A method and device for dispatching online car-hailing services

Abstract

The present invention provides a method and device for dispatching online car-hailing, and relates to the technical field of online car-hailing business processing. The method comprises: selecting a preset grabbing radius from multiple preset grabbing radiuses according to order demand and related information of online car-hailing capacity in a first preset historical time period, and using the selected preset grabbing radius as the grabbing radius corresponding to a preset future time period in a preset area; combining each order in the current time period in the preset area and multiple target online car-hailing vehicles corresponding to each order to obtain multiple combinations; dividing the multiple combinations corresponding to each order in the current time period to obtain multiple arrays; selecting an array from the multiple arrays, and dispatching orders in the current time period using the selected array. The embodiment of the present invention can preset the grabbing radius that will be selected in all future time periods according to the order demand and the capacity of the online car-hailing, and improves the success rate of dispatching orders compared to the technical solution of dispatching orders with a fixed grabbing radius.

Description

Technical Field

[0001] This invention relates to the field of ride-hailing service processing technology, and in particular to a ride-hailing dispatching method and apparatus. Background Technology

[0002] In ride-hailing services, vehicles are typically searched for within a fixed distance before order matching, such as available ride-hailing vehicles within a 3-kilometer radius of the order. The shorter the search radius, the faster a matching vehicle can be found when there are a sufficient number of available ride-hailing vehicles. This results in a closer connection between the matched vehicle and the user, leading to a better user experience.

[0003] However, when the number of orders exceeds the number of available ride-hailing vehicles, the shorter vehicle-catching radius can negatively impact the success rate of order dispatch because some orders may not have matching vehicles in their vicinity. Conversely, when the number of orders is roughly equal to the number of available ride-hailing vehicles, but the order catching radius is large, the orders may be matched with vehicles that are far away, potentially leading to order cancellation by the user. Summary of the Invention

[0004] This invention provides a method and apparatus for dispatching ride-hailing services. It can preset the vehicle-catching radius to be selected in the future time period based on order demand and ride-hailing capacity. Compared with the technical solution of dispatching orders with a fixed vehicle-catching radius, it improves the success rate of order dispatching.

[0005] In a first aspect, embodiments of the present invention provide a method for dispatching ride-hailing orders, including:

[0006] Based on the order demand and ride-hailing capacity information within the first preset historical time period, select a preset vehicle catching radius from multiple preset vehicle catching radii, and use the selected preset vehicle catching radius as the vehicle catching radius corresponding to the preset future time period within the preset area;

[0007] Each order and its corresponding multiple target ride-hailing vehicles within the preset area for the current time period are combined to obtain multiple combinations; and each order within the current time period is further divided into multiple arrays; wherein, the preset future time period includes the current time period, and the multiple target ride-hailing vehicles corresponding to each order are ride-hailing vehicles available for each order within the vehicle catching radius corresponding to the preset future time period; a combination consists of one order and one target ride-hailing vehicle corresponding to that order;

[0008] Select one array from multiple arrays and use the selected array to dispatch orders within the current time period.

[0009] The above method can select a preset vehicle retrieval radius from multiple preset vehicle retrieval radii based on the order demand and ride-hailing capacity information within a first preset historical time period. It then combines each order within the current time period with the selected vehicle retrieval radius corresponding to each order, and performs order dispatching based on the obtained multiple combinations. This method considers both order demand and ride-hailing capacity, and presets the vehicle retrieval radius that will be selected in future time periods. Compared with technical solutions that dispatch orders using fixed vehicle retrieval radii, this method improves the success rate of order dispatching.

[0010] In one possible implementation, selecting one array from multiple arrays includes:

[0011] Based on the order cancellation rate, order discount, and order business metrics for each combination in multiple arrays, a total metric corresponding to the multiple arrays is determined; wherein, the business metrics are pre-set based on business requirements.

[0012] From multiple arrays, select the array that meets the total index of the first preset requirement.

[0013] The above method can determine the total indicator corresponding to the array by considering the cancellation rate after the order is accepted by the user, the order discount, and the order's business indicators. Based on the total indicator, the appropriate array can be selected for order dispatch. In this way, the selection of the array is determined by the business indicators set in advance according to business needs, which improves the success rate of business.

[0014] In one possible implementation, the overall metrics for the multiple arrays are determined based on the order cancellation rate after user acceptance, order discounts, and business metrics for each combination in the multiple arrays, including:

[0015] Based on the order cancellation rate of each combination in each array, determine the probability that the order is not cancelled after being accepted by the user.

[0016] Multiply the probability of an order not being cancelled after being accepted by the user, the order discount, and the business metrics of the order for each combination in each array, and sum the values ​​obtained by multiplying multiple combinations in each array as the total metric for each array.

[0017] The above method can multiply the probability of a user not canceling an order after it is accepted, the order discount, and the business indicators of the order for each array, and multiply the multiple combinations in each array to get the sum of the values, which is used as the total indicator for each array. In other words, it can obtain the numerical value representing the business demand of each array and the total demand. By taking into account the overall demand, it improves the overall order dispatch success rate.

