Price determination method, device and equipment for online car-hailing fixed-price order

Through deep learning models, the completion rate of online car-hailing orders is predicted and the quotation of fixed-price orders is adjusted, which solves the problem of high driver cancellation rates and improves user experience and order stability.

CN119090532AInactive Publication Date: 2024-12-06BEIJING BAILONG MAYUN TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411103131.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-12
Publication Date
2024-12-06
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The pricing discount determination method for online car-hailing orders results in a high driver cancellation rate and is not matched with the actual situation, affecting the user experience.

Method used

By obtaining the user's target fixed price order information and supply and demand information in the city area, the pre-trained deep learning model predicts the order completion rate corresponding to each price adjustment ratio, and adjusts the initial quotation based on the price adjustment ratio with the highest order completion rate to determine the final quotation , reduce driver cancellation rate.

Benefits of technology

It effectively reduces the driver cancellation rate, improves user experience, and ensures the stability and user satisfaction of online car-hailing orders.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119090532A_ABST
    Figure CN119090532A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of intelligent traffic, and discloses a price determination method, device and equipment for an online car-hailing fixed-price order, and the method comprises the steps: obtaining first target information, the first target information comprises first order information of a target fixed-price order sent by the user and first supply and demand information of an urban area corresponding to the target fixed-price order at the current moment; inputting the first target information and each price adjustment ratio into a pre-trained first deep learning model, and predicting an order completion rate corresponding to each price adjustment ratio; and according to the price adjustment ratio with the highest order completion rate, adjusting the initial quotation of the target fixed-price order to obtain the final quotation of the target fixed-price order. According to the invention, the order completion rate of each price adjustment ratio is predicted according to the actual order information and the online car-hailing supply and demand information, and then the price adjustment ratio with the highest order completion rate is determined as the price adjustment ratio, so that the driver cancellation rate of online car-hailing fixed-price orders can be reduced.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation technology, and in particular to a price determination method, device and equipment for a fixed-price order for an online car-hailing service. Background Art

[0002] The "one-price" service of online car-hailing means that when passengers book a vehicle, the platform pre-calculates a fixed fee based on the information of the departure and destination. This fee will not change due to changes in the actual driving distance, time or traffic conditions. Passengers can clearly know the total cost of the trip when placing an order. One-price orders are often cheaper than real-time pricing orders. Specifically, the platform will calculate an estimated price based on real-time traffic conditions, estimated driving distance, estimated driving time, etc., and then randomly select a discount rate from multiple optional discount rates to calculate the final quotation. The entire discount rate is generally set in the range of 70% to 95%, so the price of a one-price order is generally lower than that of a normal real-time pricing order. Moreover, since the discount rate is randomly selected, it may not match the current actual situation, which will lead to the problem of high driver cancellation rate. Summary of the invention

[0003] In view of this, the present invention provides a price determination method, device and equipment for online car-hailing fixed-price orders to solve the problem that the pricing discount determination method for fixed-price orders (i.e., fixed-price orders) will lead to a high driver cancellation rate.

[0004] In a first aspect, the present invention provides a method for determining a price for a fixed-price order for an online car-hailing service, the method comprising:

[0005] Acquire first target information, where the first target information includes first order information of a target fixed price order issued by a user, and first supply and demand information of a city area corresponding to the target fixed price order at a current moment;

[0006] Input the first target information and each price adjustment ratio into a pre-trained first deep learning model to predict the order completion rate corresponding to each price adjustment ratio;

[0007] The initial quotation of the target fixed-price order is adjusted according to the price adjustment ratio with the highest order completion rate to obtain the final quotation of the target fixed-price order.

[0008] In an optional implementation, the first order information includes at least one of the following: order time, estimated delivery time, estimated delivery distance and initial quotation; and / or,

[0009] The first supply and demand information of the urban area corresponding to the target fixed-price order at the current moment includes at least one of the following: the number of drivers who have received orders for more than the second preset time within the first preset time before the current moment, the number of online ride-hailing orders issued within the first preset time before the current moment, the number of responses to online ride-hailing orders within the first preset time before the current moment, the waiting time for responses to online ride-hailing orders within the first preset time before the current moment, and the number of online ride-hailing fixed-price orders issued within the first preset time before the current moment; and / or,

[0010] The first target information also includes first statistical information of the city corresponding to the target fixed-price order, and the first statistical information includes at least one of the following: the average number of taxi rides in the city in one day, the average number of drivers in the city in one day, the average number of completed fixed-price orders in the city in one day, the highest fixed price among the fixed-price orders completed in the city, the lowest fixed price among the fixed-price orders completed in the city, the average fixed price among the fixed-price orders completed in the city, the median fixed price among the fixed-price orders completed in the city, and the completion rate of fixed-price orders in the city.

