Order taking and sending position point sorting method and device and distribution path determining method and device

Abstract

The invention provides an order taking and sending position point sorting method, a distribution path determining method and device, electronic equipment and a computer storage medium. The current position point and the to-be-sorted delivery position points are used as nodes, and then the obtained graph features can better reflect the correlation relationship between the current position point and each delivery position point, so that the order delivery association relationship between the plurality of to-be-delivered orders can be obtained according to the graph features and the order delivery association relationship between the plurality of to-be-delivered orders. The obtained predicted sorting information of the to-be-sorted taking and sending position points can also reflect the correlation relationship among the position points; the order taking and delivering association relationship is considered when the prediction sorting information is obtained, and the order taking and delivering association relationship is also considered when the delivery transport capacity delivers a plurality of to-be-delivered orders, so that actually, the method considers the delivery behavior habits of the delivery transport capacity, and the subsequent delivery path determined based on the prediction sorting information is more practical.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and specifically to a method for sorting order pickup and delivery locations, a method for determining a delivery path, a device for sorting order pickup and delivery locations, a device for determining a delivery path, an electronic device, and a computer storage medium. Background Art

[0002] Ordering goods online is becoming more and more common, and the delivery process of goods ordered online is becoming more and more important.

[0003] During the delivery process, the delivery rider may have multiple orders to be delivered, and each order corresponds to a pickup location and a delivery location. Based on the sorting between the pickup locations and delivery locations of multiple orders to be delivered, the delivery rider can plan the delivery routes for the multiple orders to be delivered. In the existing method of obtaining the sorting information of the pickup locations and delivery locations of multiple orders to be delivered, the pickup locations and delivery locations are generally sorted according to the shortest time or shortest distance for the delivery rider to reach these pickup locations and delivery locations, and then the sorting information of the pickup locations and delivery locations of multiple orders to be delivered is obtained. However, in fact, the delivery route planned by the sorting information obtained in this way is quite different from the actual delivery route of the delivery rider to finally deliver the multiple orders to be delivered, which results in the delivery route planned in this way not being in line with the actual situation. Therefore, how to obtain the sorting information of the pickup locations and delivery locations of multiple orders to be delivered so that the final planned delivery route conforms to the actual delivery route has become a technical problem that needs to be solved urgently. Summary of the invention

[0004] The present application provides a method for sorting order pick-up and delivery locations to obtain sorting information of pick-up locations and delivery locations of multiple orders to be delivered so that the final planned delivery path conforms to the actual delivery path. The present application also provides a delivery path determination method, an order pick-up and delivery location sorting device, a delivery path determination device, an electronic device, and a computer storage medium.

[0005] The present application provides a method for sorting order pick-up and delivery locations, which is applied to a server, and the method includes:

[0006] In response to detecting a sorting request message sent by a distribution terminal for requesting to sort the to-be-sorted pick-up and delivery location points of a plurality of to-be-delivered orders requiring a target distribution transport capacity to perform pick-up and delivery tasks, determining the plurality of to-be-delivered orders;

[0007] Based on the delivery capacity characteristics of the target delivery capacity and the order characteristics of the multiple orders to be delivered, obtaining the graph characteristics of a graph with the current location of the target delivery capacity and the pickup and delivery location points to be sorted as nodes;

[0008] According to the graph features and the order pickup and delivery association relationship between the multiple orders to be delivered, the predicted sorting information of the pickup and delivery location points to be sorted is obtained; the order pickup and delivery association relationship includes at least one of an order merging relationship, an order-associated pickup relationship, and an order-associated delivery relationship;

[0009] The predicted sorting information is provided to the delivery terminal, so that the delivery terminal determines a delivery route for delivering the multiple orders to be delivered based on the current location, the pickup and delivery location to be sorted, and the predicted sorting information.

[0010] Optionally, the obtaining of graph features of a graph with the current location of the target delivery capacity and the pickup and delivery location points to be sorted as nodes based on the delivery capacity features of the target delivery capacity and the order features of the multiple orders to be delivered includes:

[0011] Obtaining the time characteristics of the distribution capacity associated with time in the distribution capacity characteristics and the space characteristics of the distribution capacity associated with space in the distribution capacity characteristics;

[0012] Obtaining order time features associated with time in the order features, and order space features associated with space in the order features;

[0013] The graph feature is obtained according to the delivery capacity time feature, the delivery capacity space feature, the order time feature, and the order space feature.

[0014] Optionally, the graph features include node features and edge features;

[0015] The obtaining of the graph feature according to the distribution capacity time feature, the distribution capacity space feature, the order time feature, and the order space feature includes:

[0016] If the node represents the current location point, the time feature of the delivery capacity is concatenated with the spatial feature of the delivery capacity to obtain the node feature;

[0017] If the node represents a pickup location, concatenate the first order time feature associated with pickup in the order time feature and the first order space feature associated with pickup in the order space feature to obtain the node feature;

[0018] If the node represents a delivery location point, concatenate the second order time feature associated with the delivery in the order time feature and the second order space feature associated with the delivery in the order space feature to obtain the node feature;

[0019] The edge feature is obtained according to the node feature and the node association relationship between nodes.

[0020] Optionally, obtaining the edge feature according to the node feature and the node association relationship between nodes includes:

[0021] For any pair of two nodes having a node association relationship, determine two target nodes having a node association relationship;

[0022] Based on the node characteristics of the two target nodes, determine the distance and theoretical time difference between the position points represented by the two target nodes, where the theoretical time difference is the theoretical time required from the position point represented by one target node to the position point represented by another target node;

[0023] The distance and the theoretical time difference are used as the edge features.

[0024] Optionally, also include:

[0025] Determine each pickup location point and each delivery location point of the plurality of orders to be delivered;

[0026] Determine the pickup location relationship between each pickup location point and the delivery location relationship between each delivery location point;

[0027] According to at least one of the pickup position relationship and the delivery position relationship, an order pickup and delivery association relationship between the multiple orders to be delivered is determined.

[0028] Optionally, determining the order pickup and delivery association relationship between the multiple orders to be delivered according to at least one of the pickup location relationship and the delivery location relationship includes:

[0029] For any two orders among the multiple orders to be delivered, determine whether the position relationship between the two pickup locations of the two orders meets the requirement of being adjacent to each other and whether the theoretical pickup time difference between the two orders is less than a first preset time difference; if yes, confirm that the two orders have an order-associated pickup relationship;

[0030] For any two orders among the multiple orders to be delivered, determine whether the positional relationship between the two delivery location points of the two orders meets the requirement of adjacent locations and whether the theoretical delivery time difference between the two orders is less than a second preset time difference; if so, confirm that the two orders have an order-related delivery relationship.