[0018] In one possible implementation, the order discount for each combination is determined as follows:

[0019] Determine the order cancellation rate after users accept the order for each combination of multiple preset discounts;

[0020] The preset discount corresponding to the cancellation rate of orders that meet the second preset requirement after being accepted will be used as the order discount for each combination.

[0021] The above method can take into account the cancellation rate after the user answers the call and the business indicators pre-set based on business needs to improve the user experience.

[0022] In one possible implementation, the order cancellation rate after being accepted by the user is determined by the following method:

[0023] Extract multiple first order features from the estimated order completion information within this time period;

[0024] By inputting multiple features of the first order into the prediction model, the cancellation rate after the order is accepted by the user is obtained.

[0025] The above method can predict the order cancellation rate based on the predicted completion status of orders in the current time period. This method uses information from the current time period to predict the cancellation rate for the current time period, resulting in higher prediction accuracy.

[0026] In one possible implementation, based on information regarding order demand and ride-hailing capacity within a first preset historical time period, a preset vehicle-catching radius is selected from multiple preset vehicle-catching radii, including:

[0027] Input the relevant information on order demand and ride-hailing capacity within the first preset historical time period into the classification model to obtain the vehicle catching radius corresponding to the preset future time period.

[0028] The above method can input relevant information on order demand and ride-hailing capacity within a first preset historical time period into the classification model. The classification model then predicts the corresponding vehicle catching radius for future time periods from multiple preset vehicle catching radii, resulting in higher selection accuracy.

[0029] In one possible implementation, after dispatching orders within the current time period using the selected array, the method further includes:

[0030] For each order within the current time period, after the order is dispatched, the order discount is displayed to the order user.

[0031] The above method can display order discounts to users after an order is dispatched, thereby retaining users and increasing the order completion rate after they understand the order discounts.

[0032] Secondly, embodiments of the present invention provide a ride-hailing dispatch system, comprising:

[0033] The determination module is used to select a preset vehicle catching radius from multiple preset vehicle catching radii based on the relevant information of order demand and ride-hailing capacity within a first preset historical time period, and to use the selected preset vehicle catching radius as the vehicle catching radius corresponding to a preset future time period within a preset area.

[0034] The combination module is used to combine each order and the multiple target ride-hailing vehicles corresponding to each order within the preset area for the current time period to obtain multiple combinations; and to divide the multiple combinations corresponding to each order within the current time period to obtain multiple arrays; wherein, the preset future time period includes the current time period, the multiple target ride-hailing vehicles corresponding to each order are ride-hailing vehicles available for each order within the vehicle catching radius corresponding to the preset future time period; a combination is one order and one target ride-hailing vehicle corresponding to that order;

[0035] The order dispatch module is used to select one array from multiple arrays and dispatch orders within the current time period using the selected array.

[0036] Thirdly, embodiments of the present invention provide an electronic device, comprising:

[0037] processor;

[0038] A processor for executing computer programs or instructions in the memory, such that the ride-hailing dispatch method as described in any of the first aspects is executed.

[0039] Fourthly, embodiments of the present invention provide a computer-readable storage medium that, when instructions in the storage medium are executed by a processor, enables the processor to execute a ride-hailing dispatch method as described in any of the first aspects.

[0040] Fifthly, embodiments of the present invention provide a computer program product, the computer program product comprising: computer program code, which, when run on a computer, causes the computer to execute the ride-hailing dispatch method described in any of the first aspects.

[0041] Furthermore, the technical effects of any of the implementation methods in the second to fifth aspects can be found in the technical effects of different implementation methods in the first aspect, and will not be repeated here.

[0042] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit the invention. Attached Figure Description

[0043] Figure 1 A flowchart of a ride-hailing order dispatching method provided in an embodiment of the present invention;

[0044] Figure 2 This is a schematic diagram illustrating a target area classified by grid according to an embodiment of the present invention;

[0045] Figure 3 This is a schematic diagram illustrating the situation of multiple orders within a given time period, as provided in an embodiment of the present invention.

[0046] Figure 4 A flowchart illustrating a method for selecting one array from multiple arrays, provided as an embodiment of the present invention;

[0047] Figure 5 A flowchart of another ride-hailing order dispatching method provided in an embodiment of the present invention;

[0048] Figure 6 A schematic diagram of the structure of a ride-hailing dispatching device provided in an embodiment of the present invention;

[0049] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0050] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0051] In the description of the embodiments of this application, unless otherwise stated, "multiple" refers to two or more.

[0052] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature.

[0053] Definitions:

[0054] Vehicle grabbing radius: When matching an order with a ride-hailing service, a circle is drawn with the starting point of the order as the center and the vehicle grabbing radius as the radius. Available ride-hailing services within the range of this circle are found and matched with the order.

[0055] Order dispatch: The order dispatch platform gathers orders at regular intervals, matches these orders with ride-hailing vehicles, and assigns the orders to the matched vehicles. The intervals can be 20 seconds, 30 seconds, 10 seconds, etc.

[0056] However, during the order dispatch process, when the order demand and the availability of ride-hailing vehicles do not match the vehicle-catching radius, order dispatch often fails. For example, if the vehicle-catching radius is relatively large, but the number of orders is comparable to the number of available ride-hailing vehicles, some orders may be matched with ride-hailing vehicles that are far away, and the order users will cancel the orders. If the vehicle-catching radius is relatively small, but the number of orders exceeds the number of available ride-hailing vehicles, some orders may not be matched with ride-hailing vehicles, resulting in order dispatch failure.