[0011] In an optional implementation manner, before adjusting the initial quotation of the target fixed price order according to the price adjustment ratio with the highest order completion rate, the method further includes:

[0012] Obtain quotes from multiple tenants for a target fixed price order;

[0013] Determine the weight of each tenant based on the number of drivers and completed orders of each tenant;

[0014] The tenants' quotations are weighted averaged using the weight values ​​of each tenant to obtain an initial quotation.

[0015] In an optional implementation, the price adjustment ratio includes a price adjustment ratio greater than 1 and a price adjustment ratio less than 1.

[0016] In an optional implementation, the first deep learning model is a LightGBM model.

[0017] In an optional implementation, before the first target information and each price adjustment ratio are input into a pre-trained first deep learning model and the order completion rate corresponding to each price adjustment ratio is predicted, the method further includes:

[0018] Acquire historical first target information, where the historical first target information includes first order information of historical fixed price orders and first supply and demand information of the city area corresponding to the historical fixed price orders at the current moment;

[0019] Obtain the price adjustment ratio of historical fixed-price orders and the actual order completion rate of historical fixed-price orders;

[0020] Inputting the historical first target information and the price adjustment ratio of the historical fixed-price orders into the first deep learning model to be trained, and predicting the predicted order completion rate of each historical fixed-price order;

[0021] Calculate the loss value based on the predicted order completion rate of each historical fixed-price order and the corresponding actual order completion rate;

[0022] Based on the loss value, parameters of the first deep learning model to be trained are adjusted.

[0023] In an optional implementation, after the first target information and each price adjustment ratio are input into a pre-trained first deep learning model and the order completion rate corresponding to each price adjustment ratio is predicted, the method further includes:

[0024] Acquire second target information, the second target information including second order information of the target fixed price order issued by the user, and second supply and demand information of the city area corresponding to the target fixed price order at the current moment; the second order information includes the price adjustment ratio with the highest order completion rate predicted by the first deep learning model;

[0025] Inputting the second target information into a pre-trained second deep learning model to predict the driver cancellation rate of the target fixed-price order;

[0026] If the driver cancellation rate is greater than or equal to the preset threshold, the order issuance mode of the target fixed price order is determined to be the order broadcast mode, otherwise it is the order dispatch mode.

[0027] In a second aspect, the present invention provides a price determination device for a fixed price order of an online car-hailing service, the device comprising:

[0028] A first acquisition module is used to acquire first target information, where the first target information includes first order information of a target fixed price order issued by a user and first supply and demand information of a city area corresponding to the target fixed price order at a current moment;

[0029] A first prediction module is used to input the first target information and each price adjustment ratio into a pre-trained first deep learning model to predict the order completion rate corresponding to each price adjustment ratio;

[0030] The price adjustment module is used to adjust the initial quotation of the target fixed price order according to the price adjustment ratio with the highest order completion rate to obtain the final quotation of the target fixed price order.

[0031] In a third aspect, the present invention provides an online car-hailing platform, comprising: a memory and a processor, the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the price determination method for an online car-hailing fixed-price order of the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0032] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the price determination method for an online car-hailing fixed-price order of the first aspect or any corresponding embodiment thereof.

[0033] In a fifth aspect, the present invention provides a computer program product, comprising computer instructions for causing a computer to execute the price determination method for an online car-hailing fixed-price order of the first aspect or any corresponding embodiment thereof.

[0034] The price determination method, device and equipment for a fixed-price order for an online ride-hailing vehicle provided in an embodiment of the present invention can predict the order completion rate corresponding to each price adjustment ratio based on the relevant information of the order and the supply and demand information of the online ride-hailing vehicle at the current moment, and then select the price adjustment ratio corresponding to the predicted highest order completion rate, and finally determine the final quotation for the fixed-price order for the online ride-hailing vehicle based on the selected price adjustment ratio, thereby reducing the driver cancellation rate and improving the user (i.e., passenger) experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the related technologies, the drawings required for use in the specific embodiments or the related technical descriptions will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0036] Figure 1 is a flowchart of a method for determining a price for a fixed-price order for an online car-hailing service according to an embodiment of the present invention;

[0037] Figure 2 It is a flowchart of a method for determining a price of a fixed-price order for an online car-hailing service and a method for determining an order issuing method according to an embodiment of the present invention;

[0038] Figure 3 is a structural block diagram of a device for determining a price for a fixed-price order for an online car-hailing service according to an embodiment of the present invention;

[0039] Figure 4 It is a schematic diagram of the hardware structure of the online car-hailing platform according to an embodiment of the present invention. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0041] According to an embodiment of the present invention, an embodiment of a method for determining a price for a fixed-price order for an online ride-hailing service is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of executable computer instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in an order different from that shown here.