[0031] Optionally, also include:

[0032] For any two orders among the multiple orders to be delivered, determine whether the two orders are merged into one waybill;

[0033] If two orders are merged into one waybill, it is confirmed that there is an order merging relationship between the two orders.

[0034] Optionally, also include:

[0035] According to the order pick-up and delivery association relationship, an association relationship tag is added to the pick-up and delivery location points that have the order pick-up and delivery association relationship among the pick-up and delivery location points to be sorted, and the pick-up and delivery location points to which the association relationship tag is added and the pick-up and delivery location points to which the association relationship tag is not added are used as the processed pick-up and delivery location points to be sorted;

[0036] The obtaining, according to the graph features and the order pickup and delivery association relationship between the multiple orders to be delivered, predicted sorting information of the pickup and delivery location points to be sorted includes:

[0037] According to the graph features and the processed pick-up and delivery position points to be sorted, predicted sorting information of the pick-up and delivery position points to be sorted is obtained.

[0038] Optionally, adding association relationship tags to the pickup and delivery location points that have the order pickup and delivery association relationship among the pickup and delivery location points to be sorted according to the order pickup and delivery association relationship includes:

[0039] According to the order pickup and delivery association relationship, determining target orders to be delivered that have the order pickup and delivery association relationship among the multiple orders to be delivered;

[0040] Determine the pickup and delivery location points to be sorted that have the order pickup and delivery association relationship among the target orders to be delivered;

[0041] An association relationship label for indicating the type of the order pickup and delivery association relationship is added to the to-be-sorted pickup and delivery location point having the order pickup and delivery association relationship.

[0042] Optionally, obtaining predicted sorting information of the pick-up and delivery position points to be sorted according to the graph features and the processed pick-up and delivery position points to be sorted includes:

[0043] The graph features and the processed pick-up and delivery locations to be sorted are input into a trained target graph neural network model to obtain predicted sorting information of the pick-up and delivery locations to be sorted; the target graph neural network model is trained using the sample pick-up and delivery locations processed by adding sample labels, sample graph features, and actual sorting information of the sample pick-up and delivery locations; the sample labels refer to association labels added to the sample pick-up and delivery locations of multiple sample orders of the sample distribution capacity that have the order pick-up and delivery association relationship with each other; the sample graph features are obtained based on the sample distribution capacity features of the sample distribution capacity and the sample order features of the multiple sample orders, and the sample graph features refer to the graph features of the graph with the sample locations of the sample distribution capacity and the sample pick-up and delivery locations as nodes; the actual sorting information is the sorting information obtained by the sample distribution capacity after delivering the multiple sample orders.

[0044] Optionally, the target graph neural network model includes an encoder and a decoder; the graph features include node features and edge features, and the edge features are used to represent relationship features between nodes with node association relationships;

[0045] The step of inputting the graph features and the processed pick-up and delivery position points to be sorted into a trained target graph neural network model to obtain predicted sorting information of the pick-up and delivery position points to be sorted includes:

[0046] Input the graph features into an encoder to obtain a current location point feature vector for vector representation of the features of the current location point and a pick-up and delivery location point feature vector for vector representation of the features of the pick-up and delivery location points to be sorted;

[0047] The feature vector of the current location point, the feature vector of the pick-up and delivery location point and the processed pick-up and delivery location points to be sorted are input into the decoder to obtain the predicted sorting information of the pick-up and delivery location points to be sorted.

[0048] Optionally, in the step of inputting the feature vector of the current location point, the feature vector of the pick-up and delivery location point, and the processed pick-up and delivery location point to be sorted into the decoder to obtain the predicted sorting information of the pick-up and delivery location point to be sorted, the decoder obtains the predicted sorting information of the pick-up and delivery location point to be sorted in the following manner:

[0049] Based on the feature vector of the current location point, the feature vector of the pick-up and delivery location point and the processed pick-up and delivery location points to be sorted, the target pick-up and delivery location points that are ranked first in each round among the pick-up and delivery location points to be sorted are determined one by one; based on the confirmation order of the target pick-up and delivery location points, the predicted sorting information of the pick-up and delivery location points to be sorted is obtained.

[0050] Optionally, the step of determining the first target pick-up and delivery position points in each round of sorting among the pick-up and delivery position points to be sorted based on the feature vector of the current position point, the feature vector of the pick-up and delivery position point, and the processed pick-up and delivery position points to be sorted one by one; and obtaining the predicted sorting information of the pick-up and delivery position points to be sorted based on the confirmation order of the target pick-up and delivery position points, comprises:

[0051] According to the order pick-up-first-then-delivery pick-up and delivery strategy, determine the first candidate pick-up and delivery location point and the first non-candidate pick-up and delivery location point among the pick-up and delivery location points to be sorted; the candidate pick-up and delivery location points are location points among the pick-up and delivery location points to be sorted that can participate in sorting in each round of sorting;

[0052] Based on the vector relationship between the feature vector of the current location point and the feature vector of the pickup and delivery location point corresponding to the first candidate pickup and delivery location point, determine the pickup and delivery location point ranked first among the first candidate pickup and delivery location points as the first target pickup and delivery location point;

[0053] Determine a second candidate pick-up and delivery position point and a second non-candidate pick-up and delivery position point among the pick-up and delivery position points to be sorted;

[0054] Based on the processed pick-up and delivery position points to be sorted, it is determined whether there is a pick-up and delivery position point in the second candidate pick-up and delivery position points that has a corresponding association relationship tag added with the first target pick-up and delivery position point; if so, the pick-up and delivery position point in the second candidate pick-up and delivery position points that has a corresponding association relationship tag added with the first target pick-up and delivery position point is used as the second target pick-up and delivery position point;

[0055] Determine a third candidate pick-up and delivery position point and a third non-candidate pick-up and delivery position point among the pick-up and delivery position points to be sorted;

[0056] Based on the processed pick-up and delivery position points to be sorted, determine whether there is a pick-up and delivery position point in the third candidate pick-up and delivery position point that has a corresponding association relationship tag added with the second target pick-up and delivery position point; if yes, use the pick-up and delivery position point in the third candidate pick-up and delivery position point that has a corresponding association relationship tag added with the second target pick-up and delivery position point as the third target pick-up and delivery position point;

[0057] And so on; until all the pick-up and delivery position points in the to-be-sorted pick-up and delivery position points are sorted, the predicted sorting information is obtained.