[0057] Based on this, the present invention provides a technical solution that can take into account order demand and ride-hailing capacity, predict the vehicle catching radius in the future time period, and dispatch orders according to the vehicle catching radius, thereby avoiding the problem of order dispatch failure caused by the mismatch between order demand, ride-hailing capacity and vehicle catching radius, and improving the success rate of order dispatch.

[0058] The present invention will be described in detail below with reference to the accompanying drawings:

[0059] Combination Figure 1 As shown, this embodiment of the invention provides a method for dispatching ride-hailing orders, including:

[0060] S100: Based on the order demand and ride-hailing capacity information within the first preset historical time period, select a preset vehicle catching radius from multiple preset vehicle catching radii, and use the selected preset vehicle catching radius as the vehicle catching radius corresponding to the preset future time period within the preset area.

[0061] The relevant information on order demand and ride-hailing capacity includes the number of call orders within the grid corresponding to the order origin, the number of call users within the grid corresponding to the order origin, the total value of orders within the grid corresponding to the order origin, the average value of orders within the grid corresponding to the order origin, the number of cruising vehicles and vehicles accepting orders in the surrounding area of ​​the grid corresponding to the order origin, the order user acceptance rate of the grid corresponding to the order origin, the week of the call, whether it is a holiday, and the time slice index.

[0062] The time slice index is used to select the interval for the vehicle capture radius. For example, if the slice is 15 minutes, the value range is 1-96. For instance, the time slice index between 0:00 and 15:00 AM is 1, the time slice index between 15:00 and 30:00 AM is 2, and so on. There are a total of 96 15-minute intervals in a day, and the values ​​are taken sequentially. If the slice is 1 minute, the value range is 1-24*60. There are a total of 24*60 1-minute intervals in a day, and the values ​​are taken sequentially. For instance, the time slice index between 0:00 and 1:00 AM is 1.

[0063] The order starting point corresponds to a grid; for example, the grid is a predefined area divided into hexagonal shapes, such as... Figure 2As shown, the preset area is divided into 37 grids. The area surrounding the grid corresponding to the order starting point can be three rings of grids around the grid corresponding to the order starting point. When the grid corresponding to the order starting point is grid 1, the first ring of grids around grid 1 is grids 2 to 7, the second ring of grids around grid 1 is grids 8 to 19, and the third ring of grids around grid 1 is grids 20 to 37. That is, grids 2 to 37 are the area surrounding the grid corresponding to the order starting point.

[0064] Of course, the above description is only an example. This application does not care about the specific value of the grid corresponding to the order starting point. It only needs to select the vehicle grabbing radius and then dispatch orders to vehicles in the area from the order starting point to the vehicle grabbing radius.

[0065] Multiple preset vehicle detection radii, such as 3 km, 4 km, and 5 km. When executing step 100, you can select one of 3 km, 4 km, and 5 km. For example, if it is 3 km, then 3 km is the vehicle detection radius corresponding to the preset future time period.

[0066] The preset future time period is determined based on the time slice index and the current time. For example, if the time slice index is 15 minutes and the current time is 7 o'clock, then 7 o'clock to 7:15 is the preset future time period.

[0067] S101: Combine each order and the multiple target ride-hailing vehicles corresponding to each order within the preset area for the current time period to obtain multiple combinations; and divide the multiple combinations corresponding to each order within the current time period to obtain multiple arrays.

[0068] The preset future time period includes the current time period; the multiple target ride-hailing vehicles corresponding to each order are ride-hailing vehicles available within the vehicle catching radius corresponding to each order in the preset future time period; a combination is an order and a target ride-hailing vehicle corresponding to that order;

[0069] For example, the time period for this round is 20 seconds, 30 seconds, etc.

[0070] When an order may be assigned to a ride-hailing vehicle, the order and the ride-hailing vehicle are referred to as a combination.

[0071] All combinations can be arbitrarily divided into multiple arrays, and order dispatching can be performed according to each array.

[0072] For example, if there are two orders in a preset area, and each order has two ride-hailing vehicles within its corresponding vehicle-catching radius, there are a total of 2*2=4 combinations. When both orders have ride-hailing vehicles dispatched, there are a total of 4 arrays. When only one order has ride-hailing vehicles dispatched, there are a total of 4 arrays.

[0073] S102: Select one array from multiple arrays and use the selected array to dispatch orders within the current time period.

[0074] For example, combining Figure 3 As shown, the orders within this time period include orders A to D. Order A has a dispatch range P1 within the vehicle retrieval radius corresponding to the preset future time period, and the available ride-hailing vehicles within dispatch range P1 are ride-hailing vehicles a-1 to ride-hailing vehicles a-4. Order B has a dispatch range P2 within the vehicle retrieval radius corresponding to the preset future time period, and the available ride-hailing vehicles within dispatch range P2 are ride-hailing vehicles a-3, a-5 to a-7. Order C has a dispatch range P3 within the vehicle retrieval radius corresponding to the preset future time period, and the available ride-hailing vehicles within dispatch range P3 are ride-hailing vehicles a-6 to a-10. Order D has a dispatch range P4 within the vehicle retrieval radius corresponding to the preset future time period, and the available ride-hailing vehicles within dispatch range P4 are ride-hailing vehicles a-4, a-10 to a-12.