[0042] In this embodiment, a price determination method for an online car-hailing fixed price order is provided, which can be used for an online car-hailing platform. The online car-hailing platform can be an online car-hailing platform owned by a single tenant, or an aggregation platform composed of multiple tenants. The tenant is an online car-hailing operating enterprise. The online car-hailing platform can also be called an online car-hailing platform. When a user places an online car-hailing order on the aggregation platform, the user can select a tenant. The pricing methods and standards of different tenants may be different. Figure 1 is a flow chart of a method for determining a price for a fixed-price order for an online car-hailing service according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0043] Step S101, obtaining first target information, the first target information including first order information of a target fixed price order issued by a user (ie, a passenger), and first supply and demand information of a city area corresponding to the target fixed price order at the current moment.

[0044] Among them, fixed price orders are also called fixed price orders. The pricing method of ordinary online car-hailing orders is similar to the metered pricing method of taxis. The starting point and the end point are the same, but due to the delivery distance (different driving routes, etc.) and the delivery time (driving speed and road conditions, etc.), the final price cannot be determined in advance and can only be determined after reaching the destination. However, the price of fixed price orders is fixed and does not change with the delivery distance and delivery time (the starting point and the end point of the order are fixed).

[0045] The demand information at the current moment can be determined based on the demand information of the previous period of time. In addition, it can also be determined in combination with factors such as the current moment (rush hour), weather (including temperature), etc. The demand information at the current moment can also be determined by statistical historical data. For example, the average demand information of the same weather (sunny, rainy) and the same period (the period of the current moment) in the same month of the previous year can be used as the demand information at the current moment. Alternatively, a neural network model such as a multilayer perceptron, a convolutional neural network, and a recurrent neural network can be used to predict the demand information at the current moment, and its input is the current moment, the current month, the current weather, etc.

[0046] Step S102, input the first target information and each price adjustment ratio into a pre-trained first deep learning model to predict the order completion rate corresponding to each price adjustment ratio. Order completion means that after the user (passenger) places an online car-hailing order, the driver is determined, the passenger is delivered, and the passenger completes the payment for the order.

[0047] The first deep learning model mentioned above can be a deep neural network, or a binary classification model based on GBDT (Gradient Boosting Decision Trees) such as XGBoost and CatBoost.

[0048] The price adjustment ratio can be determined according to actual needs, including the level and step of the price adjustment ratio. For example, the price adjustment ratio can be between 80% and 110%, and the step can be 5%, that is, a total of seven levels: 80%, 85%, 90%, 95%, 100%, 105%, 110%; the price adjustment ratio can also be between 70% and 110%, and the step can be 10%, that is, a total of five levels: 70%, 80%, 90%, 100%, 110%.

[0049] Step S103, according to the price adjustment ratio with the highest order completion rate (i.e., the price adjustment ratio corresponding to the highest order completion rate among the order completion rates corresponding to the price adjustment ratios predicted by the first deep learning model), the initial quotation of the target fixed price order is adjusted to obtain the final quotation of the target fixed price order. The final quotation is equal to the product of the initial quotation and the price adjustment ratio.

[0050] The price determination method for a fixed-price order for an online ride-hailing vehicle provided in this embodiment can predict the order completion rate corresponding to each price adjustment ratio based on the relevant information of the order and the supply and demand information of the online ride-hailing vehicle at the current moment, and then select the price adjustment ratio corresponding to the predicted highest order completion rate. Finally, the final quotation for the fixed-price order for the online ride-hailing vehicle is determined based on the selected price adjustment ratio, thereby reducing the driver cancellation rate and improving the user (i.e., passenger) experience.

[0051] In this embodiment, a price determination method for a fixed-price order of an online car-hailing service is provided, which can be used for an online car-hailing service platform. The online car-hailing service platform can be an online car-hailing service platform to which a single tenant belongs, or an aggregation platform composed of multiple tenants. The tenant is an online car-hailing service operating enterprise, and the online car-hailing service platform can also be called an online car-hailing service platform. When a user places an online car-hailing service order on the aggregation platform, the user can select a tenant. The price calculation method and standard of different tenants may be different. The method includes the following steps:

[0052] Step S201, obtaining first target information, the first target information including first order information of a target fixed price order issued by a user, and first supply and demand information of a city area corresponding to the target fixed price order at the current moment.

[0053] In some optional specific implementations, the first order information includes at least one of the following: order time, estimated delivery time, estimated delivery distance, and initial quotation. The estimated delivery time refers to the estimated time it takes for the driver to drive from the order starting point to the order end point, and the estimated delivery distance refers to the estimated distance it takes for the driver to drive from the order starting point to the order end point.

