Boarding position recommendation method and device, electronic equipment and storage medium
By obtaining the pick-up cost and passenger willingness probability of candidate boarding locations, the boarding location with the greatest cost savings is screened out, solving the problem of inconsistent costs at different boarding points, maximizing platform revenue and improving driver and passenger experience.
Patent Information
- Application Number
- CN202511040786.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-28
- Publication Date
- 2025-10-17
AI Technical Summary
The driver's pick-up costs and passengers' walking costs at different pick-up points are inconsistent, resulting in a lower platform profitability. There is an urgent need for a method to reduce the pick-up costs and increase the platform profitability.
By obtaining the pick-up costs of candidate boarding locations, combined with the ticket compensation rate and passenger willingness probability, the candidate boarding locations with the greatest cost savings are screened out and recommended to passengers.
Effectively reduce the cost of picking up passengers, ensure that cost savings are greater than the voucher supplement, maximize platform revenue, and improve the driver and passenger experience.
Smart Images
Figure CN120804442A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of online car-hailing, and in particular to a method and device for recommending a pickup location, an electronic device, and a storage medium. BACKGROUND
[0002] Online car-hailing is a service that enables passengers to place orders online, drivers to accept orders, and online payments through a mobile APP (application), which has greatly changed the traditional mode of transportation.
[0003] In real life, the pickup point in some areas causes the driver to detour and get stuck in traffic when picking up passengers, such as a pickup point in a community, a pickup point in a small alley, a pickup point in a congested road, etc. In this case, the cost of the driver's pickup is very high, and the platform's revenue rate is greatly reduced, especially during rush hours, i.e., the cost of the driver's pickup at different pickup points is different, and the walking cost of the passenger to different pickup points is also different.
[0004] How to reduce the cost of the driver's pickup and improve the platform's revenue rate is a problem to be solved. SUMMARY
[0005] The technical problem to be solved by the embodiments of the present application is to provide a method and device for recommending a pickup location, an electronic device, and a storage medium, so as to effectively reduce the cost of pickup, while ensuring that the saved pickup cost is greater than the additional coupon compensation, so as to maximize the platform's revenue and improve the experience of drivers and passengers.
[0006] In a first aspect, the embodiments of the present application provide a method for recommending a pickup location, which comprises:
[0007] In a case where it is determined that the original pickup location corresponding to the online car-hailing order initiated by the passenger meets the pickup location replacement condition, obtaining the pickup cost of a plurality of candidate pickup locations;
[0008] According to the coupon compensation rate and the distance between the passenger and the plurality of candidate pickup locations, determining the individual willingness probability of the passenger to go to the plurality of candidate pickup locations;
[0009] According to the pickup cost, the individual willingness probability, and the original pickup cost and the coupon price of the online car-hailing order, determining the saved cost of the plurality of candidate pickup locations under different coupon compensation rates;
[0010] From the plurality of candidate pickup locations, the candidate pickup location with the maximum saved cost is selected as the replacement pickup location, and the replacement pickup location is recommended to the passenger.
[0011] In a second aspect, the embodiments of the present application provide a device for recommending a pickup location, which comprises:
[0012] The pickup cost obtaining module is configured to, in a case where it is determined that the original pickup location corresponding to the pickup order initiated by the passenger satisfies the pickup location replacement condition, obtain pickup costs of a plurality of candidate pickup locations;
[0013] The personal willingness determining module is configured to determine a personal willingness probability of the passenger to go to the plurality of candidate pickup locations according to the coupon subsidy rate and distances between the passenger and the plurality of candidate pickup locations.
[0014] The cost saving determining module is configured to determine cost savings of the plurality of candidate pickup locations under different coupon subsidy rates according to the pickup costs, the personal willingness probabilities, and an original pickup cost and a coupon price of the pickup order.
[0015] The replacement location recommending module is configured to select, from the plurality of candidate pickup locations, a candidate pickup location with the largest cost saving as a replacement pickup location, and recommend the replacement pickup location to the passenger.
[0016] In a third aspect, an electronic device is provided, including:
[0017] A processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor implements the pickup location recommending method according to any one of the preceding aspects when executing the program.
[0018] In a fourth aspect, a computer readable storage medium is provided, and when instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform the pickup location recommending method according to any one of the preceding aspects.
[0019] Compared with the prior art, the embodiments of the present application have the following advantages:
[0020] In the embodiments of the present application, in a case where it is determined that the original pickup location corresponding to the pickup order initiated by the passenger satisfies the pickup location replacement condition, pickup costs of a plurality of candidate pickup locations are obtained. A personal willingness probability of the passenger to go to the plurality of candidate pickup locations is determined according to a coupon subsidy rate and distances between the passenger and the plurality of candidate pickup locations. Cost savings corresponding to the plurality of candidate pickup locations are determined according to the pickup costs, the personal willingness probabilities, and an original pickup cost and a coupon price of the pickup order. A candidate pickup location with the largest cost saving is selected from the plurality of candidate pickup locations as a replacement pickup location, and the replacement pickup location is recommended to the passenger. The embodiments of the present application enable the passenger to go to a pickup location with a small pickup cost by means of a coupon subsidy method, thereby reducing a pickup cost (a driver reward and other fees caused by the pickup cost), while ensuring that the saved pickup cost is greater than a coupon subsidy amount paid more, so as to achieve the goal of maximizing platform revenue and improve the driver and passenger experience.
[0021] It should be understood that the foregoing general description and the following detailed description are only exemplary and explanatory and are not restrictive of the application. BRIEF DESCRIPTION OF DRAWINGS
[0022] Figure 1 A step flow chart of a pickup location recommendation method provided by an embodiment of the application;
[0023] Figure 2 A step flow chart of a position replacement condition determination method provided by an embodiment of the application;
[0024] Figure 3 A step flow chart of a pickup cost determination method provided by an embodiment of the application;
[0025] Figure 4 A step flow chart of a personal willingness probability prediction method provided by an embodiment of the application;
[0026] Figure 5 A step flow chart of a cost saving determination method provided by an embodiment of the application;
[0027] Figure 6 A step flow chart of a pickup location replacement recommendation method provided by an embodiment of the application;
[0028] Figure 7 A flow chart of dynamically adjusting a pickup point of a passenger provided by an embodiment of the application;
[0029] Figure 8 A structural schematic diagram of a pickup location recommendation device provided by an embodiment of the application;
[0030] Figure 9 A structural schematic diagram of an electronic device provided by an embodiment of the application. DETAILED DESCRIPTION
[0031] In order to make the above objectives, features and advantages of the application more apparent, further specific embodiments of the application will be described in detail below with reference to the accompanying drawings and specific embodiments.