[0058] Optionally, also include:

[0059] If there is no pick-up and delivery location point in the second candidate pick-up and delivery location points that has a corresponding association relationship tag added to the first target pick-up and delivery location point, then based on the vector relationship between the pick-up and delivery location point feature vector corresponding to the first target pick-up and delivery location point and the pick-up and delivery location point feature vector corresponding to the second candidate pick-up and delivery location point, determine the pick-up and delivery location point ranked first among the second candidate pick-up and delivery location points as the second target pick-up and delivery location point;

[0060] If there is no pick-up and delivery location point in the third candidate pick-up and delivery location points that has a corresponding association relationship tag added to the second target pick-up and delivery location point, then based on the vector relationship between the pick-up and delivery location point feature vector corresponding to the second target pick-up and delivery location point and the pick-up and delivery location point feature vector corresponding to the third candidate pick-up and delivery location point, determine the pick-up and delivery location point ranked first among the third candidate pick-up and delivery location points as the third target pick-up and delivery location point;

[0061] And so on; until all the pick-up and delivery position points in the to-be-sorted pick-up and delivery position points are sorted, the predicted sorting information is obtained.

[0062] Optionally, also include:

[0063] After determining the candidate pick-up and delivery location points and non-candidate pick-up and delivery location points in each round, the non-candidate pick-up and delivery location points are blocked by using a mask.

[0064] The present application provides a delivery path determination method, which is applied to a delivery terminal, and the method comprises:

[0065] Sending a sorting request message to a server for requesting sorting of the to-be-sorted pick-up and delivery locations of a plurality of to-be-delivered orders that require a target delivery capacity to perform pick-up and delivery tasks, so that the server determines the plurality of to-be-delivered orders in response to the request message; obtaining graph features of a graph with the current location of the target delivery capacity and the to-be-sorted pick-up and delivery locations as nodes based on the delivery capacity features of the target delivery capacity and the order features of the plurality of to-be-delivered orders; obtaining predicted sorting information of the to-be-sorted pick-up and delivery locations based on the graph features and the order pick-up and delivery association relationships between the plurality of to-be-delivered orders; the order pick-up and delivery association relationships at least include at least one of an order merging relationship, an order-associated pickup relationship, and an order-associated delivery relationship;

[0066] Obtaining the predicted ranking information provided by the server;

[0067] Based on the current location point, the pickup and delivery location points to be sorted and the predicted sorting information, a delivery path for delivering the multiple orders to be delivered is determined.

[0068] Optionally, determining a delivery path for delivering the multiple orders to be delivered based on the current location, the pickup and delivery locations to be sorted, and the predicted sorting information includes:

[0069] According to the order corresponding to the predicted order information, the pick-up and delivery location points to be ordered are arranged after the current location point in sequence, and each location point after the order is determined;

[0070] Path planning is performed based on the sorted location points to determine a delivery path for delivering the multiple orders to be delivered.

[0071] The present application provides an order pickup and delivery location point sorting device, which is applied to a service end, and the device includes:

[0072] A to-be-delivered order determination unit, configured to determine the multiple to-be-delivered orders in response to detecting a sorting request message sent by a delivery terminal for requesting to sort the to-be-sorted pick-up and delivery location points of the multiple to-be-delivered orders that require a target delivery transport capacity to perform pick-up and delivery tasks;

[0073] A graph feature obtaining unit, configured to obtain graph features of a graph having a current location of the target delivery capacity and the pickup and delivery location points to be sorted as nodes based on the delivery capacity features of the target delivery capacity and the order features of the multiple orders to be delivered;

[0074] A first prediction sorting information obtaining unit is used to obtain prediction sorting information of the pickup and delivery location points to be sorted according to the graph features and the order pickup and delivery association relationship between the multiple orders to be delivered; the order pickup and delivery association relationship includes at least one of an order merging relationship, an order-associated pickup relationship, and an order-associated delivery relationship;

[0075] The predicted sorting information providing unit is used to provide the predicted sorting information to the delivery terminal, so that the delivery terminal determines the delivery path for delivering the multiple orders to be delivered based on the current location point, the pick-up and delivery location point to be sorted and the predicted sorting information.

[0076] The present application provides a delivery path determination device, which is applied to a delivery terminal, and the device includes:

[0077] A sorting request message sending unit is used to send a sorting request message to a server for requesting sorting of the to-be-sorted pick-up and delivery locations of multiple to-be-delivered orders that require a target delivery capacity to perform pick-up and delivery tasks, so that the server determines the multiple to-be-delivered orders in response to the request message; based on the delivery capacity characteristics of the target delivery capacity and the order characteristics of the multiple to-be-delivered orders, obtains the graph characteristics of a graph with the current location of the target delivery capacity and the to-be-sorted pick-up and delivery locations as nodes; based on the graph characteristics and the order pick-up and delivery association relationship between the multiple to-be-delivered orders, obtains the predicted sorting information of the to-be-sorted pick-up and delivery locations; the order pick-up and delivery association relationship includes at least one of an order merging relationship, an order-associated pickup relationship, and an order-associated delivery relationship;

[0078] A second prediction ranking information obtaining unit, configured to obtain the prediction ranking information provided by the server;

[0079] A delivery path determination unit is used to determine a delivery path for delivering the multiple orders to be delivered based on the current location point, the pickup and delivery location points to be sorted and the predicted sorting information.

[0080] The present application provides an electronic device, including: a processor; a memory for storing a computer program, wherein the computer program is run by the processor to execute the above-mentioned order pickup and delivery location point sorting method and delivery route determination method.

[0081] The present application provides a computer storage medium, which stores a computer program. The computer program is run by a processor to execute the above-mentioned order pickup and delivery location point sorting method and delivery path determination method.