[0054] In some other optional specific implementations, the first supply and demand information of the city area corresponding to the target fixed price order at the current moment includes at least one of the following:

[0055] The number of drivers who have been waiting for orders for more than a second preset time within a first preset time before the current moment; for example, it may be the number of drivers who have been waiting for orders for more than 1 minute (second preset time) within the 5 minutes (first preset time) before the current moment, or it may be the number of drivers who have been waiting for orders for more than 3 minutes (second preset time) within the 5 minutes (first preset time) before the current moment, and it may also include the number of drivers who have been waiting for orders for more than 1 minute (second preset time) within the 5 minutes (first preset time) before the current moment, and the number of drivers who have been waiting for orders for more than 3 minutes (second preset time) within the 5 minutes (first preset time) before the current moment; waiting for orders means that the online car-hailing driver is online and can accept orders, is idle, and is waiting for the platform to dispatch or play orders;

[0056] The number of online ride-hailing orders issued within the first preset time period before the current moment; issuing an order means that the user (passenger) places an order;

[0057] The number of responses to online ride-hailing orders within the first preset time period before the current moment; response refers to the process in which the online ride-hailing platform determines the driver and pushes the result to the user and the driver after the user places an order;

[0058] The response waiting time for online ride-hailing orders within the first preset time before the current moment; the response waiting time is the time between the user placing an order and receiving the driver information pushed by the platform;

[0059] The number of fixed-price online ride-hailing orders issued within the first preset time period before the current moment.

[0060] The first preset durations in the above items may be equal or different.

[0061] Specifically, the division of the urban area can be performed in the following manner: the entire urban space is divided using a hexagonal grid (H3 grid), and the urban space within each hexagonal grid is an urban area.

[0062] In other optional specific implementations, the first target information also includes first statistical information of the city corresponding to the target fixed-price order, and the first statistical information includes at least one of the following: the average number of taxi rides in the city in one day, the average number of drivers in the city in one day, the average number of completed fixed-price orders in the city in one day, the highest fixed price among the fixed-price orders completed in the city, the lowest fixed price among the fixed-price orders completed in the city, the average fixed price of the fixed-price orders completed in the city, the median fixed price among the fixed-price orders completed in the city, and the completion rate of fixed-price orders in the city.

[0063] The above is an example of the first target information and is not intended to be limiting thereof. When implementing the present invention, the specific first target information (including the specific information of the first order information and the first supply and demand information) can be selected or adjusted as needed.

[0064] Step S202: input the first target information and each price adjustment ratio into a pre-trained first deep learning model to predict the order completion rate corresponding to each price adjustment ratio.

[0065] Specifically, the price adjustment ratio includes a price adjustment ratio greater than 1 and a price adjustment ratio less than 1.

[0066] For example, the price adjustment ratios may include 80%, 85%, 90%, 95%, 100%, 105%, and 110%.

[0067] The price of fixed-price orders in related technologies is generally lower than that of normal orders, and fixed-price orders only have a dispatching mode, forcing drivers to accept orders, which is not friendly to drivers and affects the driver's order-taking experience. The price of fixed-price orders in the embodiments of the present invention may increase as the supply and demand relationship changes.

[0068] In some optional specific implementations, the first deep learning model is a LightGBM model. LightGBM (Light Gradient Boosting Machine) is a framework for implementing the GBDT algorithm, which supports efficient parallel training and has the advantages of faster training speed, lower memory consumption, better accuracy, support for distributed and rapid processing of massive data. GBDT (Gradient Boosting Decision Tree) is a long-standing model in machine learning. Its main idea is to use weak classifiers (decision trees) for iterative training to obtain the optimal model. The model has the advantages of good training effect and not easy to overfit.

[0069] In some optional specific implementations, before step S202, that is, inputting the first target information and each price adjustment ratio into a pre-trained first deep learning model, and predicting the order completion rate corresponding to each price adjustment ratio, the method further includes:

[0070] Step 1: Obtain historical first target information, where the historical first target information includes first order information of historical fixed price orders and first supply and demand information of the city area corresponding to the historical fixed price orders at the current moment.

[0071] For specific historical first target information, please refer to the first target information above, which will not be described in detail here. When the first target information includes the first statistical information of the city corresponding to the target fixed price order, the historical first target information also includes the first statistical information of the city corresponding to the historical fixed price order. Specifically, the historical fixed price orders of one or more cities can be collected according to the actual prediction needs of the first deep learning model.

[0072] The information related to the fixed price orders in the historical first target information can be obtained through experiments. Specifically, some cities can be selected for fixed price (fixed price) experiments. The fixed price orders can be divided into multiple experimental groups, and the price adjustment ratios in different experimental groups are different. Then, the relevant information of the fixed price orders used for the experiment is obtained as the historical first target information.