[0075] Combining order A with ride-hailing vehicle a-1 yields combination 1; combining order A with ride-hailing vehicle a-2 yields combination 2; combining order A with ride-hailing vehicle a-3 yields combination 3; combining order A with ride-hailing vehicle a-4 yields combination 4; combining order B with ride-hailing vehicle a-3 yields combination 5; combining order B with ride-hailing vehicle a-5 yields combination 6; combining order B with ride-hailing vehicle a-6 yields combination 7; combining order B with ride-hailing vehicle a-7 yields combination 8; combining order C with ride-hailing vehicle a-4... Combining a-6 yields combination 9. Combining order C with ride-hailing vehicle a-7 yields combination 10. Combining order C with ride-hailing vehicle a-9 yields combination 7. Combining order C with ride-hailing vehicle a-10 yields combination 12. Combining order D with ride-hailing vehicle a-4 yields combination 13. Combining order D with ride-hailing vehicle a-10 yields combination 14. Combining order D with ride-hailing vehicle a-7 yields combination 15. Combining order D with ride-hailing vehicle a-12 yields combination 16. A total of 16 combinations are obtained.

[0076] Then, the 16 combinations are further divided into multiple arrays. For example, combination 1, combination 5, combination 9, and combination 13 are in one array; combination 2, combination 6, combination 10, and combination 14 are in another array; combination 3, combination 7, combination 11, and combination 15 are in another array; combination 4, combination 8, combination 12, and combination 16 are in another array; combination 1 and combination 5 are in another array; and combination 1 and combination 9 are in another array.

[0077] When an array is selected from the above arrays, for example, combination 2, combination 6, combination 10, and combination 14, then in this round of time period, ride-hailing vehicle a-2 will be dispatched to order A, ride-hailing vehicle a-5 will be dispatched to order B, ride-hailing vehicle a-7 will be dispatched to order C, and ride-hailing vehicle a-10 will be dispatched to order D.

[0078] In some embodiments, based on information related to order demand and ride-hailing capacity within a first preset historical time period, a preset vehicle retrieval radius is selected from multiple preset vehicle retrieval radii, including:

[0079] Input the relevant information on order demand and ride-hailing capacity within the first preset historical time period into the classification model to obtain the vehicle catching radius corresponding to the preset future time period.

[0080] The first preset historical time period can be the historical time period of the day the order was placed, and / or the historical time period before the day the order was placed, where the weekday and time period are the same.

[0081] The historical time period of the day the order was placed, for example, 1 minute or 5 minutes before the time the order was received;

[0082] The first preset historical time period is the same as the day of the week and the time period before the order was placed, for example, the same time period of the same period last week. For example, if the preset future time period is 8:00 AM to 8:15 AM on July 12, 2024, then the first preset historical time period can be 8:00 AM to 8:15 AM on July 5, 2024.

[0083] The relevant information on order demand and ride-hailing capacity for the historical time period on the day the order was placed is as follows:

[0084] 1. The number of call orders in the grid corresponding to the order start point 1 minute or 5 minutes before the order is received, the number of call users in the grid corresponding to the order start point, the total order value in the grid corresponding to the order start point, and the average order value in the grid corresponding to the order start point;

[0085] 2. The number of cruising vehicles and order-receiving vehicles in the area surrounding the grid corresponding to the order start point 1 minute or 5 minutes before the order is received;

[0086] 3. The order acceptance rate of the grid corresponding to the order start point 1 minute or 5 minutes before the order is received;

[0087] 4. What day of the week was the order received? Was it a holiday? Time slice index.

[0088] The relevant information on order demand and ride-hailing capacity for historical time periods prior to the order date, where the weekday and time period are the same, is as follows:

[0089] 1. The number of call orders, the number of call users, the total order value, and the average order value within the grid corresponding to the order starting point from 8:00 AM to 8:15 AM on July 5, 2024.

[0090] 2. The number of cruising vehicles and order-receiving vehicles in the area surrounding the grid corresponding to the order origin between 8:00 AM and 8:15 AM on July 5, 2024;

[0091] 3. Order acceptance rate for the grid corresponding to the order starting point between 8:00 AM and 8:15 AM on July 5, 2024;

[0092] 4. What day of the week is 8:00 AM to 8:15 AM on July 5, 2024? Is it a public holiday? Time slice index.

[0093] The relevant information on order demand and ride-hailing capacity for the historical time period of the day the order occurred, and the historical time period before the day the order occurred, with the same week and time period, are input into the classification model. The classification model analyzes the above features, gives the probability of multiple preset vehicle grabbing radii, and outputs the preset vehicle grabbing radius with the highest probability, which is the selected preset vehicle grabbing radius.

[0094] Before using a classification model, it is necessary to train the classification model using a training set. During each training process, a sample from the training set is input into the classification model to obtain a prediction result. Based on the prediction result and the optimal result of the sample, the parameters in the classification model are adjusted.

