Waybill processing method and device, and electronic device

Abstract

The application discloses a way of handling waybills, belonging to the field of computer technology, which helps to improve the delivery experience. The way of handling waybills disclosed by the application embodiment comprises: determining the mode of handling waybills matched by a target waybill; in response to the mode of handling waybills matched by the target waybill being a directional recommendation mode, pushing the target waybill to at least one designated recommended transport capacity, so that the waybill handling client logged in by the at least one designated recommended transport capacity displays the target waybill on a recommended waybill list display page; and in response to the mode of handling waybills matched by the target waybill being a dispatch mode, using a preset dispatch engine to perform a dispatch operation on the target waybill. The method adopts different waybill dispatch modes for waybill dispatch according to the characteristics of the waybills, and when the waybill cannot find a candidate transport capacity in the dispatch mode, the waybill is pushed to multiple designated transport capacities, which can effectively control the order-accepting transport capacity of the waybill, thereby achieving the effect of improving the delivery quality.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a waybill processing method, apparatus, electronic device, and computer-readable storage medium. Background Technology

[0002] In existing technologies, during the order dispatching process in a waybill dispatching system, orders entering the system are typically first dispatched in the order assignment mode, and then in the order-grabbing mode. When dispatching orders in the order-grabbing mode, to expedite order acceptance, the dispatching system directly exposes the order to riders within a specified radius. The final order allocation result depends entirely on the individual rider's willingness and response speed, leading to low rider quality in order acceptance and negatively impacting the delivery experience.

[0003] It is evident that the existing waybill processing methods still need improvement. Summary of the Invention

[0004] This application provides a waybill processing method that helps improve the waybill delivery experience.

[0005] In a first aspect, embodiments of this application provide a waybill processing method, including:

[0006] Determine the waybill processing mode to match the target waybill;

[0007] In response to the target waybill being matched with a waybill processing mode that is a targeted recommendation mode, the target waybill is pushed to at least one designated recommended transport capacity, so that the waybill processing client logged into the at least one designated recommended transport capacity displays the target waybill on the recommended waybill list display page;

[0008] In response to the target waybill matching the waybill's processing mode being dispatch mode, a preset dispatch engine is used to perform a dispatch operation on the target waybill.

[0009] Secondly, embodiments of this application provide a waybill processing device, including:

[0010] The waybill processing mode determination module is used to determine the waybill processing mode that matches the target waybill.

[0011] The targeted recommendation module is used to push the target waybill to at least one designated recommended transport capacity in response to the waybill processing mode matching the target waybill being a targeted recommendation mode, so that the waybill processing client logged into the at least one designated recommended transport capacity can display the target waybill on the recommended waybill list display page;

[0012] The dispatch module is used to perform dispatch operations on the target waybill in response to the waybill processing mode matching the target waybill being dispatch mode, using a preset dispatch engine.

[0013] Thirdly, embodiments of this application also disclose an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the waybill processing method described in embodiments of this application.

[0014] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, represents the steps of the waybill processing method disclosed in embodiments of this application.

[0015] The waybill processing method disclosed in this application determines the waybill processing mode matching the target waybill; in response to the target waybill matching the waybill processing mode being a targeted recommendation mode, the target waybill is pushed to at least one designated recommended delivery capacity, so that the waybill processing client logged into the at least one designated recommended delivery capacity displays the target waybill on the recommended waybill list display page; in response to the target waybill matching the waybill processing mode being a dispatch mode, a preset dispatch engine is used to perform dispatch operations on the target waybill, which helps to improve the delivery experience while taking delivery quality into account.

[0016] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0017] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0018] Figure 1 This is a flowchart of the waybill processing method according to Embodiment 1 of this application;

[0019] Figure 2 This is a flowchart of the waybill processing method according to Embodiment 2 of this application;

[0020] Figure 3 This is one of the schematic diagrams of the waybill processing device in Embodiment 2 of this application;

[0021] Figure 4 This is the second schematic diagram of the waybill processing device according to Embodiment 2 of this application;

[0022] Figure 5 A block diagram schematically illustrates an electronic device for performing the method according to this application; and

[0023] Figure 6 A storage unit for holding or carrying program code implementing the method according to this application is illustrated schematically. Detailed Implementation

[0024] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0025] Example 1

[0026] This application discloses a waybill processing method, such as... Figure 1 As shown, the method includes steps 110 to 130.

[0027] Step 110: Determine the waybill processing mode that matches the target waybill.

[0028] The target waybill mentioned in this embodiment is a waybill to be assigned. The delivery capacity mentioned in this embodiment can be a rider, delivery vehicle, delivery drone, or other entity that performs waybill delivery.

[0029] In this embodiment, for a waybill to be assigned under the current scheduling time segment, the first step is to decide whether to reach the available capacity through system dispatch or targeted recommendation. For example, the capacity audience range of the waybill under the current scheduling time segment is determined based on the waybill's attributes and intermediate scheduling results, and then the allocation method is determined. For waybills with a relatively wide capacity audience range, even with low dispatch exposure, the reach success rate is still relatively high; therefore, system dispatch can be selected. For waybills with a relatively narrow capacity audience range, even with low dispatch exposure, the success rate of accurately reaching the target capacity is low; therefore, targeted recommendation is selected to expand the capacity reach range and improve the reach success rate.