[0032] The terms used in the embodiments of the application are merely for the purpose of describing specific embodiments and are not intended to limit the application. The singular forms "a", "an" and "the" used in the embodiments of the application and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise.
[0033] Reference Figure 1 , a step flow chart of a pickup location recommendation method provided by an embodiment of the application is shown, as Figure 1As shown, the pickup location recommendation method can include steps 101-104.
[0034] Step 101: In a case where it is determined that the original pickup location corresponding to the pickup order initiated by the passenger meets the pickup location replacement condition, the pickup cost of a plurality of candidate pickup locations is obtained.
[0035] In this embodiment, the original pickup location refers to the initial pickup place (such as "XX community No. 3 door", "XX mall main entrance", etc.) positioned by the passenger through the APP or manually input when initiating the online car-hailing order.
[0036] The pickup location replacement condition refers to a standard preset by the online car-hailing platform to determine whether the original pickup location needs to be replaced.
[0037] After the online car-hailing platform receives the online car-hailing order initiated by the passenger, the original pickup location corresponding to the online car-hailing order can be obtained, and it is determined whether the original pickup location meets the pickup location replacement condition. Specifically, the size relationship between the pickup cost corresponding to the original pickup location and the cost threshold value preset by the platform can be compared, and if the pickup cost is greater than the cost threshold value, it can be determined that the original pickup location meets the pickup location replacement condition. For the implementation process, the following Figure 2 will be described in detail.
[0038] Referring to Figure 2 , a step flowchart of a position replacement condition determination method provided by an embodiment of the present application is shown. As Figure 2 shown, the position replacement condition determination method can include steps 201 and 202.
[0039] Step 201: Obtain the original pickup cost corresponding to the original pickup location corresponding to the online car-hailing order.
[0040] In this embodiment, after receiving the online car-hailing order initiated by the passenger, the pickup cost corresponding to the original pickup location corresponding to the online car-hailing order, i.e., the original pickup cost, can be obtained. Specifically, the original pickup cost of the original pickup location can be evaluated according to the information such as the traffic congestion of the road section where the vehicle in the target distance range (such as 1 kilometer or 3 kilometers, etc.) corresponding to the original pickup location drives to the original pickup location, the position information of the original pickup location (such as whether it is an internal road of a community, whether it is a narrow road section, etc.), etc.
[0041] Of course, in actual application, a deep learning model can also be used to take the surrounding information of the original pickup location as input to predict the original pickup cost corresponding to the original pickup location, etc. The present embodiment does not limit the manner of obtaining the original pickup cost.
[0042] After obtaining the original pickup cost corresponding to the original pickup location corresponding to the online car-hailing order, step 202 is performed.
[0043] Step 202: determining that the original pickup location meets the pickup location replacement condition in a case where the original pickup cost is greater than the cost threshold.
[0044] The cost threshold refers to a threshold value preset by the online car-hailing platform for judging whether the pickup cost meets the pickup location replacement condition. The specific value of the cost threshold can be determined according to business needs, which is not limited in the embodiment.
[0045] After obtaining the original pickup cost corresponding to the original pickup location of the online car-hailing order, the size relationship between the original pickup cost and the cost threshold can be compared. And whether the original pickup location meets the pickup location replacement condition is determined according to the size relationship between the two.
[0046] Specifically, if the original pickup cost is less than the cost threshold, no processing is performed.
[0047] If the original pickup cost is greater than or equal to the cost threshold, it can be determined that the original pickup location meets the pickup location replacement condition.
[0048] The embodiment of the application can replace the pickup location only when the pickup cost is large by presetting the threshold value of the pickup cost corresponding to the pickup location, which can avoid the problem that the high pickup cost leads to the reduction of the online car-hailing platform revenue.
[0049] The candidate pickup location refers to a candidate place suitable for pickup and close to the original location, which is filtered by the online car-hailing platform according to the surrounding environment of the original pickup location.
[0050] The pickup cost refers to the comprehensive cost required for a vehicle to reach a certain pickup location from the current location, which usually includes time cost (such as driving minutes), distance cost (such as driving kilometers), energy cost (such as fuel / electricity), etc. The platform will convert these costs into a unified value.
[0051] After determining that the original pickup location meets the pickup location replacement condition, the pickup costs of multiple candidate pickup locations can be obtained. In the embodiment, the pickup costs of multiple candidate pickup locations can be predicted according to the pickup costs of multiple vehicles driving to multiple candidate pickup locations and the dispatch probability of multiple vehicles. For the implementation process, the pickup cost prediction method can be combined with Figure 3 The following will be described in detail.
[0052] Referring to Figure 3 , a step flowchart of a pickup cost determination method provided by an embodiment of the application is shown. As Figure 3 described, the pickup cost determination method can include steps 301, 302 and 303.
[0053] Step 301: Obtain initial pick-up costs of a plurality of vehicles driving to a plurality of candidate pick-up locations.
[0054] In the embodiment, the plurality of vehicles refer to the ride-hailing vehicles in the ride-hailing platform system, which are currently in a state of being available for orders (not in a state of being out of service or having accepted a full order) and are located within a reasonable range of the passenger order (e.g., idle vehicles within 5 kilometers of the passenger's original pick-up location).
[0055] The initial pick-up cost refers to the basic cost of a single vehicle driving from the current real-time location to a certain candidate pick-up location, which is usually composed of distance cost (fuel / electricity cost corresponding to vehicle driving kilometers) and time cost (driver waiting cost corresponding to driving time consumption), without considering the probability of whether the vehicle will ultimately accept an order.
[0056] When calculating the pick-up cost of the plurality of candidate pick-up locations, the pick-up cost of the plurality of candidate pick-up locations can be obtained. Specifically, road section condition information of roads on which the plurality of vehicles drive to the plurality of candidate pick-up locations, order information of ride-hailing orders, and vehicle state information of the plurality of vehicles can be obtained. A pre-trained pick-up cost prediction model is called to process the road section condition information, the order information, and the vehicle state information, and the initial pick-up cost of the plurality of vehicles driving to the plurality of candidate pick-up locations is predicted.