[0082] Compared with the prior art, the embodiments of the present application have the following advantages:

[0083] The present application provides a method for sorting order pick-up and delivery location points, which is applied to a service end. In this method, in response to detecting a sorting request message sent by a distribution terminal for requesting to sort the to-be-sorted pick-up and delivery location points of multiple to-be-delivered orders that require a target distribution capacity to perform pick-up and delivery tasks, multiple to-be-delivered orders are determined; then, based on the distribution capacity characteristics of the target distribution capacity and the order characteristics of the multiple to-be-delivered orders, graph features of a graph with the current location point of the target distribution capacity and the to-be-sorted pick-up and delivery location points as nodes are obtained; the graph features obtained by taking the current location point and the to-be-sorted pick-up and delivery location points as nodes can better reflect the mutual correlation between the current location point and each pick-up and delivery location point, and then the graph features and the order characteristics of the multiple to-be-delivered orders are subsequently sorted. The order pick-up and delivery association relationship between them, the obtained predicted sorting information of the pick-up and delivery location points to be sorted can also reflect the mutual association relationship between each location point; at the same time, the order pick-up and delivery association relationship includes at least one of the order merging relationship, the order-related pickup relationship, and the order-related delivery relationship. Since the order pick-up and delivery association relationship is taken into account when obtaining the above-mentioned predicted sorting information, and the delivery capacity will also consider the order pick-up and delivery association relationship when delivering multiple orders to be delivered, in fact, this method takes into account the delivery behavior habits of the delivery capacity, that is, the obtained predicted sorting information is in line with the behavior habits of the delivery riders in the actual delivery process, so that the subsequent delivery path determined based on the predicted sorting information is more in line with reality; at the same time, it greatly facilitates the delivery process of the delivery riders, thereby improving the delivery efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present application. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0085] Figure 1 This is a schematic diagram of the first scenario of the order pickup and delivery location point sorting method of the present application;

[0086] Figure 2 A flowchart of a method for sorting order pickup and delivery locations provided in the first embodiment of the present application;

[0087] Figure 3 A schematic diagram of a second scenario of the order pickup and delivery location point sorting method of the present application;

[0088] Figure 4 A schematic diagram of a third scenario of the order pickup and delivery location point sorting method of the present application;

[0089] Figure 5 A flow chart of a method for determining a delivery path provided in a second embodiment of the present application;

[0090] Figure 6 A schematic diagram of an order pickup and delivery location point sorting device provided in a third embodiment of the present application;

[0091] Figure 7 A schematic diagram of a delivery path determination device provided in a fourth embodiment of the present application;

[0092] Figure 8 A schematic diagram of an electronic device provided in accordance with the fifth embodiment of the present application. DETAILED DESCRIPTION

[0093] Many specific details are described in the following description to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present application. Therefore, the present application is not limited by the specific implementation disclosed below.

[0094] The present application provides a method for sorting order pickup and delivery location points, a method for determining a delivery path, a device for sorting order pickup and delivery location points, a device for determining a delivery path, an electronic device, and a computer storage medium. The following describes the method for sorting order pickup and delivery location points, the method for determining a delivery path, the device for sorting order pickup and delivery location points, the device for determining a delivery path, an electronic device, and a computer storage medium through specific embodiments. In order to more clearly understand the method for sorting order pickup and delivery location points provided in the embodiments of the present application, the application scenario of the method for sorting order pickup and delivery location points provided in the embodiments of the present application is first introduced.

[0095] The order pick-up and delivery location point sorting method of the present application can be used in the scenario of sorting the pick-up and delivery location points of multiple orders to be delivered by delivery riders to determine the delivery paths of multiple orders to be delivered. Specifically, multiple orders to be delivered are orders for which a delivery rider needs to perform pick-up and delivery tasks, and each of the multiple orders to be delivered needs to be picked up and delivered, that is, each order corresponds to a pick-up location point and a delivery location point. When the delivery rider is delivering multiple orders to be delivered, it is necessary to sort the pick-up location points and the delivery location points of the multiple orders to be delivered, and then determine the delivery paths for the multiple orders to be delivered based on the sorting information of the pick-up location points and the delivery location points of the multiple orders to be delivered. The pick-up location points and the delivery location points of the multiple orders to be delivered that need to be sorted are all pick-up and delivery location points to be sorted.

[0096] For example, suppose that the delivery rider A has been assigned orders to be delivered O0, O1, O2 and O3; suppose that the pickup locations of these four orders to be delivered are P0, P1, P2 and P3, and the delivery locations are D0, D1, D2 and D3; P0, P1, P2, P3, D0, D1, D2, D3 are the pickup and delivery locations to be sorted, and it is necessary to sort the above pickup and delivery locations P0, P1, P2, P3, D0, D1, D2, D3 to plan the delivery routes for the delivery orders O0, O1, O2 and O3. In this application, for the same order, the sorting of the pickup locations takes precedence over the sorting of the delivery locations. For example, for order O0, P0 needs to be ranked ahead of D0.

[0097] Of course, it is understandable that when sorting the orders to be delivered, it is possible that delivery rider A has already picked up the goods for a certain order to be delivered. In this case, only the delivery location of the order can be used as the pickup location to be sorted. For example, when sorting the pickup locations of orders O0, O1, O2 and O3 that delivery rider A needs to deliver, if delivery rider A has already picked up the goods for order O0, then only P1, P2, P3, D0, D1, D2, and D3 need to be used as the pickup locations to be sorted.

[0098] In the scenario embodiment of the present application, the sorting of P0, P1, P2, P3, D0, D1, D2, and D3 is introduced as an example. Assuming that the current location of the delivery rider A is R0, in fact, in the process of sorting P0, P1, P2, P3, D0, D1, D2, and D3, the time-space relationship between the current location of the delivery rider A, RO, and P0, P1, P2, P3, D0, D1, D2, and D3 is mainly considered. And based on the time-space relationship, P0, P1, P2, P3, D0, D1, D2, and D3 are sorted. The so-called time-space relationship can actually refer to the consideration of the distance and theoretical time difference between two location points.

[0099] Each location point P0, P1, P2, P3, D0, D1, D2, and D3 corresponds to a theoretical arrival time. The theoretical arrival time is the time corresponding to the location of the merchant or the location of the ordering user promised by the user after placing the order. For example, for order O0, if its theoretical pickup time is 11:30 and the theoretical delivery time is 11:55, delivery rider A is required to arrive at location point P0 before 11:30, and delivery rider A is required to arrive at location point D0 before 11:55. The delivery rider picks up the goods at the location corresponding to the merchant and completes the delivery at the location of the ordering user. For example, in the field of takeout, at the meal pickup node, the theoretical arrival time is the merchant's estimated meal delivery time; at the meal delivery node, the theoretical arrival time is the delivery time of the delivery rider required by the platform.

[0100] For each of the above-mentioned pick-up and delivery locations to be sorted (P0, P1, P2, P3, D0, D1, D2, D3) and the current location point R0, its corresponding geographical location information can be obtained, and the distance between each location point can be obtained based on the geographical location information of each location point.