[0030] In some embodiments of this application, the waybill processing mode includes a targeted recommendation mode and a dispatch mode. The method for determining the waybill matching the target waybill includes: determining the waybill matching mode based on the audience range of the delivery capacity in the delivery capacity set recalled under the first and second dispatch constraints; or, determining the waybill matching mode based on the order acceptance willingness distribution information of the delivery capacity in the delivery capacity set recalled under the second dispatch constraint. The first dispatch constraint is a constraint in the existing waybill assignment optimization model, which may include conditions such as user experience, delivery capacity (e.g., rider) experience, and self-defined delivery capacity. Setting the first dispatch constraint serves as a hard constraint to protect the user and delivery capacity experience, and also helps reduce the delivery capacity search space and the solution scale during waybill assignment. For example, the first dispatch constraint includes, but is not limited to, any one or more of the following: delivery time condition, order route proximity condition, number of waybills carried by the delivery capacity condition, and number of riders reached by the waybill condition. The second dispatch constraint is obtained by relaxing the first dispatch constraint. For example, the second dispatch constraint is obtained by relaxing the rider number condition in the first dispatch constraint. Appropriately relaxing the self-defined capacity constraints can expand the optimization range and obtain better order-capacity matching results without compromising the user and capacity experience.

[0031] In some embodiments of this application, determining the matching waybill processing mode for the target waybill based on the audience range of the capacity in the capacity sets recalled under the first dispatch constraint and the second dispatch constraint respectively includes: in response to the fact that the number of capacity in the capacity set recalled under the first dispatch constraint is less than the number of capacity in the capacity set recalled under the second dispatch constraint, a targeted recommendation mode is determined for the target waybill. The capacity set recalled under the first dispatch constraint for waybill w... For example, if the recall capacity set R after relaxing the relevant capacity experience constraint threshold... w If the value is not equal to Φ, then waybill w will be allocated using a targeted recommendation mode. That is, under the current scheduling time segment, the original "air-pressurized" waybill will reach the available capacity using a targeted recommendation method. Otherwise, the target waybill will be matched with the appropriate dispatch mode.

[0032] In some embodiments of this application, determining the order processing mode matching the target waybill based on the distribution information of the willingness to accept orders among the capacity in the capacity set recalled under the second dispatch constraint for the target waybill includes: determining the target waybill matching the targeted recommendation mode by ensuring that the average willingness to accept orders among the capacity in the capacity set recalled under the second dispatch constraint for the target waybill is greater than or equal to a preset willingness to accept orders threshold. For example, under the current scheduling time segment, if the average willingness to accept orders among the capacity recalled under the second dispatch constraint for waybill w... Less than the preset order acceptance threshold This indicates that if the order w reaches the available capacity using the dispatch method, the success rate of order acceptance is low or the uncertainty is high. In this case, it is advisable to use a targeted recommendation method to expand the capacity reach and improve the success rate. Otherwise, determine the dispatch mode to match the target order.

[0033] In some embodiments of this application, the preset order acceptance willingness threshold It is determined based on the test results, and the average willingness to accept orders for the recalled capacity r of the waybill w is the willingness to accept orders. The willingness to accept orders is determined through a pre-trained model. For example, one or more features from historical order dispatch data—including environmental features, capacity features, features of already carried orders, features of currently pending orders, and the cross-features between currently pending orders and already carried orders—can be used as training data. The "whether to accept an order" identifier is used as the sample label to construct training samples. Then, a classification network is trained offline based on these training samples to obtain the willingness to accept orders prediction model. In online application, for each candidate capacity, the capacity features, features of already carried orders, the cross-features between the candidate capacity and the target order, the order features of the target order, and environmental features are extracted and input into the willingness to accept orders prediction model. The model then outputs the willingness of the candidate capacity to accept the target order.

[0034] Since there were no order allocation scenarios using the second order dispatch constraint before the targeted recommendation method was activated, order dispatch data for this scenario could not be obtained. In some embodiments of this application, sample data for the order acceptance willingness prediction model is constructed by reusing large-scale network capacity order-grabbing data samples. When constructing training samples, orders in which capacity generated order-grabbing behavior are considered positive samples, and those exposed but not grabbed are considered negative samples. Due to the severe imbalance between positive and negative samples, further sampling of negative samples is required.

[0035] Step 120: In response to the target waybill being matched with a waybill processing mode that is a targeted recommendation mode, the target waybill is pushed to at least one designated recommended transport capacity, so that the waybill processing client logged into the at least one designated recommended transport capacity displays the target waybill on the recommended waybill list display page.

[0036] Once the target waybill matching and targeted recommendation mode is determined, it is necessary to further determine the reach capacity range of the target waybill in order to execute the waybill targeted recommendation. In some embodiments of this application, pushing the target waybill to at least one designated recommended capacity includes: determining at least one recommended capacity for the target waybill by solving a combination optimization problem of many-to-many matching between waybills and capacity; and pushing the target waybill to the at least one recommended capacity. The system dispatch relies on waybill attributes and scheduling intermediate results to construct an accurate waybill profile for waybill assignment. The targeted recommendation mode determines the recommended capacity for the waybill from an operations optimization perspective. In some embodiments of this application, the problem of determining the recommended capacity for the waybill can be modeled as a many-to-many matching combination optimization problem with the goal of optimizing the global user and capacity experience and constrained by the waybill reach capacity range, thereby improving the delivery experience from two dimensions: improving reach capacity efficiency and capacity reach quality.

[0037] In some embodiments of this application, determining at least one recommended capacity for the target waybill by solving a combination optimization problem of many-to-many matching of waybills and capacity includes: constructing a combination optimization problem of many-to-many matching of waybills and capacity with the objective of optimizing the waybill delivery quality of the target waybill and with constraints of capacity reach range and capacity end resource location; and using a local search optimal solution method to search for at least one recommended capacity for the target waybill among the candidate capacity of the target waybill recalled based on the second dispatch constraint.

[0038] Taking W as an example, the set of waybills to be assigned in the current time segment can be used to construct the following combinatorial optimization problem of many-to-many matching of waybills and capacity, and solve for the optimal combination of waybills and capacity:

[0039]

[0040] st:

[0041]

[0042]

[0043] in, Let w be the decision variable, representing whether to recommend waybill w to capacity r. Indicates recommendation. This indicates that it is not recommended; R w This represents the set of candidate transport capacity recalled based on the second dispatch constraint. This represents the basic score given by the transport capacity r to the delivery of waybill w, which may include, for example, a score for timeout and a score for route convenience. In the constraints, K represents the minimum number of waybills w that can be recommended to the transport capacity (referred to as the "first number" in this paper); M represents the maximum number of waybills that can be recommended to each transport capacity.