[0073] When obtaining the historical first target information, the training samples to be collected include completed fixed-price orders and uncompleted fixed-price orders, where completed fixed-price orders are positive samples and uncompleted fixed-price orders are negative samples. The ratio of positive samples to negative samples in the sample set can be 1:1. The time of fixed-price orders in the training samples can be adjusted according to factors such as the number of samples. For example, if there are enough orders in recent fixed-price orders that can be used as training samples, then recent orders are selected as samples, otherwise orders with a longer time can be selected as samples.

[0074] Step 2: Obtain the price adjustment ratio of historical fixed-price orders and the actual order completion rate of historical fixed-price orders.

[0075] Step 3: Input the historical first target information and the price adjustment ratio of the historical fixed-price orders into the first deep learning model to be trained, and predict the predicted order completion rate of each historical fixed-price order.

[0076] Step 4: Calculate the loss value based on the predicted order completion rate of each historical fixed-price order and the corresponding actual order completion rate; the actual order completion rate of each historical fixed-price order is 0 (uncompleted order) or 1 (completed order).

[0077] Step 5: Based on the loss value, adjust the parameters of the first deep learning model to be trained.

[0078] Specifically, the loss value can be a cross entropy loss value, a logarithmic loss value, etc. The loss value is used to measure the difference between the model prediction value (i.e., the predicted order completion rate) and the actual label value (i.e., the actual order completion rate). The difference is used to guide the update of the model parameters to minimize the loss value and improve the model prediction accuracy. There are many methods for adjusting the parameters of the first deep learning model, such as gradient descent, momentum method, adaptive learning rate method, etc. Gradient descent is one of the most basic optimization algorithms, which updates the parameters by calculating the gradient of the loss value with respect to the model parameters. The gradient points to the direction where the loss value grows fastest, so the parameters need to be updated in the opposite direction of the gradient to reduce the loss. The momentum method introduces an additional speed term to accelerate the convergence process and reduce oscillation. Momentum can help gradient descent pass through flat areas faster and update smoothly when encountering slope changes. Please refer to the relevant technology for the training process of the first deep learning model, which will not be elaborated here.

[0079] Step S203: adjust the initial quotation of the target fixed price order according to the price adjustment ratio with the highest order completion rate to obtain the final quotation of the target fixed price order. The final quotation is equal to the product of the initial quotation and the price adjustment ratio.

[0080] See also Figure 2 In some optional specific implementations, step S203, i.e., before adjusting the initial quotation of the target fixed price order according to the price adjustment ratio with the highest order completion rate, further includes:

[0081] Step 1: Obtain quotations from multiple tenants for the target fixed-price order. This quotation is the estimated price of the target fixed-price order by each tenant according to its own algorithm. Since different tenants may have different algorithms, their estimated prices for the same order may also be different.

[0082] Step 2: Determine the weight of each tenant based on the number of drivers and the number of completed orders of each tenant;

[0083] Step 3: Use the weight value of each tenant to perform a weighted average on the tenants' quotations to obtain an initial quotation.

[0084] The embodiment of the present invention is applicable to an aggregation platform composed of multiple tenants. When the aggregation platform is for a fixed price order (i.e., a one-price order), it will not display the prices of each tenant separately, but all tenants use a unified price. The unified price is equal to the product of the initial quotation and the price adjustment ratio.

[0085] The aggregation platform aggregates multiple tenants together, and the user's order will be matched with drivers from multiple tenants, which can not only increase the probability of the user's order being answered, but also shorten the response time of the user's order, shorten the distance between the driver who accepts the user's order and the user, and increase the probability of a single driver accepting the order. In addition, when a user places an order on the aggregation platform, the aggregation platform will display all tenants, and the user can choose a tenant according to his or her preferences. Then the user's order will only be sent to the driver of the tenant selected by the user. If the user does not select a tenant, the user's order can be sent to the driver of any tenant by default.

[0086] Different tenants may have different quotes for the same order, but for the aggregation platform, a unified quote is required. Therefore, the platform will consider factors such as the number of drivers and the number of completed orders on the platform to weighted average the quotes of different tenants and balance the market price.

[0087] For other optional specific implementations, please refer to Figure 2 After step S203, the first target information and each price adjustment ratio are input into a pre-trained first deep learning model, and after predicting the order completion rate corresponding to each price adjustment ratio, the method further includes:

[0088] Step S204, obtaining second target information, the second target information including second order information of the target fixed price order issued by the user, and second supply and demand information of the urban area corresponding to the target fixed price order at the current moment; the second order information includes the price adjustment ratio with the highest order completion rate predicted by the first deep learning model.

[0089] Step S205: input the second target information into a pre-trained second deep learning model to predict the driver cancellation rate of the target fixed-price order.

[0090] Among them, the second deep learning model can also be a LightGBM model.

[0091] Step S206: If the driver cancellation rate is greater than or equal to the preset threshold, the order issuance mode of the target fixed price order is determined to be the order broadcast mode, otherwise it is the order dispatch mode.

