a future predetermined time period. The service provider retention rate includes the allocation retention rate, which characterizes the order allocation to the service provider, and the unallocated retention rate, which characterizes the order not allocated to the service provider. Based on the historical travel service information and the service provider retention rate, predict order revenue for the future booking period; The expected revenue is determined based on the order revenue within the future scheduled time period, combined with the pre-determined single order revenue corresponding to each of the travel order information and the order revenue within the scheduled time period. The step of predicting order revenue within the future booking period based on the historical travel service information and the service provider retention rate includes: The first revenue is determined based on the average number of completed orders within the reserved time period in the historical travel service information and the allocated retention rate; The second revenue is determined based on the average number of completed orders within the reserved time period in the historical travel service information and the unallocated retention rate; The order revenue for the future predetermined time period is determined based on the difference between the first revenue and the second revenue.

2. The method according to claim 1, wherein, The step of predicting order revenue within the future booking period based on the historical travel service information and the service provider retention rate includes: The order revenue for the future scheduled time period is determined based on the difference between the allocated retention rate and the unallocated retention rate, and the average number of historical completed orders within the scheduled time period in the historical travel service information.

3. The method according to claim 1, wherein, Based on the order revenue within the future scheduled time period, and combining the pre-determined single order revenue corresponding to each of the aforementioned travel order information and the order revenue within the scheduled time period, the expected revenue is determined, including: Based on the revenue of each single order, the revenue of orders within the predetermined time period, the revenue of orders within the future predetermined time period, and the three weights corresponding to the revenue of each single order, the revenue of orders within the predetermined time period, and the revenue of orders within the future predetermined time period, a revenue function is determined to characterize the expected revenue.

4. The method according to claim 3, wherein, Based on the expected revenue, orders are allocated to at least two of the candidate service providers to obtain order scheduling results, including: If there is only one order, the candidate service provider corresponding to the maximum function value of the revenue function will be selected as the service provider that matches the order. If there are at least two orders, multiple order matching pairs are obtained based on the travel order information and the service provider information. Each order matching pair includes one order and one candidate service provider. Different order matching pairs correspond to different revenue functions. Based on the multiple revenue functions corresponding to the multiple order matching pairs, orders are allocated to at least two of the candidate service providers to obtain the order scheduling result.

5. The method according to claim 3, wherein, The method further includes: Based on the level of attention to revenue, three weights are determined for the revenue per order, the revenue per order within the predetermined time period, and the revenue per order within the future predetermined time period.

6. The method according to claim 1, wherein, Based on the travel order information, the service provider information, and the historical travel service information, the service provider retention rate is predicted, including: The travel order information, the service provider information, and the historical travel service information are input into a pre-trained retention rate prediction model to obtain the service provider retention rate.

7. The method according to claim 1, wherein, The method further includes: The travel order information, the service provider information, and the historical travel service information are input into a pre-trained completion rate prediction model to obtain the revenue per order. Input the travel order information, the service provider information and the historical travel service information into a pre-trained completion quantity prediction model to obtain revenue for allocated orders within a predetermined time period after an order is allocated to a service provider and revenue for unallocated orders within a predetermined time period after an order is not allocated to a service provider. The order revenue for the predetermined time period is determined based on the difference between the revenue from orders allocated within the predetermined time period and the revenue from orders not allocated within the predetermined time period.

8. The method according to claim 1, wherein, The method further includes: Get current travel route information; The current travel route information is analyzed to obtain the traffic analysis results; Based on the traffic analysis results, combined with the current location information in the service provider information, and the origin and destination location information in the travel order information, the revenue of a single order and the revenue of orders within the predetermined time period are predicted respectively.