[0044] In the above optimization problem, the objective function To ensure delivery quality after a rider accepts an order, the constraints consist of two parts. The first part (i.e., the constraint that a waybill w can be recommended to at least K riders) ensures the reach of the waybill and thus ensures delivery efficiency. The second part (i.e., the constraint that the number of waybills recommended to riders at the same time in each round cannot exceed M) is the restriction on the display of resources on the rider's end.

[0045] Because in actual scheduling system operation, the initial number of candidate transport capacity for a waybill may be less than K, the constraints of the above optimization problem may be too stringent. Therefore, in some embodiments of this application, the constraints are relaxed as follows during the actual solution process: the reach of transport capacity is adjusted from a hard constraint that must be met to an optimization objective that needs to be maximized. That is, the above optimization problem can be relaxed into the following problem:

[0046]

[0047]

[0048] Where β is a maximal positive number.

[0049] Next, the matching optimization problem was further solved.

[0050] In some embodiments of this application, the method of using local search for optimal solutions to search for at least one recommended transport capacity for the target waybill among candidate transport capacities includes: an initialization sub-step, used to determine a specified number of waybills initially recommended to each candidate transport capacity according to the resource slot constraints of the transport capacity end and the number of waybills that can be recommended by each candidate transport capacity, in descending order of delivery quality, to obtain an initial recommendation relationship between waybills and candidate transport capacities; a waybill set determination sub-step, used to determine a first waybill set, a second waybill set, and a third waybill set matching each candidate transport capacity based on the specified number of waybills initially recommended to each candidate transport capacity; wherein, each waybill in the first waybill set is a waybill with insufficient capacity recommended to less than a first number of candidate transport capacities, each waybill in the second waybill set is recommended to greater than or equal to the first number of candidate transport capacities, and the waybills in the third waybill set are: the waybill set matching the first number of candidate transport capacities. The process involves several steps: a search optimization sub-step, whereby for each of the insufficient capacity waybills in the first set of waybills, a local capacity search is performed according to the order of candidate capacity shortage from largest to smallest, searching for candidate capacity to recommend the insufficient capacity waybills with the goal of optimizing waybill delivery quality; updating the recommendation relationship between waybills and candidate capacity based on the searched candidate capacity; updating the first set of waybills, the second set of waybills, and the third set of waybills matched by each candidate capacity; an iterative judgment sub-step, whereby in response to the failure to meet the iterative search termination condition, the process jumps to the search optimization sub-step; and in response to meeting the iterative search termination condition, at least one recommended capacity for the target waybill is determined based on the updated recommendation relationship between the waybill and candidate capacity.

[0051] First, in the initialization sub-step, the capacity r∈∪ recalled based on the second dispatch constraint is traversed. w∈W The specified number of waybills with the lowest scores for each transport capacity are assigned to that capacity. The specified number is the minimum of the maximum number of waybills recommended for each capacity and the number of waybills that capacity r can recommend; that is, the number of waybills assigned to capacity r is determined according to the formula min{M, number of waybills that capacity r can recommend}. Thus, the initial recommendation relationship between each transport capacity and at least one waybill can be determined.

[0052] Then, based on the initial recommendation relationship between each capacity and at least one waybill, different categories of waybill sets are determined from the waybill dimension. In some embodiments of this application, waybills with fewer than K recommended capacity items (i.e., less than a first quantity) are defined as capacity-deficient waybills, and the waybill set consisting of capacity-deficient waybills is defined as the first waybill set L, denoted as: <waybill w, recommended capacity deficit amount We define waybills with a recommended capacity quantity exceeding K (i.e., greater than or equal to the first quantity) as surplus capacity waybills, and define the set of waybills constituting the surplus capacity waybills as the second waybill set U, denoted as: <waybill w, recommended capacity surplus quantity At the same time, traversing the recommendation relation, the capacity r∈∪ w∈W R w Let the set of recommended waybills with a capacity of r that have a recommended capacity exceeding K be the third set of waybills, U. r , denoted as: <waybill i, <score Recommended rider surplus in, This indicates that waybill i is assigned a base score by capacity r during delivery, which reflects the delivery quality; each candidate capacity is matched with a third waybill set U. r .

[0053] As can be seen from the above method for determining the waybill set, the waybills in the current time segment are divided into two waybill sets, namely the first waybill set and the second waybill set. The waybills in the first waybill set establish recommendation relationships with fewer than a first number of transport capacities, while the waybills in the second waybill set establish recommendation relationships with equal to or more than the first number of transport capacities. Therefore, for the waybills in the first waybill set, in order to improve their transport capacity reach, it is necessary to further search for matching candidate transport capacities.

[0054] Next, in the search optimization sub-step, the first order set L is traversed in descending order of recommended rider shortage. For any order w∈L, with the goal of optimizing the order delivery quality, a search is conducted among the candidate capacity to recommend the order with insufficient capacity. In some embodiments of this application, the step of searching among the candidate capacity to recommend the order with insufficient capacity with the goal of optimizing the order delivery quality includes: determining candidate capacity included in the candidate capacity of the order recalled based on the first dispatch constraint, but not included in the recommended capacity of the order, as candidate recommended capacity; determining the candidate recommended capacity whose delivery quality for the order with surplus capacity among the recommended orders is lower than its delivery quality for the order with insufficient capacity, and whose delivery quality for the order with insufficient capacity is the highest, as a recommended capacity for the order with insufficient capacity.