[0057] In actual application, a pick-up cost C_jk model (i.e., the pick-up cost prediction model in the embodiment) can be pre-set, which can be a strategy-based relational model. For example, according to a series of data such as pick-up distance, pick-up time, number of pick-up road obstacles, number of pick-up traffic lights, pick-up road section type, and pick-up road section traffic flow, the initial pick-up cost of the driver driving to each candidate pick-up point is finally converted (the pick-up cost can be linked to the order coupon price or the driver reward fee according to business conditions). Or train deep models such as mmoe / essm to predict the initial pick-up cost of each candidate pick-up point.
[0058] After obtaining the initial pick-up cost of the plurality of vehicles driving to the plurality of candidate pick-up locations, step 302 is performed.
[0059] Step 302: According to the vehicle state information of the plurality of vehicles at the current time, determine the order allocation probability of the plurality of vehicles for the ride-hailing order.
[0060] The order allocation probability refers to the possibility (a value between 0 and 1) of a vehicle being allocated to the current ride-hailing order by the platform, which is determined by the vehicle state information (e.g., a vehicle with a high order allocation probability is idle and close).
[0061] When obtaining the pick-up costs of multiple candidate boarding locations, the vehicle status information of multiple vehicles at the current moment can be obtained, and based on the vehicle status information of multiple vehicles at the current moment, the dispatch probability of multiple vehicles for online car-hailing orders can be determined.
[0062] In the specific implementation, an MMoe multi-scenario prediction model can be trained to predict the dispatch probability of each vehicle based on the current status of the vehicle (information such as the distance between the vehicle and multiple candidate boarding locations).
[0063] Step 303: Determine the pick-up cost of each candidate boarding location based on the initial pick-up cost of the multiple vehicles traveling to each of the multiple candidate boarding locations and the dispatch probabilities corresponding to the multiple vehicles.
[0064] After obtaining the initial pickup cost of each of the multiple candidate pickup locations and the corresponding dispatch probabilities of multiple vehicles, the pickup cost of each candidate pickup location can be determined based on the initial pickup cost of multiple vehicles traveling to each of the multiple candidate pickup locations and the corresponding dispatch probabilities of multiple vehicles. The calculation formula is as follows:
[0065] C_j=ΣP_k*C_jk
[0066] In the above formula, C_j is the pick-up cost, P_k is the dispatch probability, and C_jk is the initial pick-up cost.
[0067] The embodiment of the present application combines the vehicle's initial pick-up cost and the actual dispatch probability to calculate the pick-up cost of the candidate location, which is closer to the actual order-taking scenario (avoiding the cost calculation deviation caused by "vehicles with low dispatch probability"). Ultimately, it can more accurately screen out the boarding location that is better for both the platform and passengers, and reduce the overall pick-up efficiency loss (such as reducing the driver's idle time and shortening the passenger's waiting time).
[0068] Of course, the method for determining the pick-up cost of multiple candidate boarding locations is not limited to the above Figure 3 In the specific implementation of the method described in the embodiment, the online ride-hailing platform can also calculate the distance and estimated time from the driver's current location to each candidate pickup location through map navigation, and convert it into the pick-up cost based on local oil / electricity prices, vehicle depreciation, etc. This embodiment does not limit the method for obtaining the pick-up cost for multiple candidate pickup locations.
[0069] Step 102: Determine the passenger's personal willingness probability to go to the multiple candidate boarding locations based on the ticket compensation rate and the distance between the passenger and the multiple candidate boarding locations.
[0070] The coupon subsidy rate refers to a coupon subsidy ratio provided by the online car-hailing platform to encourage the passenger to change to a candidate drop-off location (the subsidy cost borne by the platform accounts for the coupon face value).
[0071] The personal willingness probability refers to the possibility of the passenger willing to go to a candidate drop-off location (represented by a value between 0 and 1, 1 represents 100% willingness, and 0 represents complete unwillingness), which is mainly related to the distance of the passenger to the candidate location and the coupon subsidy rate (the closer the distance, the higher the coupon subsidy rate, and the higher the willingness probability).
[0072] After determining that the original drop-off location meets the drop-off location change condition and obtaining a plurality of candidate drop-off locations, the coupon subsidy rate preset by the online car-hailing platform and the distance between the passenger and the plurality of candidate drop-off locations can be obtained. The coupon subsidy rate can be a plurality of grades of coupon subsidy rates, which can be used by the online car-hailing platform to evaluate which grade of coupon subsidy rate to use for subsequent user subsidies. The distance between the passenger and the plurality of candidate drop-off locations can be calculated by a positioning system, and the method of obtaining the distance is not limited in this embodiment.
[0073] After obtaining the preset coupon subsidy rate and the distance between the passenger and the plurality of candidate drop-off locations, the personal willingness probability of the passenger to go to the plurality of candidate drop-off locations can be determined according to the preset coupon subsidy rate and the distance between the passenger and the plurality of candidate drop-off locations. Specifically, a causal inference model can be used to predict the personal willingness probability. For the prediction process, the causal inference model can be combined with the distance between the passenger and the plurality of candidate drop-off locations and the coupon subsidy rate. Figure 4 The following will be described in detail.
[0074] Referring to Figure 4 , a step flowchart of a personal willingness probability prediction method provided by an embodiment of the present application is shown. As Figure 4 described, the personal willingness probability prediction method can include steps 401 and 402.
[0075] Step 401: Obtain order information of the online car-hailing order, walking information of the passenger walking to the plurality of candidate drop-off locations, and location information of the plurality of candidate drop-off locations.
[0076] In this embodiment, the order information of the online car-hailing order refers to the basic data contained in the online car-hailing order initiated by the passenger, which can include: order starting point (original drop-off location), terminal, expected travel distance, expected travel time, passenger-selected vehicle type (such as express car / private car), order initiation time (such as early morning peak 7:30), passenger historical travel preferences, and the like.
[0077] The walking information refers to relevant data of a passenger walking from a current location to a candidate pickup location, including: walking distance (e.g., 200 meters), walking time (e.g., 3 minutes), comfort level of a walking path (e.g., whether there is a traffic light, whether it needs to cross the road, whether the road surface is flat, whether it is shaded and rainproof), and the like.