9. An order scheduling method, comprising: The system receives travel order information sent by service requesters through an order scheduling platform, which is an aggregation platform that aggregates multiple travel platforms. The travel order information is sent to the multiple travel platforms, and feedback information is received from the multiple travel platforms. The feedback information carries service provider information of at least two candidate service providers corresponding to the travel order information and historical travel service information of at least two candidate service providers. The service provider retention rate is predicted based on the travel order information, the service provider information, and the historical travel service information. The service provider retention rate is used to characterize the probability that a candidate service provider will provide travel services within a future predetermined time period. The service provider retention rate includes the allocation retention rate, which characterizes the order allocation to the service provider, and the unallocated retention rate, which characterizes the order not allocated to the service provider. The first revenue is determined based on the average number of completed orders within the reserved time period in the historical travel service information and the allocated retention rate; The second revenue is determined based on the average number of completed orders within the reserved time period in the historical travel service information and the unallocated retention rate; The order revenue within the future predetermined time period is determined based on the difference between the first revenue and the second revenue. Based on the order revenue within the future predetermined time period, and combined with the pre-determined single order revenue corresponding to each of the travel order information and the order revenue within the predetermined time period, the expected revenue is determined; wherein, the expected revenue includes short-term revenue, medium-term revenue and long-term revenue, short-term revenue is the revenue brought by the current travel order, and medium-term revenue and long-term revenue both represent whether the travel order is allocated to a candidate service provider and affect the revenue brought by the candidate service provider in the future period of time. Orders are allocated to at least two of the candidate service providers based on the expected revenue, resulting in an order scheduling outcome.

10. An order scheduling method, comprising: Send the current location information and current service status information of the service provider terminal to the travel platform, so that the travel platform can send feedback information to the order scheduling platform based on multiple current location information, multiple current service status information, and travel order information received from the order scheduling platform; The order scheduling platform predicts the service provider retention rate based on the travel order information, the service provider information of at least two candidate service providers carried in the feedback information, and the historical travel service information of at least two candidate service providers. The service provider retention rate is used to characterize the probability that the candidate service providers will provide travel services in a future predetermined time period. The service provider retention rate includes the allocation retention rate, which characterizes the order allocation to the service provider, and the unallocated retention rate, which characterizes the order not allocated to the service provider. The first revenue is determined based on the average number of completed orders within the reserved time period in the historical travel service information and the allocated retention rate; The second revenue is determined based on the average number of completed orders within the reserved time period in the historical travel service information and the unallocated retention rate; The order revenue within the future predetermined time period is determined based on the difference between the first revenue and the second revenue. Based on the order revenue within the future predetermined time period, and combined with the pre-determined single order revenue corresponding to each of the travel order information and the order revenue within the predetermined time period, the expected revenue is determined; wherein, the expected revenue includes short-term revenue, medium-term revenue and long-term revenue, short-term revenue is the revenue brought by the current travel order, and medium-term revenue and long-term revenue both represent whether the travel order is allocated to a candidate service provider and affect the revenue brought by the candidate service provider in the future period of time. Based on the expected revenue, orders are allocated to at least two of the candidate service providers to obtain an order scheduling result, which represents the service provider that matches each of the travel order information. The system receives matching travel order information sent by the order scheduling platform through the travel platform corresponding to the matched service provider, as well as the expected revenue corresponding to the matching travel order information.

11. An order scheduling method, comprising: Receive travel requests initiated by service requesters and generate travel order information based on the travel requests; The travel order information is sent to the order scheduling platform, which then sends the travel order information to multiple travel platforms. Based on the service provider information of at least two candidate service providers and the historical travel service information of at least two candidate service providers fed back by the multiple travel platforms, the service provider retention rate is predicted in combination with the travel order information. The service provider retention rate is used to characterize the probability that the candidate service provider will provide travel services in a future predetermined time period. The service provider retention rate includes the allocation retention rate, which characterizes the order allocation to the service provider, and the unallocated retention rate, which characterizes the order not allocated to the service provider. The first revenue is determined based on the average number of historical completed orders within a predetermined time period in the historical travel service information and the allocated retention rate; the second revenue is determined based on the average number of historical completed orders within a predetermined time period in the historical travel service information and the unallocated retention rate; and the order revenue within the future predetermined time period is determined based on the difference between the first revenue and the second revenue. Based on the order revenue within the predetermined future time period, and combining the pre-determined single order revenue corresponding to each of the travel order information and the order revenue within the predetermined time period, the expected revenue is determined. Orders are then allocated to at least two candidate service providers based on the expected revenue to obtain the order scheduling result. The expected revenue includes short-term revenue, medium-term revenue, and long-term revenue. Short-term revenue is the revenue generated by the current travel order. Medium-term and long-term revenues both indicate whether the travel order is allocated to a candidate service provider and affect the revenue generated by that candidate service provider in the future. Receive the service provider information allocated in the order scheduling result sent by the order scheduling platform.

12. A computer storage medium having a computer program stored thereon, which, when executed by a processor, implements the order scheduling method as described in any one of claims 1-11.

Images