[0055] First, for any waybill w∈L, retrieve the set of transport capacity based on the first dispatch constraint. The capacity that is not currently on its recommended list is set D. w ; Traverse set D w For candidate capacity r∈D wRecord the set U of waybills whose recommended capacity exceeds K among the waybills recommended for candidate capacity r. r Received <Rider R, rating> U r >, in, This represents the basic score for the delivery order W to be handled by the candidate capacity r. Then, the candidate capacity r and the set of delivery orders U recorded in the previous steps are iterated through. r The relationship between the basic score of the delivery order W and the capacity is determined by the formula. Determine the set of waybills U r There is a recommendation relationship between the medium and greater than K capacity members, and the capacity r with a lower basic delivery score for waybill W than the capacity r with a lower basic delivery score for waybill i. * (That is, the capacity whose delivery quality for waybill W is better than that for waybill i), and the capacity r * As the recommended capacity for waybill W, waybill i will also be included. * From transport capacity r * Remove it from the recommended list.

[0056] For example, for orders W1 and W2, the recommended rider lists obtained using the second dispatch constraint are Rw1 and Rw2, respectively, and the recommended rider sets obtained using the first dispatch constraint are respectively... and If the number of recommended riders in Rw1 and Rw2 is less than K, then orders W1 and W2 are considered orders with insufficient capacity and are assigned to the first order set L. For order W1 in the first order set L, if the recommended rider set obtained using the first dispatch constraint condition... If rider r1 is not in the current list of recommended candidate riders for order W1, then rider r1 will be added to set D. w In this process, rider r1 may become the recommended rider for order W1. Furthermore, if among the orders recommended to rider r1 (including orders w3 and w4, ...), there is a surplus of recommended riders for orders w3 and w4, exceeding the number of K candidate riders, then record the basic score of rider r1, the delivery order W1 delivered by rider r1, and the order set U. r1 The relationship between them.

[0057] Next, it is determined whether rider r1 can be changed from the recommended rider for orders w3 and w4 to the recommended rider for order W1. Specifically, by further analyzing the relationship between rider r1's base delivery scores for orders W1, W3, and W4, it is determined whether having rider r1 deliver order W1 would improve overall delivery quality. For example, if rider r1's base score for order W1 is lower than that for order W3, then rider r1 is removed from the recommended rider list for order W3 and added to the recommended rider list for order W1. In other words, order W3 is removed from rider r1's recommended order list, and order W1 is added to rider r1's recommended order list.

[0058] At this point, the recommendation relationship between waybills and delivery capacity (i.e., candidate riders) has changed. Next, based on the changed recommendation relationship, update the first waybill set L, the second waybill set U, and the second waybill set U matched with each delivery capacity determined in the previous steps. r .

[0059] Subsequently, based on the iterative search termination condition, it is further determined whether there are still waybills with insufficient capacity that need to be searched for recommended capacity. In this embodiment of the application, the iterative search termination condition includes any one or more of the following: the recommended capacity of the waybill with insufficient capacity is equal to the first quantity; the number of recommendable capacity for the waybill with insufficient capacity is 0; and there are no recommended waybills among the recommended waybills of the recommended capacity of the waybill with insufficient capacity that have been recommended to a number exceeding the first quantity of recommended capacity. Wherein, the recommendable capacity of a waybill is: candidate capacity included in the candidate capacity recalled by the waybill based on the first dispatch constraint condition, but not included in the recommended capacity of the waybill.

[0060] For example, when the recommended capacity for all waybills in the first set of waybills (i.e., the set of waybills with insufficient capacity) reaches K (i.e., the first quantity), that is, when the recommended capacity list for each waybill in the first set of waybills satisfies... Then it can be determined that the termination condition of the iterative search has been met. For example, this includes candidate capacity for the waybills recalled based on the first dispatch constraint. The number of candidate capacities not included in the recommended capacity of the waybill, i.e., the D determined in the aforementioned steps. w When the set is empty, it can be determined that the termination condition of the iterative search has been met. For example, for a waybill w, among any recommended waybills with capacity that can be recommended but has not yet entered the recommendation list, there are no waybills with more than K recommended capacity members, i.e. Then it can be determined that the termination condition of the iterative search has been met.

[0061] Once it is determined that the search results of the current iteration have met the termination condition of the iteration search, the iteration search operation is stopped, and the target waybill is pushed to the recommended capacity determined by the push relationship based on the search results of the current iteration (i.e., the updated recommendation relationship between waybills and capacity).

[0062] If it is determined that the search result of the current iteration does not meet the iteration search termination condition, the search optimization sub-step is repeated until the search result meets the iteration search termination condition.

[0063] In some embodiments of this application, the waybill processing client that logs in with at least one designated candidate transport capacity displays the target waybill on the recommended waybill list display page, including: the waybill processing client sorts and displays the received recommended waybills based on the sorting index of the target waybill, wherein the sorting index of the target waybill is obtained by weighted calculation based on the delivery experience factor, delivery efficiency factor, and transport capacity willingness to accept the order factor of the target waybill.

[0064] After determining the recommended capacity for the target waybill, the waybill scheduling system further determines the recommended waybill for each candidate capacity in the current time segment. Then, for each candidate capacity, the recommended waybill is sent to the waybill processing client currently logged in for display. Upon receiving the recommended waybills, the waybill processing client sorts all recommended waybills according to their ranking criteria and displays the sorted results on the recommended waybill list page. The recommendation criteria for each recommended waybill are obtained by weighting factors such as delivery experience, delivery efficiency, and capacity willingness to accept orders. For example, according to the formula... The calculation yielded that, This represents the ranking index of recommended freight orders W for recommended capacity r; This represents the delivery experience factor for recommended capacity r and recommended order W. This represents the delivery efficiency factor of the recommended delivery order W for the recommended capacity r. This represents the willingness to accept capacity orders for recommended capacity r and recommended waybill W; α, β, and γ represent the weights of the corresponding ranking factors, and their values ​​can be set according to specific business needs. Among them, the delivery experience factor is used to characterize the degree of delivery timeout, the delivery efficiency factor is used to characterize the route proximity of the capacity, and the willingness to accept capacity order factor is used to characterize the willingness to accept capacity.