[0095] The methods for obtaining samples and optimal results from the training set are as follows:

[0096] Orders are dispatched to each region according to a time slice index, rotating through multiple preset pickup radii at preset time intervals. The average transaction value of successful orders in each region under different preset pickup radii is used to identify the maximum average transaction value, and the preset pickup radius corresponding to this maximum average transaction value is considered the optimal result. The relevant information from the previous time slice index and the optimal result of the current time slice index are combined into a single sample, and this process is repeated to obtain multiple samples, forming a training set.

[0097] The classification model can be a commonly used machine learning algorithm such as XGBM or LightGBM.

[0098] In some embodiments, one array is selected from multiple arrays and combined. Figure 4 As shown, it includes:

[0099] S400: Determine the total metrics corresponding to the multiple arrays based on the order cancellation rate, order discount, and business metrics of each combination in the multiple arrays.

[0100] Among them, the business metrics are pre-set based on business needs;

[0101] The order cancellation rate of a group is the probability that, after a ride-hailing vehicle from that group is dispatched to a user in that group, the user in that group will cancel the order if they are not satisfied with the ride-hailing vehicle.

[0102] The business metric for an order can be its value, which is the cost of the order before discounts.

[0103] S401: Select the array that satisfies the total index of the first preset requirement from multiple arrays.

[0104] The first preset requirement can be the maximum value or a value greater than the preset value;

[0105] For example, from multiple arrays, the array with the largest total index is selected as the array to choose from.

[0106] Alternatively, select the array from multiple arrays whose total index exceeds a preset value.

[0107] When the number of arrays with total indicators greater than the preset value is greater than 1, an array can be randomly selected from the arrays greater than 1 as the final selected array, and orders will be dispatched according to the array within the preset time period of this round.

[0108] In some embodiments, the overall metrics corresponding to the multiple arrays are determined based on the order cancellation rate after user acceptance, order discount, and order business metrics for each combination in the multiple arrays, including:

[0109] Based on the order cancellation rate of each combination in each array, determine the probability that the order is not cancelled after being accepted by the user.

[0110] Multiply the probability of an order not being cancelled after being accepted by the user, the order discount, and the business metrics of the order for each combination in each array, and sum the values ​​obtained by multiplying multiple combinations in each array as the total metric for each array.

[0111] In detail, the probability that a user will not cancel an order after it has been accepted is determined by the following formula:

[0112] prob Ni =1-prob i ;

[0113] Among them, prob Ni Let prob be the probability that user i does not cancel an order after being picked up by a ride-hailing service. i The cancellation rate of order user i after being picked up by a ride-hailing vehicle.

[0114] The overall index corresponding to the array is determined using the following formula:

[0115]

[0116] Among them, IND S For an array of total metrics, flow i For order user i's business metrics, discount i Discount for order user i.

[0117] by Figure 3 For example, when the array consists of combination 1, combination 5, combination 9, and combination 13, the overall index IND corresponding to this array is determined using the following formula. S

[0118] IND S =prob NA *flow A *discount A

[0119] +prob NB *flow B *discount B

[0120] +prob NC *flow C *discount C

[0121] +prob ND *flow D *discount D

[0122] The probability that user A does not cancel order A after accepting the order. NA The transaction flow of ride-hailing vehicle a-1 when it completes order A. A and order discount A Multiply the parameters for order A above; the probability that order B is not canceled after being accepted by the user. NB The order flow when ride-hailing vehicle a-5 completes order B. B and order discount B Multiply the parameters for order B mentioned above; the probability that order C is not canceled after being accepted by the user. NC When ride-hailing vehicle a-6 completes order C, the order flow is... C and order discount C Multiply the parameters for order C mentioned above; the probability that the user does not cancel order D after accepting the call is prob.ND When ride-hailing vehicle a-4 completes order D, the order flow is... D and order discount D Multiply the parameters for order D mentioned above; multiply the four combinations mentioned above and then add them together to obtain the total index IND corresponding to this array. S .

[0123] Similarly, the total indicators for other arrays are calculated using the same method.

[0124] In some embodiments, order discounts are determined in the following ways:

[0125] Determine the order cancellation rate after users accept the order for each combination of multiple preset discounts;

[0126] Based on the order cancellation rate after the multiple preset discounts for each combination are accepted, determine the expected indicators corresponding to the multiple preset discounts for each combination;

[0127] The preset discount corresponding to the expected index that meets the second preset requirement will be used as the order discount for each combination.

[0128] Among these, the business metric can be the order cost. This refers to the estimated cost of the order when the order dispatch platform generates the order, and the order discount is an additional discount on top of this order cost.

[0129] For example, the second preset requirement is the maximum value, the preset discount includes three discounts, the cancellation rate of the order after the user accepts the order under each combination of the three discounts is determined, the maximum value of the cancellation rate of the order after the user accepts the order under the three discounts is found, and the discount corresponding to the maximum value of the cancellation rate of the order after the user accepts the order is used as the order discount.

[0130] For example, the preset discounts are 5%, 10%, 20%, etc.