[0078] The location information refers to geographical feature data of the candidate pickup location itself and surroundings, including: latitude and longitude coordinates, whether it is a legal parking area (e.g., a parking space is marked on the roadside), surrounding road traffic conditions (e.g., whether it is congested, the number of lanes), surrounding landmark buildings (e.g., convenience stores, bus stops, for passenger positioning), historical pickup success rate (e.g., whether the passenger successfully gets on the bus in the past month, whether the driver is easy to find), and the like.
[0079] In the calculation of the individual willingness probability, order information of a network car-hailing order, walking information of a passenger walking to a plurality of candidate pickup locations, and location information of the plurality of candidate pickup locations can be obtained. The order information can be directly extracted from an order submitted by the passenger. The walking information can be calculated through a map service API (Application Programming Interface). The location information can be obtained through a map database and historical pickup data.
[0080] Step 402: calling a pre-trained causal inference model to process the pre-set coupon rate, the order information, the walking information, and the location information, to obtain the individual willingness probability of the passenger to the plurality of candidate pickup locations.
[0081] After obtaining the order information of the online car-hailing order, the walking information of the passenger walking to the plurality of candidate pickup locations, and the location information of the plurality of candidate pickup locations, a pre-trained causal inference model can be called to process the pre-set coupon subsidy rate, order information, walking information and location information to obtain the individual willingness probability of the passenger to go to the plurality of candidate pickup locations. Specifically, for a passenger i, there is an original pickup point, and the original pickup point has a pickup cost Ci. For the scenario of encouraging passengers to change pickup points, the passenger is recommended to go to the jth pickup point, and the nth coupon subsidy rate is given. Based on the causal inference model (such as EFIN (Enhanced Fairness-aware Uplift Network) and the like), it is speculated that the individual willingness of the passenger to accept the nth coupon subsidy rate and go to the jth pickup point is Pijn. It is known that Pijn is related to order\passenger information i, walking duration duration, walking mileage distance, marketing coupon subsidy rate n, pickup cost C_ij of walking to the jth pickup point, pickup point surrounding information, etc., that is, Pijn=f(i,duration,mileage,n,C_ij).
[0082] Among them, for each candidate pickup location, different coupon subsidy rates correspond to different individual willingness probabilities.
[0083] Among them, the causal inference model can be trained based on historical data, the input variables are coupon subsidy rate, order information, walking information, location information, and the output is individual willingness probability under different coupon subsidy rates.
[0084] Model training basis: the training data comes from platform historical orders, including: in the past 6 months, when passengers face different coupon subsidy rates (such as 0% / 10% / 20% / 30%), different walking information (such as 50 meters / 200 meters / 500 meters), and different location information (such as easy parking / difficult parking) in the "original position needs to be changed" scenario, whether to actually go to the candidate location (recorded as 1 for "yes" and 0 for "no").
[0085] Causal relationship capture: the model excludes confounding factors through "control variable method", for example: when the walking distance is fixed at 150 meters, analyze whether the passenger acceptance rate is improved by the coupon subsidy rate alone (rather than the walking distance change) when the coupon subsidy rate is increased from 10% to 20%.
[0086] The model inference process can be: for candidate location A, input: coupon supplement rate 10%, order information (morning peak, fast car, passenger price sensitive), walking information (150 meters / 2 minutes / high comfort level), location information (easy to park / high success rate), the model outputs the willingness probability 0.7 (70%). If the coupon supplement rate is increased to 20% and other information remains unchanged, the model outputs the willingness probability 0.85 (85%) (because the coupon supplement rate is increased, the willingness is increased).
[0087] For candidate location B, input: coupon supplement rate 10%, order information (same as above), walking information (300 meters / 5 minutes / comfort level medium), location information (slightly congested / medium success rate), the model outputs the willingness probability 0.4 (40%). If the coupon supplement rate is increased to 20% and other information remains unchanged, the model outputs the willingness probability 0.6 (60%) and so on.
[0088] It can be understood that the above examples are only examples listed for better understanding of the technical solutions of the embodiments of the present application, and are not the only limitation of the embodiments.
[0089] The embodiments of the present application can accurately capture the real influence of coupon supplement rate, walking cost and location characteristics on passenger willingness through causal inference model, can more accurately predict the acceptance probability of passengers to candidate locations under different coupon supplement rates, thereby helping the platform to formulate more reasonable coupon supplement strategy (such as setting higher coupon supplement rate for candidate locations with long walking distance), improving the acceptance rate of location recommendation, reducing invalid recommendation, and finally optimizing passenger experience and platform operation efficiency.
[0090] Step 103: determining the saved cost of the plurality of candidate pickup locations under different coupon supplement rates according to the pick-up cost, the individual willingness probability, and the original pick-up cost and coupon pre-price of the online car-hailing order.
[0091] The original estimated price of the online car-hailing order before using any coupon (such as the sum of the basic mileage fee + time fee + long-distance fee, etc.). For example, the coupon pre-price of Xiaowang's order from the pickup point to the destination is 30 yuan.
[0092] The saved cost refers to the comprehensive cost that the online car-hailing platform is expected to reduce by recommending the passenger to change to a certain candidate pickup location.
[0093] After obtaining the pick-up costs of the plurality of candidate pick-up locations and the individual willingness probabilities of the passenger to go to the plurality of candidate pick-up locations, the saving costs of the plurality of candidate pick-up locations at different coupon subsidies can be determined according to the pick-up costs, the individual willingness probabilities, and the original pick-up cost and the pre-coupon price of the online car-hailing order. Specifically, the additional coupon subsidy costs of the plurality of candidate pick-up locations at different coupon subsidies and the target pick-up costs can be determined first, and then the saving costs of the plurality of candidate pick-up locations at different coupon subsidies can be calculated in combination with the original pick-up cost of the original pick-up location. For the implementation process, the following Figure 5 will be described in detail.
[0094] Referring to Figure 5 , a step flowchart of a saving cost determination method provided by an embodiment of the present application is shown. As shown in Figure 5 , the saving cost determination method can include steps 501, 502, and 503.
[0095] Step 501: determining, based on the individual willingness probability, the coupon subsidy rate, and the pre-coupon price, an additional coupon subsidy cost of the plurality of candidate pick-up locations at different coupon subsidies.