[0065] In some embodiments of this application, the delivery experience factor The delivery efficiency factor can be represented by the total number of minutes of delay added to the entire route of the delivery capacity after the waybill is delivered; This can be represented by the total additional travel distance along the entire route of the transport capacity after the waybill is delivered; the transport capacity's willingness to accept orders factor. It can be predicted using the aforementioned order acceptance willingness prediction model.

[0066] Step 130: In response to the target waybill being matched with a waybill processing mode of dispatch mode, a preset dispatch engine is used to perform a dispatch operation on the target waybill.

[0067] Once the target waybill matching dispatch mode is determined, a preset dispatch engine can be used to perform the dispatch operation on the target waybill. The preset dispatch engine can be a dispatch engine from the prior art.

[0068] For specific implementation methods of using a preset dispatch engine to perform dispatch operations on the target waybill, please refer to the prior art, and will not be repeated in the embodiments of this application.

[0069] The waybill processing method disclosed in this application determines the waybill processing mode matching the target waybill; in response to the target waybill matching the waybill processing mode being a targeted recommendation mode, the target waybill is pushed to at least one designated recommended delivery capacity, so that the waybill processing client logged into the at least one designated recommended delivery capacity displays the target waybill on the recommended waybill list display page; in response to the target waybill matching the waybill processing mode being a dispatch mode, a preset dispatch engine is used to perform dispatch operations on the target waybill, which helps to improve the delivery experience while taking delivery quality into account.

[0070] The waybill processing method disclosed in this application dispatches waybills using different waybill dispatch modes based on their characteristics. When a waybill cannot find candidate capacity using the dispatch mode, it pushes the waybill to multiple designated capacity providers. Compared with the prior art of pushing waybills into a bidding pool, this method can effectively control the capacity to accept waybills, thereby improving delivery quality.

[0071] Furthermore, when using a targeted recommendation model for order dispatch, the order allocation optimization problem is modeled as a combinatorial optimization problem involving many-to-many matching of orders and riders. The global optimization objective is the quality of order delivery to riders, while the constraint is the range of riders reached by the order. This improves the delivery experience from two dimensions: increasing rider delivery efficiency and enhancing rider delivery quality. On one hand, targeted recommendation relaxes the order dispatch constraints, expanding the recall range of delivery capacity and facilitating further optimization of delivery capacity delivery quality. On the other hand, by recommending orders to multiple delivery capacities (e.g., K), the delivery capacity reach of the order is expanded, effectively improving rider delivery efficiency and increasing the scheduling system's tolerance for order rejections. Compared to existing technologies that rely on a localized decision-making perspective on delivery capacity and where delivery capacity is visible within a 3-kilometer radius of a delivery order, targeted recommendation introduces a global decision-making perspective from the platform to seek the overall optimal delivery capacity reach quality. Furthermore, by optimizing the list of riders who can recommend a delivery order, it targets and recommends the order to a specific delivery capacity, reasonably limiting the delivery capacity exposure range of the order, further ensuring the quality of riders reached, and optimizing the delivery experience.

[0072] On the other hand, by solving the combination optimization problem of many-to-many matching of waybills and transport capacity, at least one recommended transport capacity for the target waybill is determined. When determining the initial solution, the premise is to ensure the delivery quality and meet some constraints (such as the maximum number of waybills recommended for each transport capacity). When finding the optimal solution, a local search method is adopted, which can ensure the solution accuracy and solution efficiency.

[0073] Example 2

[0074] This application discloses a waybill processing method, such as... Figure 2 As shown, after the step of pushing the target waybill to at least one designated candidate capacity, the method includes: step 140.

[0075] Step 140: In response to the target waybill not being picked up by transportation capacity within a preset time period, the process jumps to the step of determining the waybill processing mode that matches the target waybill, so as to repeat the processing of the target waybill.

[0076] After a waybill is delivered to one or more transport capacities, if the waybill is not picked up by a transport capacity within the allowed time, the waybill will be pulled back to the backlog pool for rescheduling. The above-mentioned waybill dispatch mode selection and waybill dispatch process will be repeated to ensure the efficiency and quality of the waybill's reach to transport capacity.

[0077] Example 3

[0078] This application discloses a waybill processing device, such as... Figure 3 As shown, the device includes:

[0079] The waybill processing mode determination module 310 is used to determine the waybill processing mode that matches the target waybill;

[0080] The targeted recommendation module 320 is used to push the target waybill to at least one designated recommended transport capacity in response to the waybill processing mode matching the target waybill being a targeted recommendation mode, so that the waybill processing client logged into the at least one designated recommended transport capacity displays the target waybill on the recommended waybill list display page;

[0081] The dispatch module 330 is used to perform dispatch operations on the target waybill in response to the waybill processing mode being dispatch mode matched with the target waybill, using a preset dispatch engine.

[0082] In some embodiments of this application, the targeted recommendation module 320 is further configured to:

[0083] By solving the combination optimization problem of many-to-many matching between waybills and transport capacity, at least one recommended transport capacity for the target waybill is determined;

[0084] The target waybill is pushed to the at least one recommended transport capacity.

[0085] In some embodiments of this application, determining at least one recommended capacity for the target waybill by solving a combination optimization problem of many-to-many matching between waybills and capacity includes:

[0086] With the goal of optimizing the delivery quality of the target waybill, and constrained by the capacity reach range and capacity resource location limitations, a combinatorial optimization problem of many-to-many matching between waybills and capacity is constructed.

[0087] A local search for optimal solutions is employed to search for at least one recommended capacity for the target waybill among the candidate capacity recalled based on the second dispatch constraint.