[0131] For combination 1, calculate the cancellation rate of order A and ride-hailing vehicle a-1 after the order is accepted by the user under multiple preset discounts. Based on the cancellation rate of order A and ride-hailing vehicle a-1 under multiple preset discounts, determine the total indicator of order A and ride-hailing vehicle a-1 under multiple preset discounts. Select the discount corresponding to the largest total indicator as the order discount that the platform will distribute when order A hailes ride-hailing vehicle a-1.

[0132] It should be noted that this step is essentially determining the discount for passengers, and this discount can be configured. Specifically, different discounts can be configured, such as 5%, 9%, or 8% off; alternatively, it can be configured such that only orders over 30 yuan receive 5%, 9%, or 8% off, etc. This invention does not impose specific limitations on this.

[0133] It is important to note that, in combination Figure 3As shown, for ride-hailing vehicle a-3 at the current stage, the cancellation rate after connecting with order A and order B will be calculated simultaneously. When dispatching orders, the algorithm will use constraints to ensure that the ride-hailing vehicle will not be dispatched to two orders repeatedly. That is, there is only one vehicle for one order, and one vehicle can only serve one order.

[0134] In some embodiments, the order cancellation rate after user acceptance is determined in the following ways:

[0135] The estimated order completion information for the current time period and the historical order completion information for users in the second preset historical time period are input into the prediction model to obtain the order cancellation rate after the user accepts the order.

[0136] The second preset historical time period can be a preset number of days prior to the day the order was placed.

[0137] For example, 14 or 30 days before the order date; if the order date is August 25, 2024, then the second preset historical time period is from August 7, 2024 to August 24, 2024.

[0138] The estimated order completion information for this time period is as follows:

[0139] 1. Order discounts and order discount amounts;

[0140] 2. The hour in which the order was received, the time slice in which the order was received, and the hour in which the order is expected to be completed; for example, if the time slice is 15 minutes, the order was received at 8:00 AM in August 2024, and the time slice is from 8:00 AM to 8:15 AM in August 2024.

[0141] 3. Call to pick-up interval, estimated pick-up mileage, estimated pick-up time, pick-up speed (pick-up mileage / pick-up time), drop-off mileage, estimated drop-off arrival time, pick-up and drop-off mileage, time ratio (pick-up distance / drop-off distance, considering that the shorter the drop-off mileage, the more sensitive users are to the pick-up distance), and estimated cost;

[0142] 4. Within the grid where the order originates, the number of calls, the number of answered calls (rate), and the number of canceled calls (rate) within that grid over the past one and five minutes.

[0143] The historical order completion information for users within the second preset historical time period is as follows:

[0144] The number of times a user completes an order, the number of calls made, and the number or percentage of orders canceled after being answered in the past 14 and 30 days;

[0145] The number or percentage of orders that users have proactively cancelled in the past 14 or 30 days after the order was received;

[0146] The average pick-up mileage of users who placed the order in the past 14 days, and the pick-up time of the current order / the average pick-up time of users who placed the order in the past 14 days; (the larger this value is, the greater the deviation from the user's average data, and the more likely the user will cancel the order).

[0147] The estimated order completion information within the current time period and the historical order completion information of the order user within the second preset historical time period are input into the prediction model. The prediction model analyzes the probability of whether the order will be cancelled after being accepted based on the user's current order status and previous order status, and gives a specific value of the cancellation rate after the order user accepts the order.

[0148] Before using the prediction model, it is necessary to train the prediction model using a training set. During each training process, a sample from the training set is input into the prediction model to obtain the prediction result. Based on the prediction result and the actual result of the sample, the parameters in the prediction model are adjusted.

[0149] The methods for obtaining samples in the training set and actual results are as follows:

[0150] Initially, data can be collected through randomized production experiments, i.e., randomly giving users discounts to observe their order completion status and record whether the user canceled the pre-order (1 for cancellation, 0 for non-cancellation). Then, collect the historical order completion information of the user to form a sample. Use the recorded cancellation status of the pre-order as the actual result of the sample, and so on, collecting multiple samples and actual results.

[0151] The collected samples and actual results are used to train the prediction model.

[0152] The prediction model can be a binary classification model, such as commonly used machine learning algorithms like XGB and LightGBM.

[0153] In actual calculations, the prediction model provides the probability of an order being cancelled after it is accepted, and the probability of an order not being cancelled after it is accepted. The prediction model outputs the order cancellation rate after the user accepts the order.

[0154] Combination Figure 5 As shown, an embodiment of the present invention provides a dispatching method, including:

[0155] S500: Input the relevant information of order demand and ride-hailing capacity within the first preset historical time period into the classification model to obtain the vehicle catching radius corresponding to the preset future time period within the preset area;

[0156] S501: Combine each order within the preset area and the multiple target ride-hailing vehicles corresponding to each order in the current time period to obtain multiple combinations; and divide the multiple combinations corresponding to each order in the current time period to obtain multiple arrays;

[0157] S502: Input the estimated order completion information within the current time period and the historical order completion information of the order users within the second preset historical time period into the prediction model to obtain the order cancellation rate after the order user accepts the order.

[0158] S503: Based on the order cancellation rate of each combination in each array, determine the probability that the order is not cancelled after being accepted by the user;

[0159] S504: Multiply the probability of a user not canceling an order after it has been accepted, the order discount, and the order cost for each combination in each array, and sum the values ​​obtained by multiplying multiple combinations in each array as the total indicator for each array;

[0160] S505: Select the array with the largest total index from multiple arrays, and use the selected array to dispatch orders within the current time period.