[0096] In this embodiment, the additional coupon subsidy cost refers to the coupon subsidy cost paid by the online car-hailing platform to encourage the passenger to select the candidate pick-up location, which is directly related to the coupon subsidy rate and the passenger acceptance probability (individual willingness probability).
[0097] After obtaining the individual willingness probability, the coupon subsidy rate, and the pre-coupon price, the additional coupon subsidy cost of the plurality of candidate pick-up locations at different coupon subsidies can be determined based on the individual willingness probability, the coupon subsidy rate, and the pre-coupon price. The calculation formula can be Pijn*n*B, where Pijn is the individual willingness probability, n is the coupon subsidy rate, and B is the pre-coupon price.
[0098] Step 502: determining, according to the pick-up cost and the individual willingness probability, a target pick-up cost of the plurality of candidate pick-up locations at different coupon subsidies.
[0099] The target pick-up cost refers to the pick-up cost that the platform is expected to pay for the candidate location after considering the actual acceptance probability of the passenger.
[0100] After obtaining the pick-up costs of the plurality of vehicles driving to the plurality of candidate pick-up locations, the target pick-up costs of the plurality of candidate pick-up locations at different coupon subsidies can be determined according to the pick-up costs and the individual willingness probabilities. The calculation formula can be Pijn*∑P_k*C_jk, where P_k is the initial pick-up cost and C_jk is the dispatch probability.
[0101] Step 503: According to the original pickup cost, the additional coupon subsidy cost and the target pickup cost, determine the saving cost of the multiple candidate pickup locations under different coupon subsidy rates.
[0102] After obtaining the additional coupon subsidy cost and the target pickup cost, the saving cost of the multiple candidate pickup locations under different coupon subsidy rates can be determined according to the original pickup cost, the additional coupon subsidy cost and the target pickup cost.
[0103] In this embodiment, based on the idea of maximizing the saving pickup cost, it can be determined which pickup point and which n transfer coupon subsidy rate to select. The operational optimization problem is as follows:
[0104] The decision variable is as follows:
[0105] j: transfer destination, from the limited candidate pickup point J.
[0106] n coupon subsidy rate, real value, range in [0.01, 0.15], every 0.01 is a grade.
[0107] The objective function is as follows:
[0108] Maximize the saving cost: maximize(C-Pijn*n*B-Pijn*ΣP_k*C_jk).
[0109] subject to dij≤300
[0110] j∈J
[0111] n∈[0.01, 0.15]
[0112] Where subject is the constraint condition, i.e. the constraint of maximization, indicating that the distance between the passenger and the candidate pickup point is limited within 300 meters.
[0113] Passenger i, pickup point j, vehicle k, coupon subsidy rate n, coupon price B, original pickup cost of pickup point (i.e. original pickup cost) C, Pijn is the prediction of i's transfer probability, P_k is the dispatch probability, C_jk is the estimated pickup cost of each candidate pickup point to each vehicle k at the current time.
[0114] Algorithm implementation:
[0115] In actual solving, the following steps can be used:
[0116] Generate a feasible solution space: for order i, filter all candidate pickup points j that satisfy dij≤300.
[0117] Optimize n for each j: for each feasible j, maximize the objective function on n∈[0.01, 0.15] (one-dimensional optimization algorithm can be used to realize it).
[0118] Select the optimal j and n: compare the objective function values of all feasible (j, n) pairs and select the maximum value.
[0119] The embodiment of the application quantifies the cost benefit under different coupon subsidies, and the platform can accurately screen out the candidate position and coupon subsidy combination that saves the most cost, optimizes the operation cost while improving the passenger acceptance. Compared with single consideration of pickup efficiency or subsidy intensity, this scheme realizes the dynamic balance of the two, finally reduces the total expenditure of the platform, and improves the resource utilization efficiency.
[0120] After determining the cost savings of the multiple candidate pickup locations under different coupon subsidies according to the pickup cost, the individual willingness probability, and the original pickup cost and coupon pre-price of the online car-hailing order, step 104 is performed.
[0121] Step 104: From the multiple candidate pickup locations, screen out the candidate pickup location that saves the most cost as the replacement pickup location, and recommend the replacement pickup location to the passenger.
[0122] After determining the cost savings of the multiple candidate pickup locations under different coupon subsidies, the candidate pickup location that saves the most cost can be screened out from the multiple candidate pickup locations as the replacement pickup location, and the replacement pickup location is recommended to the passenger. Specifically, the replacement pickup location and the corresponding coupon subsidy amount can be recommended to the passenger to decide whether to replace the pickup location. For this implementation process, the following can be combined Figure 6 The following will be described in detail.
[0123] Referring to Figure 6 , a step flowchart of a replacement pickup location recommendation method provided by an embodiment of the application is shown. As Figure 6 shown, the replacement pickup location recommendation method can include steps 601, 602, 603, and 604.
[0124] Step 601: From the multiple candidate pickup locations, screen out the candidate pickup location that saves the most cost as the replacement pickup location.
[0125] In this embodiment, after obtaining the cost savings of each candidate pickup location under different coupon subsidies, the candidate pickup location that saves the most cost can be screened out from the multiple candidate pickup locations as the replacement pickup location. Specifically, the cost savings of all candidate locations under different coupon subsidies (such as candidate A saves 2.01 yuan under 20% coupon subsidy, and candidate B saves 5.58 yuan under 20% coupon subsidy) can be traversed, and the combination with the highest cost savings (such as candidate B + 20% coupon subsidy, saving 5.58 yuan) is selected, etc.
[0126] It can be understood that the above examples are only examples for better understanding the technical solutions of the embodiments of the present application, and are not the only limitation of the embodiments.
[0127] Step 602: Obtain the target coupon subsidy rate corresponding to the replacement pickup location.
[0128] The target coupon subsidy rate refers to the coupon subsidy rate that saves the most cost for the platform among the multiple candidate pickup locations (i.e. the optimal coupon subsidy rate screened out in the above steps).
[0129] The coupon subsidy rate corresponding to the maximum cost saving in the above step 601 can be obtained (e.g. candidate B saves the most at a 20% coupon subsidy rate), and the maximum cost saving (5.58 yuan) of candidate B corresponds to a coupon subsidy rate of 20%, so the target coupon subsidy rate = 20% and the like.
[0130] After obtaining the target coupon subsidy rate corresponding to the replacement pickup location, step 603 is performed.
[0131] Step 603: Determine the additional subsidy price according to the target coupon subsidy rate and the coupon pre-price.