[0092] For example, if the driver cancellation rate of the target fixed-price order predicted by the second deep learning model is greater than 70% (preset threshold), the order issuance mode of the target fixed-price order is determined to be the order broadcast mode; otherwise, the order issuance mode of the target fixed-price order can be determined to be the order dispatch mode.

[0093] Both the dispatch mode and the broadcast mode are ways for the platform to distribute online ride-hailing orders initiated by users to drivers. The dispatch mode is that after the user (passenger) places an order, the online ride-hailing platform matches the driver and directly dispatches the order to the driver, notifying the driver to pick up the passenger. The broadcast mode is that after the user (passenger) places an order, the online ride-hailing platform broadcasts the online ride-hailing order to multiple drivers who are listening to the order (generally drivers in the city where the order starts or in neighboring cities). The drivers who receive the broadcast can choose whether to accept the order according to their preferences, and the orders are first-come, first-served.

[0094] Many drivers are unwilling to accept cheap or highly discounted orders. If all drivers are assigned orders in a dispatching mode, it will be difficult for the driver to cancel the order, resulting in a poor experience. The embodiment of the present invention predicts the driver cancellation rate before dispatching the order. If it is predicted that the driver does not like the order, the order is distributed in a broadcasting mode, so that idle drivers who want to take the order will take the initiative to grab it, thereby reducing driver dissatisfaction and improving the driver's order-taking experience.

[0095] Specifically, in addition to the price adjustment ratio with the highest order completion rate predicted by the first deep learning model, the second order information may also include at least one of the following: order time, estimated delivery time, estimated delivery distance and initial quotation.

[0096] The second supply and demand information may be the same as the first supply and demand information, or it may be different from the first supply and demand information. For example, the first supply and demand information may not include: the number of orders for fixed-price online ride-hailing orders within the first preset time period before the current moment, while the second supply and demand information needs to include this item. The second supply and demand information may not include: the number of responses to online ride-hailing orders within the first preset time period before the current moment; the waiting time for responses to online ride-hailing orders within the first preset time period before the current moment. The second supply and demand information may also include the number of orders that are waiting at the current moment (no driver has been confirmed, and the user has not canceled).

[0097] In addition, the second target information may also include statistical information of the city corresponding to the target fixed-price order, referred to herein as the second statistical information, which may be different from the first statistical information. For example, the first statistical information may not include the highest fixed price among the fixed-price orders completed in the city, the lowest fixed price among the fixed-price orders completed in the city, the average fixed price among the fixed-price orders completed in the city, the median fixed price among the fixed-price orders completed in the city, and the completion rate of the fixed-price orders in the city. The second statistical information may not include: the average number of taxi rides in the city in one day, the average number of drivers in the city in one day, and the average number of completed orders for fixed-price orders in the city in one day.

[0098] The training samples of the second deep learning model include fixed-price orders canceled by drivers and fixed-price orders that are not canceled by drivers but not completed, wherein fixed-price orders canceled by drivers are positive samples, and fixed-price orders that are not canceled by drivers but not completed (i.e., fixed-price orders that have been confirmed as drivers, or have been answered, but have been canceled for reasons other than driver cancellation) are negative samples. The ratio of positive samples to negative samples in the sample set can be 1:1.

[0099] In this embodiment, a price determination device for a fixed-price order for an online car-hailing service is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and will not be repeated hereafter. As used below, the term "module" may be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation in hardware, or a combination of software and hardware, is also possible and conceivable.

[0100] This embodiment provides a price determination device for a fixed price order of an online car-hailing service. Figure 3 As shown, including:

[0101] The first acquisition module 301 is used to acquire first target information, where the first target information includes first order information of a target fixed price order issued by a user, and first supply and demand information of a city area corresponding to the target fixed price order at the current moment;

[0102] The first prediction module 302 is used to input the first target information and each price adjustment ratio into a pre-trained first deep learning model to predict the order completion rate corresponding to each price adjustment ratio;

[0103] The price adjustment module 303 is used to adjust the initial quotation of the target fixed price order according to the price adjustment ratio with the highest order completion rate to obtain the final quotation of the target fixed price order.