[0101] When sorting the pickup and delivery locations to be sorted, the sorting is mainly based on the spatiotemporal relationship between the locations. For example, when the delivery rider is currently at R0, assuming that order O0 requires the earliest pickup time among the four orders, and P0 is closer to the current location R0, P0 can be ranked first, and then the sorting of the pickup and delivery locations to be sorted is P1, D0, P2, D1, P3, D2, D3. Finally, the sorting of the pickup and delivery locations to be sorted and the current location is R0, P0, P1, D0, P2, D1, P3, D2, D3, respectively.

[0102] When sorting the various pickup and delivery locations to be sorted, the logical relationship of picking up the goods before delivering the goods must be followed for the same order, and the order pickup and delivery association relationship between the various orders to be delivered is not considered. When the delivery rider actually delivers multiple orders to be delivered, the order pickup and delivery association relationship between the orders will be considered. For example, when two orders are orders from the same merchant, when the delivery rider picks up the goods, even if the pickup locations of the two orders are not adjacent after sorting the various pickup and delivery locations to be sorted using the above method, the delivery rider will pick up the goods of the two orders together when picking up the goods. Therefore, the sorting information obtained by sorting the various pickup and delivery locations to be sorted using this method does not conform to the actual situation. In short, in the existing method of sorting the various pickup and delivery locations to be sorted, two or more pickup locations of the same merchant order cannot be guaranteed to be adjacent when sorting; and the solution of the present application ensures that two or more pickup locations of the same merchant order are adjacent when sorting by considering the characteristics of the order pickup and delivery association relationship.

[0103] In order to make the predicted sorting information of the to-be-sorted pick-up and delivery locations of multiple orders to be delivered predicted by the order pick-up and delivery location sorting method of the present application more in line with the actual situation, in the present application, when predicting the predicted sorting information of the to-be-sorted pick-up and delivery locations of multiple orders to be delivered, not only the graph characteristics of the graph with the current location of the target delivery capacity and the to-be-sorted pick-up and delivery locations as nodes are obtained based on the delivery capacity characteristics of the target delivery capacity and the order characteristics of the multiple orders to be delivered; at the same time, the predicted sorting information of the to-be-sorted pick-up and delivery locations is obtained based on the graph characteristics and the order pick-up and delivery association relationship between the multiple orders to be delivered; the order pick-up and delivery association relationship includes at least one of the following relationships: an order merging relationship, an order-associated pickup relationship, and an order-associated delivery relationship; because in the present scenario embodiment, in the process of obtaining the predicted sorting information, the order pick-up and delivery association relationship between the multiple orders to be delivered is taken into account, in fact, the method takes into account the delivery behavior habits of the delivery capacity, that is, the obtained predicted sorting information is in line with the behavior habits of the delivery riders in the actual delivery process, and the subsequent delivery path determined based on the predicted sorting information is also more in line with reality. It should be noted that in the present application, when sorting the to-be-sorted pick-up and delivery locations of multiple orders to be delivered, it is necessary to consider the order pick-up and delivery association relationship between the multiple orders to be delivered, but it is not required that there must be an order pick-up and delivery association relationship between the multiple orders to be delivered.

[0104] For example, assuming that in the above-mentioned orders to be delivered, O0 and O1 have an order-associated delivery relationship, and O0 and O2 have an order-merged relationship, then after picking up order O0, the next step is to pick up order O2, that is, P2 needs to be sorted after P0; at the same time, after order O0 is delivered, the next step is to deliver order O1, that is, D1 needs to be sorted after D0; and then the final sorting of the pick-up and delivery locations to be sorted and the current location points are R0, P0, P2, P1, D0, D1, P3, D2, D3 respectively.

[0105] To understand the order pick-up and delivery location sorting method for this application, please refer to Figure 1 , which is a schematic diagram of the first scenario of the order pickup and delivery location point sorting method of the present application. In this scenario, taking the order pickup and delivery location point sorting method executed on the server as an example, the so-called server is a computing device used to provide data processing, storage and other services for the user terminal. Generally, the server can refer to a server or a server cluster. The user terminal is generally an electronic device that is convenient for the user to directly operate. In this scenario, the user terminal can be a delivery terminal used by a delivery rider.

[0106] Specifically, see Figure 1, the following process is performed through the server: first, multiple to-be-delivered orders that require the target delivery capacity to perform pick-up and delivery tasks are determined; multiple to-be-sorted pick-up and delivery locations that require pick-up and delivery tasks are determined among the multiple to-be-delivered orders; then, based on the delivery capacity characteristics of the target delivery capacity and the order characteristics of the multiple to-be-delivered orders, the graph features of the graph with the current location of the target delivery capacity and the to-be-sorted pick-up and delivery locations as nodes are obtained; then, based on the graph features and the order pick-up and delivery association relationships between the multiple to-be-delivered orders, the predicted sorting information of the to-be-sorted pick-up and delivery locations is obtained.

[0107] When the delivery terminal needs to determine the delivery paths of multiple orders to be delivered and to navigate as the target delivery capacity based on the determined delivery paths, the delivery terminal sends a sorting request message to the server for requesting to sort the pick-up and delivery location points to be sorted of the multiple orders to be delivered that require the target delivery capacity to perform the pick-up and delivery tasks. After obtaining the sorting request message, the server determines the multiple orders to be delivered; thereafter, the predicted sorting information of the pick-up and delivery location points to be sorted is obtained according to the above process, and the predicted sorting information of the pick-up and delivery location points to be sorted is provided to the delivery terminal, so that the delivery terminal determines the delivery path for delivering the multiple orders to be delivered based on the current location, the pick-up and delivery location points to be sorted and the predicted sorting information.

[0108] The above-mentioned Figure 1 That is, it is a diagram of an application scenario of the order pickup and delivery location point sorting method of the present application. The application scenario of the order pickup and delivery location point sorting method is not specifically limited in the embodiments of the present application. The application scenario of the above-mentioned order pickup and delivery location point sorting method is only an embodiment of the application scenario of the order pickup and delivery location point sorting method provided by the present application. The purpose of providing this application scenario embodiment is to facilitate the understanding of the order pickup and delivery location point sorting method provided by the present application, and it is not used to limit the order pickup and delivery location point sorting method provided by the present application. Other application scenarios of the order pickup and delivery location point sorting method in the embodiments of the present application are not described one by one.

[0109] First embodiment

[0110] The first embodiment of the present application provides a method for sorting order pick-up and delivery locations. The execution subject of this embodiment is a server, and for some specific examples or details of this embodiment, please refer to the above scenario embodiment.