[0132] The additional subsidy price refers to the coupon amount calculated based on the target coupon subsidy rate (i.e. the discount amount that the passenger can enjoy) to encourage the passenger to replace to the target location by the online car-hailing platform.
[0133] After obtaining the target coupon subsidy rate corresponding to the replacement pickup location, the additional subsidy price can be determined according to the target coupon subsidy rate and the coupon pre-price. Specifically, the additional subsidy price = coupon pre-price x target coupon subsidy rate, for example, given that the coupon pre-price = 40 yuan and the target coupon subsidy rate = 20%, then: the additional subsidy price = 40 x 20% = 8 yuan and the like.
[0134] Step 604: Recommend the replacement pickup location and the additional subsidy price to the passenger to decide whether to replace the pickup location.
[0135] After obtaining the additional subsidy price, the replacement pickup location and the additional subsidy price can be recommended to the passenger to decide whether to replace the pickup location. Specifically, the passenger can be pushed information through an APP, and the passenger can choose to "confirm replacement" or "reject", if accepted, the system automatically updates the pickup location of the order; if rejected, the original location remains unchanged.
[0136] Through quantitative analysis, the platform can accurately find the cost-optimal location recommendation strategy (target location + subsidy amount) according to the embodiments of the present application, which can improve the passenger acceptance with the minimum subsidy cost while ensuring the driver pickup efficiency.
[0137] Next, combined with Figure 7 For the flowchart of dynamically adjusting the pickup point of the passenger. As shown in Figure 7As shown, the flow can include:
[0138] 1. Estimate the pickup cost Cjk of each pickup point j at the current time for each candidate vehicle k. Specifically, a pickup cost Cjk model can be designed, which can be a strategy-based relational model, such as finally converting the pickup cost to be paid to the driver according to a series of data such as pickup distance, pickup time, number of pickup roadblocks, number of pickup traffic lights, pickup road segment type, and pickup road segment traffic volume.
[0139] 2. Prediction model of the dispatch probability P_k of each candidate vehicle k. Specifically, an mmoe multi-scene prediction model can be trained to predict the dispatch probability P_k of each vehicle according to the current status of the vehicle.
[0140] 3. Total pickup cost estimate Cj of each pickup point j at the current time. The total pickup cost Cj of each pickup point is obtained from Cjk and P_k, that is, Cj = ΣP_k*Cjk.
[0141] 4. Individual willingness probability of passenger i walking to the jth pickup point and choosing to give the nth coupon rate. Based on the causal inference model EFIN, it is speculated that the passenger accepts a certain n coupon rate and goes to the jth pickup point, and the individual willingness is Pijn. It is known that Pijn is related to order\passenger information i, walking duration duration, walking mileage distance, marketing coupon rate n, pickup cost C_ij to j pickup point, and pickup point surrounding information, that is, Pijn = f(i,duration,mileage,n,C_ij).
[0142] 5. The whole problem is modeled as a problem of maximizing the saved cost, and based on the operational optimization algorithm, it is calculated how to use the least and less than the saved driver pickup cost marketing coupon amount to encourage users to walk and maximize the platform revenue. After obtaining the transfer willingness cost of passenger i, based on the idea of maximizing the saved pickup cost, it is determined which pickup point and which n transfer coupon rate to choose.
[0143] In the embodiment of the present application, the existing pickup point distribution only decides the optimal pickup point according to the current pickup point information or surrounding road condition information, and only aims to improve the driver and passenger experience. The present application introduces marketing means, aims to maximize the platform profit, encourages passengers to change pickup points by issuing marketing coupons, not only makes passengers feel the preferential strength, but also reduces the pickup cost of drivers, improves the platform profit, and realizes the three-win of drivers, passengers and platform.
[0144] The method for recommending a pickup location provided in the embodiments of the present application comprises: in a case where it is determined that a pickup location corresponding to a ride-hailing order initiated by a passenger satisfies a pickup location replacement condition, obtaining pickup costs of a plurality of candidate pickup locations; determining a personal willingness probability of the passenger to go to the plurality of candidate pickup locations according to a coupon rate and distances between the passenger and the plurality of candidate pickup locations; determining a cost saving of the plurality of candidate pickup locations according to the pickup costs, the personal willingness probability, and an original pickup cost and a coupon price of the ride-hailing order; and selecting a candidate pickup location with the largest cost saving from the plurality of candidate pickup locations as a replacement pickup location, and recommending the replacement pickup location to the passenger. The embodiments of the present application enable the passenger to go to a pickup location with a small pickup cost by means of a coupon compensation method, thereby reducing a pickup cost (a driver reward and other fees caused by the pickup cost), while ensuring that the pickup cost saved is greater than a coupon compensation amount paid more, so as to achieve the goal of maximizing platform revenue and improve the driver and passenger experience.
[0145] Reference Figure 8 , a structure schematic diagram of a pickup location recommendation device provided by the embodiments of the present application is shown, as shown in Figure 8 , the pickup location recommendation device 800 can comprise the following modules:
[0146] The pickup cost obtaining module 810 is configured to, in a case where it is determined that a pickup location corresponding to a ride-hailing order initiated by a passenger satisfies a pickup location replacement condition, obtain pickup costs of a plurality of candidate pickup locations.
[0147] The personal willingness determining module 820 is configured to determine a personal willingness probability of the passenger to go to the plurality of candidate pickup locations according to a coupon rate and distances between the passenger and the plurality of candidate pickup locations.
[0148] The cost saving determining module 830 is configured to determine a cost saving of the plurality of candidate pickup locations under different coupon rates according to the pickup costs, the personal willingness probability, and an original pickup cost and a coupon price of the ride-hailing order.
[0149] The replacement location recommending module 850 is configured to select a candidate pickup location with the largest cost saving from the plurality of candidate pickup locations as a replacement pickup location, and recommend the replacement pickup location to the passenger.
[0150] Optionally, the replacement condition determining module comprises:
[0151] The original cost obtaining unit is configured to obtain an original pickup cost corresponding to a pickup location corresponding to a ride-hailing order.
[0152] The replacement condition determining unit is configured to, in a case where the original pickup cost is greater than a cost threshold, determine that the pickup location satisfies a pickup location replacement condition.