[0161] The above-described solution in this embodiment of the invention fully considers the current supply and demand situation and determines the optimal dispatch radius under the current time and space. Furthermore, it mines user profiles from historical behavioral data to quantitatively characterize the probability of users canceling pre-accepted orders. Simultaneously, by providing users with price subsidies, it encourages users to patiently wait for drivers, increases order completion rates, and boosts the platform's overall revenue, achieving a win-win-win situation for users, drivers, and the platform.

[0162] In one possible implementation, after dispatching orders within the current time period using a selected array, the method further includes:

[0163] For each order within this round, after the order is dispatched, the order discount is displayed to the order user.

[0164] In detail, after an order is dispatched, it is automatically generated on the order user's side. At this point, an order discount is displayed to the order user. This discount is added to the order user's account after the order fee is generated. This way, considering the dynamically changing vehicle pickup radius, if the order user is matched with a long-distance ride-hailing vehicle, the discount can be supplemented by additional fees, thus improving the order completion rate.

[0165] like Figure 6 As shown, the present invention also provides a ride-hailing dispatching device, comprising:

[0166] The determination module 600 is used to select a preset vehicle catching radius from multiple preset vehicle catching radii based on the relevant information of order demand and ride-hailing capacity within a first preset historical time period, and to use the selected preset vehicle catching radius as the vehicle catching radius corresponding to a preset future time period within a preset area.

[0167] The combination module 601 is used to combine each order and the multiple target ride-hailing vehicles corresponding to each order within the preset area for the current time period to obtain multiple combinations; and to divide the multiple combinations corresponding to each order within the current time period to obtain multiple arrays; wherein, the preset future time period includes the current time period, the multiple target ride-hailing vehicles corresponding to each order are ride-hailing vehicles available for each order within the vehicle catching radius corresponding to the preset future time period; a combination is one order and one target ride-hailing vehicle corresponding to that order;

[0168] The order dispatch module 602 is used to select one array from multiple arrays and dispatch orders within the current time period using the selected array.

[0169] Optional, the dispatch module 602 is specifically used for:

[0170] Based on the order cancellation rate, order discount, and order business metrics for each combination in multiple arrays, a total metric corresponding to the multiple arrays is determined; wherein, the business metrics are pre-set based on business requirements.

[0171] From multiple arrays, select the array that meets the total index of the first preset requirement.

[0172] Optional, the dispatch module 602 is specifically used for:

[0173] Based on the order cancellation rate of each combination in each array, determine the probability that the order is not cancelled after being accepted by the user.

[0174] Multiply the probability of an order not being cancelled after being accepted by the user, the order discount, and the business metrics of the order for each combination in each array, and sum the values ​​obtained by multiplying multiple combinations in each array as the total metric for each array.

[0175] Optionally, the dispatch module 602 is also used for:

[0176] Based on the correlation between the cancellation rate after the user answers the call, business metrics, and discounts, the discounts corresponding to the cancellation rate after the user answers the call and the business metrics of the order are determined, and the discounts corresponding to the cancellation rate after the user answers the call and the business metrics of the order are used as the order discounts.

[0177] Optionally, the dispatch module 602 is also used for:

[0178] The estimated order completion information for the current time period and the historical order completion information for users in the second preset historical time period are input into the prediction model to obtain the order cancellation rate after the user accepts the order.

[0179] Optionally, module 600 is defined, specifically for:

[0180] Input the relevant information on order demand and ride-hailing capacity within the first preset historical time period into the classification model to obtain the vehicle catching radius corresponding to the preset future time period.

[0181] Optionally, the device further includes: a display module, which is connected to the dispatch module;

[0182] The display module is used to display the order discount to the order user after the order is dispatched for each order within the current time period.

[0183] In addition, combined Figures 1-6 The ride-hailing dispatch method and apparatus described in this embodiment of the invention can be implemented by electronic devices.

[0184] Electronic devices, including: processors;

[0185] Memory used to store the processor's executable instructions;

[0186] The processor is configured to execute the instructions to implement the ride-hailing dispatch method as described above.

[0187] Based on the above introduction, an example is proposed. Figure 7 The structure of electronic devices.

[0188] The electronic device may include a processor 710 and a memory 720 storing computer program instructions.

[0189] Specifically, the processor 710 may include a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement embodiments of the present invention.

[0190] Memory 720 may include mass storage for data or instructions. For example, and not limitingly, memory 720 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where suitable, memory 720 may include removable or non-removable (or fixed) media. Where suitable, memory 720 may be internal or external to a data processing device. In a particular embodiment, memory 720 is a non-volatile solid-state memory. In a particular embodiment, memory 720 includes read-only memory (ROM). Where suitable, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically rewritable ROM (EAROM), or flash memory, or a combination of two or more of these.

[0191] The processor 710 implements any of the methods for performing tasks in the above embodiments by reading and executing computer program instructions stored in the memory 720.

[0192] In one example, the electronic device may also include a communication interface 730 and a bus 740. Wherein, as... Figure 7 As shown, the processor 710, memory 720, and communication interface 730 are connected via bus 740 and communicate with each other.