[0111] Please refer to Figure 2 , which is a flow chart of the order pickup and delivery location point sorting method provided in the first embodiment of the present application.

[0112] The order pickup and delivery location point sorting method of the embodiment of the present application is applied to the server, and includes the following steps:

[0113] Step S201: In response to detecting a sorting request message sent by a distribution terminal for requesting to sort the to-be-sorted pick-up and delivery location points of a plurality of to-be-delivered orders requiring a target distribution transport capacity to perform pick-up and delivery tasks, a plurality of to-be-delivered orders are determined.

[0114] The pickup and delivery location points to be sorted are the location points that require the target delivery capacity to perform the pickup and delivery tasks; the pickup and delivery location points to be sorted include at least one of the pickup location points and the delivery location points; the number of the pickup and delivery location points to be sorted is multiple. In this embodiment, it can be understood that the delivery capacity is not limited to delivery riders, but can also be delivery machines such as delivery unmanned vehicles.

[0115] The target delivery capacity is delivery rider A. Delivery rider A may need to deliver multiple orders at the same time, that is, multiple orders to be delivered. In order to plan the delivery routes for the delivery rider to deliver multiple orders to be delivered, for example, before delivery rider A delivers orders O0, O1, O2 and O3, it is necessary to plan the delivery routes for delivery rider A in advance, and display the delivery routes on the delivery terminal for the delivery rider to refer to during the delivery of orders O0, O1, O2 and O3.

[0116] In the present application, the order pick-up and delivery location point sorting method is actually used to make the predicted sorting information of multiple pick-up and delivery location points to be sorted finally obtained more in line with reality, so that the delivery route planned based on the predicted sorting information enables the delivery rider to save time and effort during the delivery process.

[0117] Step S202: Based on the delivery capacity characteristics of the target delivery capacity and the order characteristics of multiple orders to be delivered, the graph characteristics of a graph with the current location of the target delivery capacity and the pickup and delivery location points to be sorted as nodes are obtained.

[0118] In this embodiment, when planning the delivery routes for the delivery rider to deliver multiple orders to be delivered, the planning is actually carried out in the order from the delivery rider's current location to the pick-up and delivery locations of each order to be delivered. Therefore, the pick-up and delivery locations of each order to be delivered need to be sorted as pick-up and delivery locations to be sorted.

[0119] Specifically, in the process of sorting multiple pick-up and delivery locations to be sorted, the sorting is actually based on the relevant characteristics of the multiple pick-up and delivery locations to be sorted, and the relevant characteristics of the multiple pick-up and delivery locations to be sorted are reflected through the delivery capacity characteristics of the target delivery capacity and the order characteristics of the multiple orders to be delivered. Of course, it can also be reflected through the delivery environment characteristics of the multiple orders to be delivered.

[0120] In this embodiment, when sorting multiple pick-up and delivery locations to be sorted, the current location of the delivery rider and the multiple pick-up and delivery locations to be sorted are used as graph nodes to construct a graph (the graph represents the relationship graph between nodes), and then based on the delivery capacity characteristics of the target delivery capacity and the order characteristics of the multiple orders to be delivered, the graph features of the graph with the current location of the target delivery capacity and the pick-up and delivery locations to be sorted as nodes are obtained. The graph features include node features and edge features. The edge features refer to the features of each edge of the constructed graph, and the edges are used to represent the relationship between adjacent nodes.

[0121] The order features include at least one of the following features: order weight, quantity of goods in the order, price of goods in the order, order type, merchant ID (Identity Document) of the merchant to which the order belongs, interest surface ID of the interest surface to which the merchant location belongs (an interest surface is composed of polygonal fence boundaries and feature data, such as a community, a shopping mall, and a school can all be an interest surface, and the interest surface is represented by AOI, which stands for Area Of Interest), interest surface ID of the interest surface to which the ordering user location belongs, the promised shipping time of the order, the promised delivery time of the order, and the first ETS time (Estimated Time of Stay, estimated stay time, such as the time it takes for a delivery rider to enter the interest surface of the merchant location and leave the interest surface of the merchant location when picking up the goods. Generally speaking, the estimated stay time actually corresponds to the time it takes for a delivery rider to walk in the interest surface area), the second ETS time corresponding to the order delivery (such as the time it takes for a delivery rider to enter the interest surface of the ordering user location and leave the interest surface of the ordering user location when delivering the goods), order allocation time, order pickup location, order delivery location and other information; delivery capacity characteristics include at least one of the following characteristics: the current location information of the delivery rider, the upper limit of the number of orders that the delivery rider can carry, the delivery speed, the delivery capacity level, and the delivery history timeout information, etc. Of course, it can be understood that the above are only examples of some order characteristics and delivery capacity characteristics, and the order characteristics and delivery capacity characteristics are not limited to the listed situations.

[0122] The delivery environment characteristics mentioned above include at least one of the following characteristics: road conditions, weather characteristics, holiday characteristics, etc. related to delivery. For example, during peak hours in the morning and evening, due to the large number of orders, the delivery speed of the delivery rider is slower; similarly, rainy and snowy weather may also cause the delivery speed of the delivery rider to be slower.

[0123] In this embodiment, based on the delivery capacity characteristics of the target delivery capacity and the order characteristics of multiple orders to be delivered, graph characteristics of a graph with the current location of the target delivery capacity and the pickup and delivery locations to be sorted as nodes are obtained, including: obtaining delivery capacity time characteristics associated with time in the delivery capacity characteristics, and delivery capacity space characteristics associated with space in the delivery capacity characteristics; obtaining order time characteristics associated with time in the order characteristics, and order space characteristics associated with space in the order characteristics; obtaining graph characteristics based on the delivery capacity time characteristics, the delivery capacity space characteristics, the order time characteristics, and the order space characteristics.

[0124] To help you understand the graph formed by the current location and the pickup and delivery locations to be sorted as nodes, and how to use the graph features and the order pickup and delivery associations between multiple orders to be delivered to sort the pickup and delivery locations to be sorted, see Figure 3 , which is a schematic diagram of the second scenario of the order pick-up and delivery location point sorting method of the present application, Figure 3 The diagram takes delivery rider A as the target delivery capacity as an example.

[0125] First, the delivery capacity characteristics of delivery rider A, the order characteristics of multiple orders to be delivered, and the environmental characteristics are obtained; the delivery capacity characteristics, the order characteristics of multiple orders to be delivered, and the environmental characteristics can be used as the original characteristics; then, the delivery capacity characteristics and the order characteristics of multiple orders to be delivered are split and combined according to the time dimension and the space dimension respectively, and then the time characteristics and space characteristics are obtained. The time characteristics, space characteristics, and environmental characteristics are used together as the processed characteristics. In fact, the environmental characteristics do not need to be split and combined in this process.