[0153] Optionally, the pickup cost obtaining module comprises:
[0154] An initial cost obtaining unit, configured to obtain initial pickup costs of a plurality of vehicles driving to a plurality of candidate pickup locations;
[0155] A dispatch probability determining unit, configured to determine, according to vehicle condition information of the plurality of vehicles at a current time, a dispatch probability of the plurality of vehicles for the online car-hailing order;
[0156] A pickup cost determining unit, configured to determine, according to the initial pickup cost of each of the plurality of vehicles driving to the plurality of candidate pickup locations and the corresponding dispatch probability of the plurality of vehicles, the pickup cost of each candidate pickup location.
[0157] Optionally, the initial cost obtaining unit comprises:
[0158] An information obtaining sub-unit, configured to obtain road section condition information of roads on which the plurality of vehicles drive to the plurality of candidate pickup locations, order information of the online car-hailing order, and vehicle state information of the plurality of vehicles;
[0159] A cost prediction sub-unit, configured to call a pre-trained pickup cost prediction model to process the road section condition information, the order information, and the vehicle state information, and predict the initial pickup cost of the plurality of vehicles driving to the plurality of candidate pickup locations.
[0160] Optionally, the personal willingness determining module comprises:
[0161] An information obtaining unit, configured to obtain order information of the online car-hailing order, walking information of the passenger walking to the plurality of candidate pickup locations, and location information of the plurality of candidate pickup locations;
[0162] A personal willingness prediction unit, configured to call a pre-trained causal inference model to process the pre-set coupon rate, the order information, the walking information, and the location information, and obtain a personal willingness probability of the passenger going to the plurality of candidate pickup locations;
[0163] For each candidate pickup location, the personal willingness probability corresponds to different coupon rates.
[0164] Optionally, the saved cost determining module comprises:
[0165] An additional coupon cost determining unit, configured to determine, based on the personal willingness probability, the coupon rate, and the coupon price, an additional coupon cost of the plurality of candidate pickup locations under different coupon rates;
[0166] The target pick-up cost determination unit is configured to determine a target pick-up cost of the multiple candidate drop-off locations under different coupon subsidy rates according to the pick-up cost and the individual willingness probability.
[0167] The cost saving determination unit is configured to determine a cost saving of the multiple candidate drop-off locations under different coupon subsidy rates according to the original pick-up cost, the additional coupon subsidy cost and the target pick-up cost.
[0168] Optionally, the replacement location recommendation module comprises:
[0169] The replacement location screening unit is configured to screen a candidate drop-off location with the largest cost saving from the multiple candidate drop-off locations as the replacement drop-off location.
[0170] The target coupon subsidy rate acquisition unit is configured to acquire a target coupon subsidy rate corresponding to the replacement drop-off location.
[0171] The additional subsidy price determination unit is configured to determine an additional subsidy price according to the target coupon subsidy rate and the pre-coupon price.
[0172] The replacement location recommendation unit is configured to recommend the replacement drop-off location and the additional subsidy price to the passenger to determine whether to replace the drop-off location.
[0173] The drop-off location recommendation apparatus provided by the embodiments of the present application determines the pick-up cost of multiple candidate drop-off locations in the case that the original drop-off location corresponding to the online car-hailing order initiated by the passenger satisfies the drop-off location replacement condition. The individual willingness probability of the passenger to go to the multiple candidate drop-off locations is determined according to the coupon subsidy rate and the distance between the passenger and the multiple candidate drop-off locations. The cost saving of the multiple candidate drop-off locations is determined according to the pick-up cost, the individual willingness probability, the original pick-up cost and the pre-coupon price of the online car-hailing order. The candidate drop-off location with the largest cost saving is screened from the multiple candidate drop-off locations as the replacement drop-off location, and the replacement drop-off location is recommended to the passenger. The embodiments of the present application enable the passenger to go to the drop-off point with a small pick-up cost by the coupon subsidy method, thereby reducing the pick-up cost (the driver's reward and other fees caused by the pick-up cost), while ensuring that the saved pick-up cost is greater than the additional coupon subsidy, so as to achieve the goal of maximizing the platform revenue and improve the driver and passenger experience.
[0174] The embodiments of the present application further provide an electronic device, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, and the computer program is executed by the processor to implement the above-mentioned drop-off location recommendation method.
[0175] Figure 9 The structure schematic diagram of an electronic device 900 of the embodiments of the present application is shown. As shown in FIG. 9, the electronic device 900 comprises a processor 901, a memory 902 and a bus 903. Figure 9As shown, the electronic device 900 includes a central processing unit (CPU) 901, which can perform various appropriate actions and processes according to computer program instructions stored in a read-only memory (ROM) 902 or loaded from a storage unit 908 into a random access memory (RAM) 903. Various programs and data required by the electronic device 900 to operate can also be stored in the RAM 903. The CPU 901, the ROM 902, and the RAM 903 are connected to each other via a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.
[0176] Various components in the electronic device 900 are connected to the I / O interface 905, including an input unit 906, such as a keyboard, a mouse, a microphone, etc.; an output unit 907, such as various types of displays, a speaker, etc.; the storage unit 908, such as a magnetic disk, a compact disk, etc.; and a communication unit 909, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 909 allows the electronic device 900 to exchange information / data with other devices via a computer network, such as the Internet, and / or various telecommunication networks.
[0177] The various processes and processes described above can be performed by the processing unit 901. For example, the method of any of the embodiments described above can be implemented as a computer software program tangibly embodied in a computer readable medium, such as the storage unit 908. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 900 via the ROM 902 and / or the communication unit 909. When the computer program is loaded into the RAM 903 and executed by the CPU 901, one or more actions of the method described above can be performed.
[0178] In addition, the embodiments of the present application also provide a computer readable storage medium, which stores a computer program, and the program is executed by a processor to implement the above-mentioned boarding position recommendation method.
[0179] Each embodiment in the specification is described in a progressive manner, and each embodiment focuses on the difference from other embodiments. The same and similar parts between each embodiment can be referred to each other.
[0180] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, device, or computer program product. Therefore, the embodiments of the present application can be in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the embodiments of the present application can be in the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.
[0181] The embodiments of the present application are described with reference to the flowchart illustrations and / or block diagrams of the methods, terminals (systems), and computer program products according to the embodiments of the present application. It is 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 terminal to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal, create means for implementing the functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams.
[0182] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams.