[0104] In some optional implementations, the first order information includes at least one of the following: order time, estimated delivery time, estimated delivery distance, and initial quotation; and / or,

[0105] The first supply and demand information of the urban area corresponding to the target fixed-price order at the current moment includes at least one of the following: the number of drivers who have received orders for more than the second preset time within the first preset time before the current moment, the number of online ride-hailing orders issued within the first preset time before the current moment, the number of responses to online ride-hailing orders within the first preset time before the current moment, the waiting time for responses to online ride-hailing orders within the first preset time before the current moment, and the number of online ride-hailing fixed-price orders issued within the first preset time before the current moment; and / or,

[0106] The first target information also includes first statistical information of the city corresponding to the target fixed-price order, and the first statistical information includes at least one of the following: the average number of taxi rides in the city in one day, the average number of drivers in the city in one day, the average number of completed fixed-price orders in the city in one day, the highest fixed price among the fixed-price orders completed in the city, the lowest fixed price among the fixed-price orders completed in the city, the average fixed price among the fixed-price orders completed in the city, the median fixed price among the fixed-price orders completed in the city, and the completion rate of fixed-price orders in the city.

[0107] In some optional implementations, the price determination device for the online car-hailing fixed price order further includes:

[0108] The quotation acquisition module is used to obtain quotations from multiple tenants for the target fixed price order;

[0109] The weight value acquisition module is used to determine the weight value of each tenant based on the number of drivers and the number of completed orders of each tenant;

[0110] The weighted average module is used to perform weighted average of the tenants' quotations using the weight values ​​of each tenant to obtain an initial quotation.

[0111] In some optional implementations, the price adjustment ratio includes a price adjustment ratio greater than 1 and a price adjustment ratio less than 1.

[0112] In some optional embodiments, the first deep learning model is a LightGBM model.

[0113] In some optional implementations, the price determination device for the online car-hailing fixed price order further includes:

[0114] A second acquisition module is used to acquire historical first target information, where the historical first target information includes first order information of historical fixed price orders and first supply and demand information of the city area corresponding to the historical fixed price orders at the current moment;

[0115] The third acquisition module is used to obtain the price adjustment ratio of historical fixed-price orders and the actual order completion rate of historical fixed-price orders;

[0116] A second prediction module is used to input the historical first target information and the price adjustment ratio of the historical fixed-price orders into the first deep learning model to be trained, and predict the predicted order completion rate of each historical fixed-price order;

[0117] A calculation module, used to calculate the loss value based on the predicted order completion rate of each historical fixed price order and the corresponding actual order completion rate;

[0118] The model parameter adjustment module is used to adjust the parameters of the first deep learning model to be trained based on the loss value.

[0119] In some optional implementations, the price determination device for the online car-hailing fixed price order further includes:

[0120] A fourth acquisition module is used to acquire second target information, where the second target information includes second order information of a target fixed price order issued by a user and second supply and demand information of a city area corresponding to the target fixed price order at the current moment; the second order information includes a price adjustment ratio with the highest order completion rate predicted by the first deep learning model;

[0121] A third prediction module is used to input the second target information into a pre-trained second deep learning model to predict a driver cancellation rate of a target fixed-price order;

[0122] The order issuance mode determination module is used to determine that the order issuance mode of the target fixed-price order is the order broadcast mode if the driver cancellation rate is greater than or equal to a preset threshold, otherwise it is the order dispatch mode.

[0123] The further functional description of each of the above modules is the same as that of the above corresponding embodiments and will not be repeated here.

[0124] The price determination device for the online car-hailing fixed-price order in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above-mentioned functions.

[0125] The embodiment of the present invention also provides an online car-hailing platform, having the above Figure 3 The price determination device for a fixed-price order for an online ride-hailing service is shown.

[0126] See also Figure 4 , Figure 4is a structural diagram of an online car-hailing platform provided by an optional embodiment of the present invention, such as Figure 4 As shown, the online car-hailing platform includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. The various components are connected to each other using different buses for communication, and can be installed on a common motherboard or installed in other ways as needed. The processor can process instructions executed in the online car-hailing platform, including instructions stored in or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Figure 4 A processor 10 is taken as an example.

[0127] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0128] The memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiment.

[0129] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the online car-hailing platform, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the online car-hailing platform via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0130] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0131] The online car-hailing platform also includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 can be connected via a bus or other means. Figure 4The example of connecting through bus is taken in the following.

[0132] The input device 30 can receive input digital or character information, and generate key signal input related to the user settings and function control of the online car-hailing platform, such as a touch screen, a keypad, a mouse, a trackpad, a touchpad, an indicator rod, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (e.g., an LED) and a tactile feedback device (e.g., a vibration motor), etc. The above-mentioned display device includes but is not limited to a liquid crystal display, a light emitting diode, a display and a plasma display. In some optional embodiments, the display device may be a touch screen.

[0133] The online ride-hailing platform also includes a communication interface for the online ride-hailing platform to communicate with other devices or communication networks.

[0134] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0135] A part of the present invention may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the existence of the computer program instruction in a computer-readable medium includes, but is not limited to, a source file, an executable file, an installation package file, etc., and accordingly, the way in which the computer program instruction is executed by the computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium may be any available computer-readable storage medium or communication medium accessible to the computer.