[0193] The communication interface 730 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of the present invention.

[0194] Bus 740 includes hardware, software, or both, that couples components of an electronic device together. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 740 may include one or more buses. Although specific buses are described and illustrated in embodiments of the invention, the invention contemplates any suitable bus or interconnect.

[0195] The electronic device can execute the ride-hailing dispatch method in this embodiment of the invention based on the received task, thereby achieving a combination of... Figures 1-6 The description describes the dispatching methods and devices for ride-hailing services.

[0196] In addition, in conjunction with the electronic devices in the above embodiments, the present invention can provide a storage medium that, when the instructions in the storage medium are executed by the processor of the electronic device, enables the electronic device to execute the ride-hailing dispatch method as described in any of the above embodiments.

[0197] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0198] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0199] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0200] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0201] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for dispatching online car-hailing services, characterized in that: include: According to the order demand and the related information of the online car-hailing capacity in the first preset historical time period, a preset car grabbing radius is selected from multiple preset car grabbing radiuses, and the selected preset car grabbing radius is used as the car grabbing radius corresponding to the preset future time period in the preset area; Combining each order in the current time period within the preset area and multiple target online-hailing vehicles corresponding to each order to obtain multiple combinations; and dividing the multiple combinations corresponding to each order in the current time period to obtain multiple arrays; wherein the preset future time period includes the current time period, and the multiple target online-hailing vehicles corresponding to each order are online-hailing vehicles available within the capture radius corresponding to each order in the preset future time period; and one combination is one order and one target online-hailing vehicle corresponding to the order; An array is selected from multiple arrays, and the selected array is used to dispatch orders within the current time period.

2. The method for dispatching online car-hailing services according to claim 1, characterized in that: Select an array from multiple arrays, including: Determine the total index corresponding to the multiple arrays according to the order cancellation rate after the user picks up the order, the order discount, and the business index of the order for each combination in the multiple arrays; wherein the business index is pre-set based on business needs; From the multiple arrays, an array with a total index that meets the first preset requirement is selected as the array.

3. The method for dispatching online car-hailing services according to claim 2, characterized in that: According to the order cancellation rate after user pick-up, order discount, and order business indicators of each combination in multiple arrays, the total indicators corresponding to the multiple arrays are determined, including: According to the cancellation rate of the order users of each combination in each array after they pick up the order, determine the probability that the order users of each combination in each array do not cancel after they pick up the order; Multiply the probability of not canceling the order after the user picks up the order, the order discount, and the business indicators of the order for each combination in each array, and take the sum of the values ​​obtained by multiplying multiple combinations in each array as the total indicator corresponding to each array.

4. The method for dispatching online car-hailing services according to claim 2, characterized in that: The order discount for each combination is determined as follows: Determine the cancellation rate of orders received by users corresponding to multiple preset discounts for each combination; The preset discount corresponding to the cancellation rate after the order is picked up by the user that meets the second preset requirement is used as the order discount for each combination.

5. The method for dispatching online car-hailing services according to claim 2, characterized in that: The cancellation rate of orders after being picked up is determined by: The estimated order completion information within the current time period and the historical order completion information of the order users within the second preset historical time period are input into the prediction model to obtain the cancellation rate of the order users after picking up the order.

6. The method for dispatching online car-hailing services according to claim 1, characterized in that: According to the order demand and the related information of the online car-hailing capacity in the first preset historical time period, a preset car grabbing radius is selected from multiple preset car grabbing radiuses, including: The order demand and online car-hailing capacity information within the first preset historical time period are input into the classification model to obtain the vehicle capture radius corresponding to the preset future time period.

7. The method for dispatching online car-hailing services according to any one of claims 2 to 6, characterized in that: After dispatching orders within the current time period using the selected array, the method further includes: For each order within the current round, after the order is dispatched, the order discount is displayed to the order user.

8. A device for dispatching online car-hailing vehicles, characterized in that: include: A determination module, configured to select a preset grabbing radius from a plurality of preset grabbing radiuses according to order demand and information related to online car-hailing capacity within a first preset historical time period, and use the selected preset grabbing radius as the grabbing radius corresponding to a preset future time period within a preset area; A combination module, used for combining each order in the preset area during the current time period and multiple target online-hailing vehicles corresponding to each order to obtain multiple combinations; and dividing the multiple combinations corresponding to each order in the current time period to obtain multiple arrays; wherein the preset future time period includes the current time period, and the multiple target online-hailing vehicles corresponding to each order are online-hailing vehicles available within the capture radius corresponding to each order in the preset future time period; and one combination is one order and one target online-hailing vehicle corresponding to the order; The dispatching module is used to select an array from multiple arrays and use the selected array to dispatch orders within the current time period.

9. An electronic device, characterized in that: include: Memory, used to store computer programs or instructions; A processor is used to execute the computer program or instructions in the memory so that the online car-hailing dispatching method as described in any one of claims 1-6 is executed.

10. A computer-readable storage medium, characterized in that: When the instructions in the storage medium are executed by the processor, the processor is enabled to execute the online car-hailing dispatching method as described in any one of claims 1-6.

Images