[0126] The delivery capacity time characteristics associated with time in the delivery capacity characteristics can be used as time characteristics, and the order time characteristics associated with time in the order characteristics can also be used as time characteristics; time characteristics include: the promised shipment time of the order, the promised delivery time of the order, the first ETS time corresponding to the order pickup, the second ETS time corresponding to the order delivery, and the order allocation time; the delivery capacity space characteristics associated with space in the delivery capacity characteristics can be used as space characteristics, and the order space characteristics associated with space in the order characteristics can also be used as space characteristics; space characteristics include: the order pickup location, the order delivery location, and the current location information of the delivery rider.

[0127] Graph features include node features and edge features; as a way to obtain graph features based on the time features of delivery capacity, the space features of delivery capacity, the time features of order, and the space features of order: if the node represents the current location, the time features of delivery capacity and the space features of delivery capacity are concatenated to obtain the node features; of course, if the environmental features are related to the current location (the environmental features are related to the current location, which means that the features of the current location are affected by the environmental features), the time features of delivery capacity, the space features of delivery capacity, and the environment features are concatenated together to obtain the node features. For the current location, the time features of delivery capacity, the space features of delivery capacity, and the environment features each account for a certain proportion (i.e., weight). When obtaining the features of the current location, the time features of delivery capacity, the space features of delivery capacity, and the environment features are weighted summed with their respective weights to obtain the features of the current location.

[0128] As another way to obtain graph features based on the time characteristics of delivery capacity, the space characteristics of delivery capacity, the time characteristics of orders, and the space characteristics of orders: if the node represents a pickup location, then the first order time feature associated with pickup in the order time feature and the first order space feature associated with pickup in the order space feature are concatenated to obtain the node feature; if the node represents a delivery location, then the second order time feature associated with delivery in the order time feature and the second order space feature associated with delivery in the order space feature are concatenated to obtain the node feature; of course, if the environmental characteristics are related to the pickup location, when obtaining the features of the pickup location, the first order time feature associated with pickup in the order time feature and the first order space feature associated with pickup in the order space feature are concatenated with the environmental characteristics to obtain the node feature; similarly, if the environmental characteristics are related to the delivery location, when obtaining the features of the delivery location, the second order time feature associated with delivery in the order time feature and the second order space feature associated with delivery in the order space feature are concatenated with the environmental characteristics to obtain the node feature.

[0129] Of course, in fact, when obtaining node features, some original features may be unrelated to time, space and environment. At this time, they can also be considered as the weights they occupy in each node, and when obtaining the node features, they can be spliced ​​together to obtain the node features.

[0130] After obtaining the node features, the edge features are obtained according to the node features and the node association relationships between the nodes.

[0131] Specifically, edge features are obtained based on node features and node association relationships between nodes, including: for any pair of two nodes with a node association relationship, two target nodes with a node association relationship are determined; based on the node features of the two target nodes, the distance and theoretical time difference between the location points represented by the two target nodes are determined, the theoretical time difference being the theoretical time required to reach the location point represented by one target node from another target node; and the distance and theoretical time difference are used as edge features.

[0132] See also Figure 3 , Figure 3 The relationship diagram shown in the figure is a diagram constructed based on the current location point (R0) of the delivery rider A and the pickup and delivery locations to be sorted (P0, P1, P2, P3, D0, D1, D2, D3) as nodes. In this relationship diagram, the nodes with connecting lines have a node association relationship, such as the node represented by R0 has a node association relationship with the node represented by P0, and the node represented by R0 has a node association relationship with the node represented by P3. It should be noted that Figure 3 The connection relationship between the nodes is determined based on the scene needs. In fact, in other scenarios, Figure 3 The nodes in the may be connected to each other, which is not limited here.

[0133] When obtaining edge features, obtain the features corresponding to the edges between all nodes with node association relationships in the relationship graph. For example, for the edge between the node represented by R0 and the node represented by P0, when obtaining the edge feature, obtain the theoretical time required for delivery rider A to get from the R0 position to the P0 position and the distance between the R0 position and the P0 position. Assuming that the current time point is 11:15, and the theoretical pickup time point for order O0 is 11:30, and the distance from the R0 position to the P0 position is 1.5 kilometers, then the distance corresponding to the edge feature is 1.5 kilometers and the theoretical time is 15 minutes.

[0134] Since the node features corresponding to the R0 position point and the P0 position point include the geographic location information corresponding to the R0 position point and the P0 position point, the distance between the R0 position point and the P0 position point can be obtained; at the same time, the node features corresponding to the R0 position point and the P0 position point include the current time information and the theoretical pickup time point information of the order O0, so the theoretical time required for the delivery rider A to get from the R0 position point to the P0 position point can be obtained.

[0135] Step S203: obtaining predicted sorting information of the pick-up and delivery locations to be sorted according to the graph features and the order pick-up and delivery association relationships among the multiple orders to be delivered.

[0136] In fact, obtaining predicted sorting information of the pick-up and delivery locations to be sorted based on graph features and the order pick-up and delivery association relationships between multiple orders to be delivered may mean: determining the time relationship and spatial relationship between the current location and each location point in the pick-up and delivery locations to be sorted based on graph features; obtaining predicted sorting information of the pick-up and delivery locations to be sorted based on the time relationship, spatial relationship and order pick-up and delivery association relationship.

[0137] The order pickup and delivery association relationship includes at least one of an order merging relationship, an order-associated pickup relationship, and an order-associated delivery relationship.

[0138] In this embodiment, in order to know whether there is an order pick-up and delivery relationship between multiple orders to be delivered, it is necessary to predetermine whether there is an order pick-up and delivery relationship between multiple orders to be delivered. As a way to predetermine whether there is an order pick-up and delivery relationship between multiple orders to be delivered, it can be: first, determine each pickup location point and each delivery location point of the multiple orders to be delivered; then, determine the pickup position relationship between each pickup location point and the delivery position relationship between each delivery location point; finally, based on at least one position relationship among the pickup position relationship and the delivery position relationship, determine the order pick-up and delivery relationship between the multiple orders to be delivered.