[0183] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal, such that a series of operational steps are carried out on the computer or other programmable terminal to produce a computer implemented process so that the instructions which execute on the computer or other programmable terminal provide steps for implementing the functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams. Figure 1 Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams.
[0184] Although preferred embodiments of the present application have been described, those skilled in the art will be able to make additional modifications and variations to these embodiments without departing from the basic inventive concepts disclosed. Accordingly, it is intended to be covered in the appended claims all such modifications and variations as fall within the scope of the embodiments of the present application.
[0185] Finally, it is to be understood that the terms such as first and second, and the like, herein are used only to distinguish one from another entity or action, and do not necessarily require or imply these entities or actions to be in any physical or logical order. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises... a" does not, without more constraints, exclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the element.
[0186] The above provides a boarding position recommendation method, a boarding position recommendation device, an electronic device and a computer readable storage medium. The principles and implementation manners of the present application are described by applying specific examples. The above description of the embodiments is only used to help understand the method and its core idea of the present application. For those skilled in the art, according to the idea of the present application, the specific implementation manners and application ranges can be changed. In summary, the content of the specification should not be understood as a limitation of the present application.
Claims
1. A method for recommending a boarding location, characterized in that: The method comprises: When it is determined that the original pickup location corresponding to the online ride-hailing order initiated by the passenger meets the pickup location change conditions, the pickup costs of multiple candidate pickup locations are obtained; determining the passenger's personal willingness probability to go to the multiple candidate boarding locations based on the ticket compensation rate and the distance between the passenger and the multiple candidate boarding locations; Determine the cost savings of the multiple candidate boarding locations at different coupon compensation rates based on the pick-up cost, the individual willingness probability, and the original pick-up cost and pre-coupon price of the online ride-hailing order; A candidate boarding location with the greatest cost savings is selected from the plurality of candidate boarding locations as a replacement boarding location, and the replacement boarding location is recommended to the passenger.
2. The method according to claim 1, characterized in that Determining that the original boarding location corresponding to the online ride-hailing order initiated by the passenger meets the boarding location change conditions includes: Obtain the original pick-up cost corresponding to the original boarding location corresponding to the online ride-hailing order; When the original pick-up cost is greater than the cost threshold, it is determined that the original boarding location meets the boarding location change condition.
3. The method according to claim 1, characterized in that The method of obtaining the pick-up costs of multiple candidate boarding locations includes: Obtaining initial pick-up costs of multiple vehicles traveling to the multiple candidate boarding locations; Determining, based on the vehicle status information of the multiple vehicles at the current moment, the dispatch probability of the multiple vehicles for the online ride-hailing order; The pick-up cost of each candidate boarding location is determined according to the initial pick-up cost of the multiple vehicles traveling to each of the multiple candidate boarding locations and the dispatch probabilities corresponding to the multiple vehicles.
4. The method according to claim 3, characterized in that The obtaining of initial pick-up costs of multiple vehicles traveling to the multiple candidate boarding locations includes: Obtaining road condition information of the roads on which the multiple vehicles travel to the multiple candidate pickup locations, order information of the online ride-hailing orders, and vehicle status information of the multiple vehicles; A pre-trained pick-up cost prediction model is called to process the road condition information, the order information, and the vehicle status information to predict initial pick-up costs for the multiple vehicles traveling to the multiple candidate boarding locations.
5. The method according to claim 1, wherein Determining the passenger's personal willingness probability to go to the plurality of candidate boarding locations based on the ticket compensation rate and the distance between the passenger and the plurality of candidate boarding locations includes: Obtaining order information of the online ride-hailing order, walking information of the passenger to the multiple candidate boarding locations, and location information of the multiple candidate boarding locations; Invoking a pre-trained causal inference model to process the preset ticket compensation rate, the order information, the walking information, and the location information to obtain the passenger's personal willingness probability to go to the multiple candidate boarding locations; Among them, for each candidate boarding position, different ticket compensation rates correspond to different personal willingness probabilities.
6. The method according to claim 1, characterized in that Determining the cost savings of the plurality of candidate boarding locations at different coupon compensation rates based on the pick-up cost, the personal willingness probability, and the original pick-up cost and pre-coupon price of the online ride-hailing order includes: determining, based on the personal willingness probability, the coupon subsidy rate, and the pre-coupon price, additional coupon subsidy costs for the plurality of candidate boarding locations at different coupon subsidy rates; Determining target pick-up costs for the multiple candidate boarding locations at different ticket compensation rates based on the pick-up costs and the individual willingness probabilities; The cost savings of the plurality of candidate boarding locations at different ticket compensation rates are determined according to the original pick-up cost, the additional ticket compensation cost, and the target pick-up cost.
7. The method according to claim 6, characterized in that The step of selecting a candidate boarding location with the greatest cost savings from the plurality of candidate boarding locations as a replacement boarding location, and recommending the replacement boarding location to the passenger includes: Selecting a candidate boarding location with the greatest cost savings from the plurality of candidate boarding locations as a replacement boarding location; Obtain the target coupon rate corresponding to the changed boarding location; Determine the additional subsidy price based on the target coupon subsidy rate and the pre-coupon price; The changed boarding location and the additional subsidy price are recommended to the passenger, so that the passenger can decide whether to change the boarding location.
8. A device for recommending boarding positions, characterized in that: The device comprises: A pick-up cost acquisition module is used to obtain the pick-up costs of multiple candidate pick-up locations when it is determined that the original pick-up location corresponding to the online ride-hailing order initiated by the passenger meets the pick-up location change conditions; a personal willingness determination module, configured to determine the personal willingness probability of the passenger to go to the plurality of candidate boarding locations based on the ticket compensation rate and the distance between the passenger and the plurality of candidate boarding locations; a cost saving determination module, configured to determine the cost saving of the plurality of candidate boarding locations under different coupon compensation rates based on the pick-up cost, the individual willingness probability, and the original pick-up cost and pre-coupon price of the online ride-hailing order; The replacement position recommendation module is used to screen out the candidate boarding position with the greatest cost savings from the multiple candidate boarding positions as the replacement boarding position, and recommend the replacement boarding position to the passenger.
9. An electronic device, characterized in that: include: A processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for recommending a boarding position according to any one of claims 1 to 7 is implemented.
10. A computer-readable storage medium, characterized in that When the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the method for recommending a boarding position according to any one of claims 1 to 7.