[0136] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A method for determining the price of a fixed-price order for an online car-hailing service, characterized in that: The method comprises: Acquire first target information, where the first target information includes first order information of a target fixed price order issued by a user, and first supply and demand information of a city area corresponding to the target fixed price order at a current moment; Inputting the first target information and each price adjustment ratio into a pre-trained first deep learning model to predict the order completion rate corresponding to each price adjustment ratio; The initial quotation of the target fixed-price order is adjusted according to the price adjustment ratio with the highest order completion rate to obtain a final quotation of the target fixed-price order.

2. The method according to claim 1, characterized in that: The first order information includes at least one of the following: order time, estimated delivery time, estimated delivery distance and initial quotation; and / or, The first supply and demand information of the urban area corresponding to the target fixed-price order at the current moment includes at least one of the following: the number of drivers who have received orders for more than the second preset time within the first preset time before the current moment, the number of online ride-hailing orders issued within the first preset time before the current moment, the number of responses to online ride-hailing orders within the first preset time before the current moment, the response waiting time for online ride-hailing orders within the first preset time before the current moment, and the number of online ride-hailing fixed-price orders issued within the first preset time before the current moment; and / or, The first target information also includes first statistical information of the city corresponding to the target fixed-price order, and the first statistical information includes at least one of the following: the average number of taxi rides in the city in one day, the average number of drivers in the city in one day, the average number of completed fixed-price orders in the city in one day, the highest fixed price among the completed fixed-price orders in the city, the lowest fixed price among the completed fixed-price orders in the city, the average fixed price of the completed fixed-price orders in the city, the median fixed price among the completed fixed-price orders in the city, and the completion rate of fixed-price orders in the city.

3. The method according to claim 1, characterized in that Before adjusting the initial quotation of the target fixed price order according to the price adjustment ratio with the highest order completion rate, the method further includes: Obtaining quotations from multiple tenants for the target fixed price order; Determine the weight value of each tenant according to the number of drivers and the number of completed orders of each tenant; The quotations of the tenants are weighted averaged using the weight values ​​of the tenants to obtain the initial quotation.

4. The method according to claim 1, characterized in that The price adjustment ratio includes a price adjustment ratio greater than 1 and a price adjustment ratio less than 1.

5. The method according to claim 1, characterized in that: The first deep learning model is the LightGBM model.

6. The method according to claim 1, characterized in that Before inputting the first target information and each price adjustment ratio into a pre-trained first deep learning model to predict the order completion rate corresponding to each price adjustment ratio, the method further includes: Acquire historical first target information, wherein the historical first target information includes first order information of historical fixed price orders and first supply and demand information of the city area corresponding to the historical fixed price order at the current moment; Obtaining the price adjustment ratio of the historical fixed-price order and the actual order completion rate of the historical fixed-price order; Inputting the historical first target information and the price adjustment ratio of the historical fixed-price orders into the first deep learning model to be trained, and predicting the predicted order completion rate of each of the historical fixed-price orders; Calculate the loss value based on the predicted order completion rate of each of the historical fixed-price orders and the corresponding actual order completion rate; Based on the loss value, adjust the parameters of the first deep learning model to be trained.

7. The method according to claim 1, characterized in that After inputting the first target information and each price adjustment ratio into a pre-trained first deep learning model and predicting the order completion rate corresponding to each price adjustment ratio, the method further includes: Acquire second target information, where the second target information includes second order information of a target fixed price order issued by a user, and second supply and demand information of a city area corresponding to the target fixed price order at a current moment; the second order information includes a price adjustment ratio with a highest order completion rate predicted by the first deep learning model; Inputting the second target information into a pre-trained second deep learning model to predict the driver cancellation rate of the target fixed-price order; If the driver cancellation rate is greater than or equal to a preset threshold, the order issuance mode of the target fixed-price order is determined to be the order broadcast mode, otherwise it is the order dispatch mode.

8. A price determination device for a fixed price order of an online car-hailing service, characterized in that: The device comprises: A first acquisition module is used to acquire first target information, wherein the first target information includes first order information of a target fixed price order issued by a user and first supply and demand information of a city area corresponding to the target fixed price order at a current moment; A first prediction module is used to input the first target information and each price adjustment ratio into a pre-trained first deep learning model to predict the order completion rate corresponding to each price adjustment ratio; The price adjustment module is used to adjust the initial quotation of the target fixed price order according to the price adjustment ratio with the highest order completion rate to obtain the final quotation of the target fixed price order.

9. An online car-hailing platform, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the price determination method for a fixed-price order for an online car-hailing service as claimed in any one of claims 1 to 7 by executing the computer instructions.

10. A computer program product, characterized in that It includes computer instructions, which are used to enable a computer to execute the price determination method for an online car-hailing fixed-price order as described in any one of claims 1 to 7.