[0088] In some embodiments of this application, the method of using local search for optimal solutions to search for at least one recommended capacity for the target waybill among candidate capacities includes:

[0089] The initialization sub-step is used to determine the initial number of waybills recommended to each candidate transportation capacity based on the resource location constraints of the transportation capacity and the number of waybills that can be recommended for each candidate transportation capacity, in descending order of delivery quality, so as to obtain the initial recommendation relationship between waybills and candidate transportation capacity;

[0090] The waybill set determination sub-step is used to determine a first waybill set, a second waybill set, and a third waybill set matching each candidate capacity based on a specified number of waybills initially recommended to each candidate capacity; wherein, each waybill in the first waybill set is a capacity-insufficient waybill recommended to less than a first number of candidate capacity, each waybill in the second waybill set is recommended to a capacity greater than or equal to the first number of candidate capacity, and the waybills in the third waybill set are: capacity-surplus waybills among the waybills recommended to candidate capacity matching the third waybill set, and the capacity-surplus waybills are waybills recommended to a capacity greater than the first number of candidate capacity;

[0091] The search optimization sub-step is used to perform a local capacity search for each of the insufficient capacity waybills in the first waybill set, in descending order of candidate capacity shortage, to search for candidate capacity to recommend the insufficient capacity waybills with the goal of optimizing the waybill delivery quality, and to update the recommendation relationship between waybills and candidate capacity based on the searched candidate capacity, as well as to update the first waybill set, the second waybill set, and the third waybill set matched by each candidate capacity;

[0092] An iterative judgment sub-step is used to jump to the execution of the search optimization sub-step in response to the failure to meet the iterative search termination condition; and, in response to the satisfaction of the iterative search termination condition, to determine at least one recommended capacity for the target waybill based on the updated recommendation relationship between the waybill and the candidate capacity.

[0093] In some embodiments of this application, the step of searching for candidate transport capacity among the candidate transport capacity to recommend the insufficient transport capacity for waybills with the objective of optimizing the delivery quality of the waybill includes:

[0094] Candidate capacity included in the candidate capacity of the waybill recalled based on the first dispatch constraint, and not included in the recommended capacity of the waybill, are identified as candidate recommended capacity.

[0095] The candidate recommended capacity is determined to be the one with the highest delivery quality for the recommended waybill with surplus capacity, which is lower than the delivery quality for the waybill with insufficient capacity, and is used as a recommended capacity for the waybill with insufficient capacity.

[0096] In some embodiments of this application, the waybill processing mode determination module 310 is further configured to:

[0097] Based on the target waybill's capacity audience range within the capacity sets recalled under the first and second dispatch constraints, the waybill processing mode matching the target waybill is determined; or,

[0098] Based on the distribution information of the willingness to accept orders in the capacity set recalled under the second dispatch constraint, the order processing mode matching the target order is determined.

[0099] The second dispatch constraint is the dispatch constraint obtained by relaxing the first dispatch constraint.

[0100] In some embodiments of this application, determining the matching waybill processing mode based on the capacity audience range in the capacity set recalled under the first dispatch constraint and the second dispatch constraint respectively includes:

[0101] If the number of transport capacities in the capacity set recalled under the first dispatch constraint for the target waybill is less than the number of transport capacities in the capacity set recalled under the second dispatch constraint, then the waybill processing mode matching the target waybill is determined.

[0102] In some embodiments of this application, determining the order processing mode matching the target order based on the order acceptance willingness distribution information of the capacity in the capacity set recalled under the second dispatch constraint condition includes:

[0103] If the average willingness to accept orders in the capacity set recalled under the second dispatch constraint for the target waybill is greater than or equal to the preset willingness to accept orders threshold, the waybill processing mode matching the target waybill is determined.

[0104] In some embodiments of this application, the waybill processing client that logs in with at least one designated candidate transport capacity displays the target waybill on the recommended waybill list display page, including:

[0105] The waybill processing client sorts and displays the received recommended waybills based on the ranking index of the target waybill. The ranking index of the target waybill is obtained by weighting the target waybill's delivery experience factor, delivery efficiency factor, and delivery capacity willingness factor.

[0106] In some embodiments of this application, such as Figure 4 As shown, the device further includes:

[0107] The waybill pullback processing module 340 is used to respond to the step of determining the waybill processing mode that matches the target waybill if the target waybill is not picked up by the transportation capacity within a preset time period, so as to repeat the processing of the target waybill.

[0108] The waybill processing device disclosed in this application is used to implement the waybill processing method described in Embodiment 1 and Embodiment 2 of this application. The specific implementation methods of each module of the device will not be repeated here. Please refer to the specific implementation methods of the corresponding steps in the method embodiments.

[0109] The waybill processing device disclosed in this application determines the waybill processing mode matching the target waybill; in response to the target waybill matching the waybill processing mode being a targeted recommendation mode, the target waybill is pushed to at least one designated recommended delivery capacity, so that the waybill processing client logged into the at least one designated recommended delivery capacity displays the target waybill on the recommended waybill list display page; in response to the target waybill matching the waybill processing mode being a dispatch mode, a preset dispatch engine is used to perform dispatch operations on the target waybill, which helps to improve the delivery experience while taking delivery quality into account.

[0110] The waybill processing device disclosed in this application dispatches waybills by adopting different waybill dispatch modes according to the characteristics of the waybill. When the waybill cannot find candidate capacity using the dispatch mode, it pushes the waybill to multiple designated capacity. Compared with the prior art of pushing waybills into the order-grabbing pool, it can effectively control the capacity for accepting waybills, thereby achieving the effect of improving delivery quality.

[0111] Furthermore, when using a targeted recommendation model for order dispatch, the order allocation optimization problem is modeled as a combinatorial optimization problem involving many-to-many matching of orders and riders. The global optimization objective is the quality of order delivery to riders, while the constraint is the range of riders reached by the order. This improves the delivery experience from two dimensions: increasing rider delivery efficiency and enhancing rider delivery quality. On one hand, targeted recommendation relaxes the order dispatch constraints, expanding the recall range of delivery capacity and facilitating further optimization of delivery capacity delivery quality. On the other hand, by recommending orders to multiple delivery capacities (e.g., K), the delivery capacity reach of the order is expanded, effectively improving rider delivery efficiency and increasing the scheduling system's tolerance for order rejections. Compared to existing technologies that rely on a localized decision-making perspective on delivery capacity and where delivery capacity is visible within a 3-kilometer radius of a delivery order, targeted recommendation introduces a global decision-making perspective from the platform to seek the overall optimal delivery capacity reach quality. Furthermore, by optimizing the list of riders who can recommend a delivery order, it targets and recommends the order to a specific delivery capacity, reasonably limiting the delivery capacity exposure range of the order, further ensuring the quality of riders reached, and optimizing the delivery experience.

[0112] On the other hand, by solving the combination optimization problem of many-to-many matching of waybills and transport capacity, at least one recommended transport capacity for the target waybill is determined. When determining the initial solution, the premise is to ensure the delivery quality and meet some constraints (such as the maximum number of waybills recommended for each transport capacity). When finding the optimal solution, a local search method is adopted, which can ensure the solution accuracy and solution efficiency.

[0113] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus embodiments, since they are fundamentally similar to the method embodiments, the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0114] The above provides a detailed description of a waybill processing method and apparatus provided by this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method of this application and its core idea. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the idea of ​​this application. Therefore, the content of this specification should not be construed as a limitation of this application.

[0115] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0116] The various component embodiments of this application can be implemented in hardware, or as software modules running on one or more processors, or a combination thereof. Those skilled in the art will understand that microprocessors or digital signal processors (DSPs) can be used in practice to implement some or all of the functions of some or all of the components in the electronic device according to the embodiments of this application. This application can also be implemented as a device or apparatus program (e.g., a computer program and computer program product) for performing part or all of the methods described herein. Such a program implementing this application can be stored on a computer-readable medium, or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.

[0117] For example, Figure 5An electronic device is shown that can implement the methods according to this application. The electronic device may be a PC, mobile terminal, personal digital assistant, tablet computer, etc. The electronic device conventionally includes a processor 510 and a memory 520, and program code 530 stored in the memory 520 and executable on the processor 510, which, when executing the program code 530, implements the methods described in the above embodiments. The memory 520 may be a computer program product or a computer-readable medium. The memory 520 may be an electronic memory such as flash memory, EEPROM (Electrically Erasable Programmable Read-Only Memory), EPROM, hard disk, or ROM. The memory 520 has a storage space 5201 for the program code 530 of a computer program for performing any of the method steps described above. For example, the storage space 5201 for the program code 530 may include various computer programs for implementing the various steps in the above methods. The program code 530 is computer-readable code. These computer programs can be read from or written to one or more computer program products. These computer program products include program code carriers such as hard disks, CDs, memory cards, or floppy disks. The computer program includes computer-readable code that, when executed on an electronic device, causes the electronic device to perform the method according to the above embodiments.

[0118] This application also discloses a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the waybill processing method as described in Embodiment 1 of this application.

[0119] Such a computer program product can be a computer-readable storage medium, which can have the same characteristics as... Figure 5 The memory 520 in the illustrated electronic device is similarly arranged with storage segments, storage spaces, etc. Program code can be stored, for example, in a compressed form on the computer-readable storage medium. The computer-readable storage medium is typically as shown in the reference. Figure 6 The portable or fixed storage unit is described above. Typically, the storage unit includes computer-readable code 530', which is code read by a processor and, when executed by the processor, implements the various steps of the method described above.

[0120] The terms "an embodiment," "embodiment," or "one or more embodiments" as used herein mean that a particular feature, structure, or characteristic described in connection with an embodiment is included in at least one embodiment of this application. Furthermore, please note that the examples of the phrase "in one embodiment" do not necessarily all refer to the same embodiment.

[0121] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of this application may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0122] In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. This application can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names.

[0123] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A waybill processing method, characterized in that, include: Determine the waybill processing mode to match the target waybill; In response to the target waybill matching, the waybill processing mode is a targeted recommendation mode. This involves solving a combination optimization problem of many-to-many matching between waybills and transport capacity, with the goal of optimizing the delivery quality of the target waybill, and constraints such as the reach of transport capacity and resource limitations. A local search method is used to search for at least one recommended transport capacity among the candidate transport capacities recalled based on the second dispatch constraint. The target waybill is then pushed to at least one designated recommended transport capacity, allowing the waybill processing client logged into the at least one designated recommended transport capacity to display the target waybill on the recommended waybill list display page. In response to the target waybill being matched with a waybill processing mode of dispatch mode, a preset dispatch engine is used to perform a dispatch operation on the target waybill: The step of searching for at least one recommended capacity for the target waybill among candidate capacity using the local search for optimal solutions includes: The initialization sub-step is used to determine the initial number of waybills recommended to each candidate transportation capacity based on the resource location constraints of the transportation capacity and the number of waybills that can be recommended for each candidate transportation capacity, in descending order of delivery quality, so as to obtain the initial recommendation relationship between waybills and candidate transportation capacity; The waybill set determination sub-step is used to determine a first waybill set, a second waybill set, and a third waybill set matching each candidate capacity based on a specified number of waybills initially recommended to each candidate capacity; wherein, each waybill in the first waybill set is a capacity-insufficient waybill recommended to less than a first number of candidate capacity, each waybill in the second waybill set is recommended to a capacity greater than or equal to the first number of candidate capacity, and the waybills in the third waybill set are: capacity-surplus waybills among the waybills recommended to candidate capacity matching the third waybill set, and the capacity-surplus waybills are waybills recommended to a capacity greater than the first number of candidate capacity; The search optimization sub-step is used to perform a local capacity search for each of the insufficient capacity waybills in the first waybill set, in descending order of candidate capacity shortage, to search for candidate capacity to recommend the insufficient capacity waybills with the goal of optimizing the waybill delivery quality, and to update the recommendation relationship between waybills and candidate capacity based on the searched candidate capacity, as well as to update the first waybill set, the second waybill set, and the third waybill set matched by each candidate capacity; An iterative judgment sub-step is used to jump to the execution of the search optimization sub-step in response to the failure to meet the iterative search termination condition; and, in response to the satisfaction of the iterative search termination condition, to determine at least one recommended capacity for the target waybill based on the updated recommendation relationship between the waybill and the candidate capacity.

2. The method according to claim 1, characterized in that, The step of searching for candidate transport capacity among the candidate transport capacity to recommend the insufficient transport capacity for the waybill with the goal of optimizing the delivery quality of the waybill includes: Candidate capacity included in the candidate capacity of the waybill recalled based on the first dispatch constraint, and not included in the recommended capacity of the waybill, are identified as candidate recommended capacity. The candidate recommended capacity is determined to be the one with the highest delivery quality for the recommended waybill with surplus capacity, which is lower than the delivery quality for the waybill with insufficient capacity, and is used as a recommended capacity for the waybill with insufficient capacity.

3. The method according to any one of claims 1 to 2, characterized in that, The step of determining the waybill processing mode that matches the target waybill includes: Based on the target waybill's capacity audience range within the capacity sets recalled under the first and second dispatch constraints, the waybill processing mode matching the target waybill is determined; or, Based on the distribution information of the willingness to accept orders in the capacity set recalled under the second dispatch constraint, the order processing mode matching the target order is determined. The second dispatch constraint is the dispatch constraint obtained by relaxing the first dispatch constraint.

4. The method according to claim 3, characterized in that, The step of determining the matching waybill processing mode for the target waybill based on the capacity audience range in the capacity set recalled under the first and second dispatch constraints respectively includes: If the number of transport capacities in the capacity set recalled under the first dispatch constraint for the target waybill is less than the number of transport capacities in the capacity set recalled under the second dispatch constraint, then the waybill processing mode matching the target waybill is determined.

5. The method according to claim 3, characterized in that, The step of determining the order processing mode matching the target order based on the order acceptance willingness distribution information of the capacity in the capacity set recalled under the second dispatch constraint condition includes: If the average willingness to accept orders in the capacity set recalled under the second dispatch constraint for the target waybill is greater than or equal to the preset willingness to accept orders threshold, the waybill processing mode matching the target waybill is determined.

6. The method according to any one of claims 1 to 2, characterized in that, The step of displaying the target waybill on the recommended waybill list display page by the waybill processing client logged in with at least one designated candidate transport capacity includes: The waybill processing client sorts and displays the received recommended waybills based on the ranking index of the target waybill. The ranking index of the target waybill is obtained by weighting the target waybill's delivery experience factor, delivery efficiency factor, and delivery capacity willingness factor.

7. The method according to any one of claims 1 to 2, characterized in that, After the step of pushing the target waybill to at least one designated candidate capacity, the method further includes: If the target waybill is not picked up by transportation capacity within a preset time period, the process jumps to the step of determining the waybill processing mode that matches the target waybill, so as to repeat the processing of the target waybill.

8. A waybill processing device, characterized in that, include: The waybill processing mode determination module is used to determine the waybill processing mode that matches the target waybill. The targeted recommendation module is used to respond to the target waybill matching waybill being in a targeted recommendation mode. It solves a combination optimization problem of many-to-many matching of waybills and transport capacity, aiming to optimize the delivery quality of the target waybill, and constrained by the capacity reach range and capacity resource limitations. The module employs a local search optimal solution method to search for at least one recommended transport capacity among the candidate transport capacities recalled based on the second dispatch constraint. The target waybill is then pushed to at least one designated recommended transport capacity, so that the waybill processing client logged into the at least one designated recommended transport capacity displays the target waybill on the recommended waybill list display page. The step of searching for at least one recommended transport capacity among the candidate transport capacities using the local search optimal solution method includes: The initialization sub-step is used to determine the initial number of waybills recommended to each candidate transportation capacity based on the resource location constraints of the transportation capacity and the number of waybills that can be recommended for each candidate transportation capacity, in descending order of delivery quality, so as to obtain the initial recommendation relationship between waybills and candidate transportation capacity; The waybill set determination sub-step is used to determine a first waybill set, a second waybill set, and a third waybill set matching each candidate capacity based on a specified number of waybills initially recommended to each candidate capacity; wherein, each waybill in the first waybill set is a capacity-insufficient waybill recommended to less than a first number of candidate capacity, each waybill in the second waybill set is recommended to a capacity greater than or equal to the first number of candidate capacity, and the waybills in the third waybill set are: capacity-surplus waybills among the waybills recommended to candidate capacity matching the third waybill set, and the capacity-surplus waybills are waybills recommended to a capacity greater than the first number of candidate capacity; The search optimization sub-step is used to perform a local capacity search for each of the insufficient capacity waybills in the first waybill set, in descending order of candidate capacity shortage, to search for candidate capacity to recommend the insufficient capacity waybills with the goal of optimizing the waybill delivery quality, and to update the recommendation relationship between waybills and candidate capacity based on the searched candidate capacity, as well as to update the first waybill set, the second waybill set, and the third waybill set matched by each candidate capacity; An iterative judgment sub-step is used to jump to the search optimization sub-step in response to the failure to meet the iterative search termination condition; and, in response to the satisfaction of the iterative search termination condition, to determine at least one recommended capacity for the target waybill based on the updated recommendation relationship between the waybill and the candidate capacity. The dispatch module is used to perform dispatch operations on the target waybill in response to the waybill processing mode matching the target waybill being dispatch mode, using a preset dispatch engine.

9. An electronic device, comprising a memory, a processor, and program code stored in the memory and executable on the processor, characterized in that, When the processor executes the program code, it implements the waybill processing method according to any one of claims 1 to 7.

10. A computer-readable storage medium having program code stored thereon, characterized in that, When the program code is executed by the processor, it implements the steps of the waybill processing method according to any one of claims 1 to 7.

Images