[0139] More specifically, determining the order pickup and delivery association relationship between multiple orders to be delivered based on at least one of the pickup location relationship and the delivery location relationship may mean: for any two orders among the multiple orders to be delivered, judging whether the location relationship between the two pickup location points of the two orders meets the requirement of being adjacent to each other and whether the theoretical pickup time difference between the two orders is less than a first preset time difference; if so, confirming that the two orders have an order-associated pickup relationship; at the same time, for any two orders among the multiple orders to be delivered, judging whether the location relationship between the two delivery location points of the two orders meets the requirement of being adjacent to each other and whether the theoretical delivery time difference between the two orders is less than a second preset time difference; if so, confirming that the two orders have an order-associated delivery relationship.

[0140] If the pickup locations corresponding to the two orders are less than a preset distance (such as 1 km), the location relationship between the two pickup locations is adjacent. The first preset time difference and the second preset time difference may be the same or different.

[0141] In this embodiment, as another way to predetermine whether there is an order pickup and delivery association relationship between multiple orders to be delivered, it can also refer to: for any two orders among the multiple orders to be delivered, determine whether the two orders are merged into one waybill; if the two orders are merged into one waybill, confirm that the two orders have an order merging relationship. It should be noted that the merging of two or more orders into one waybill is handled by the order scheduling platform. When the order scheduling platform determines that two or more orders are merged into one waybill, it will assign them to a delivery rider at the same time; of course, the two or more orders merged into one waybill require the pickup time, address, etc. to be close. For two or more orders merged into one waybill, their corresponding pickup locations are guaranteed to be adjacent during sorting to improve the accuracy of the planned path; for multiple orders that cannot be merged into one waybill, a logic is constructed to determine whether there is an order-associated pickup relationship or an order-associated delivery relationship between multiple orders to help sort the pickup locations and delivery locations of these orders.

[0142] In the present application, whether two or more orders are merged into one waybill is determined by the order scheduling platform; multiple orders or two orders merged into one waybill may have an order-associated pickup relationship or an order-associated delivery relationship; however, whether they are merged into one waybill has no necessary relationship with the order-associated pickup relationship or the order-associated delivery relationship.

[0143] In this embodiment, in order to facilitate the sorting of the pick-up and delivery locations to be sorted based on the order pick-up and delivery association relationship, according to the order pick-up and delivery association relationship, association relationship tags are added to the pick-up and delivery locations to be sorted that have order pick-up and delivery association relationships with each other, and the pick-up and delivery locations to be sorted with added association relationship tags and the pick-up and delivery locations to be sorted without adding association relationship tags are used as processed pick-up and delivery locations to be sorted; therefore, as a method of obtaining predicted sorting information of the pick-up and delivery locations to be sorted based on graph features and the order pick-up and delivery association relationships between multiple orders to be delivered, it can refer to: obtaining predicted sorting information of the pick-up and delivery locations to be sorted based on the graph features and the processed pick-up and delivery locations to be sorted.

[0144] Specifically, according to the order pickup and delivery association relationship, adding association relationship labels to the pickup and delivery location points that have order pickup and delivery association relationships with each other among the pickup and delivery location points to be sorted may mean: first, according to the order pickup and delivery association relationship, determining the target orders to be delivered that have order pickup and delivery association relationships with each other among multiple orders to be delivered; then, determining the pickup and delivery location points to be sorted that have order pickup and delivery association relationships with each other among the target orders to be delivered; and then, adding association relationship labels for indicating the order pickup and delivery association relationship type to the pickup and delivery location points to be sorted that have order pickup and delivery association relationships.

[0145] For example, since orders O0 and O1 have an order-associated delivery relationship, and O0 and O2 have an order-merged relationship, orders O0, O1, and O2 can all be used as target orders to be delivered. The delivery locations D0 and D1 of orders O0 and O1 have an order-associated delivery relationship. Add an association tag to indicate the type of order-associated delivery relationship to the two locations D0 and D1. For example, after adding the association tag, D0 can become D0 与D1关联送达 , D1 can be changed to D1 after adding the association relationship tag 与D0关联送达 , that is, add a subscript to D0 and D1. This subscript is only an example of the added association label. It is understandable that other methods can be used to add association labels. For another example, since O0 and O2 have an order merging relationship, and the pickup locations P0 and P2 of orders O0 and O2 have an order-associated pickup relationship, an association label is added to the two locations P0 and P2 to indicate the type of order-associated pickup relationship. For example, after adding the association label, P0 can become P0 与P2关联取货 , P2 can become P2 after adding the association relationship tag 与P0关联取货 .

[0146] More specifically, obtaining predicted sorting information of the pick-up and delivery position points to be sorted based on the graph features and the processed pick-up and delivery position points to be sorted may refer to: inputting the graph features and the processed pick-up and delivery position points to be sorted into a trained target graph neural network model to obtain predicted sorting information of the pick-up and delivery position points to be sorted.

[0147] In this embodiment, the target graph neural network model is obtained by training the sample pickup and delivery location points, sample graph features, and actual sorting information of the sample pickup and delivery location points after adding sample labels; the sample labels refer to the association labels added to the sample pickup and delivery location points of multiple sample orders of the sample distribution capacity that have order pickup and delivery association relationships with each other; the sample graph features are obtained based on the sample distribution capacity features of the sample distribution capacity and the sample order features of multiple sample orders, and the sample graph features refer to the graph features of the graph with the sample location points and sample pickup and delivery location points where the sample distribution capacity is located as nodes; the actual sorting information is the sorting information obtained by the sample distribution capacity after delivering multiple sample orders. The sample pickup and delivery location points after adding sample labels are similar to the pickup and delivery location points to be sorted after processing, and the sample graph features are similar to the graph features in the previous text; the sample pickup and delivery location points are similar to the pickup and delivery location points to be sorted; the sample location points where the sample distribution capacity is located are similar to the current location points. The sample location points where the sample distribution capacity is located actually refer to the location points where the sample distribution capacity is located when the sample distribution capacity is planned to deliver multiple sample orders.

[0148] In this embodiment, the target graph neural network model includes an encoder and a decoder; the graph features include node features and edge features, and the edge features are used to represent the relationship features between nodes with node association relationships.

[0149] As an implementation method of inputting graph features and processed pick-up and delivery position points to be sorted into a trained target graph neural network model to obtain predicted sorting information of the pick-up and delivery position points to be sorted, it can refer to: inputting graph features into an encoder to obtain a current position point feature vector for vector representation of features of the current position point, and a pick-up and delivery position point feature vector for vector representation of features of the pick-up and delivery position points to be sorted; inputting the current position point feature vector, the pick-up and delivery position point feature vector and the processed pick-up and delivery position points to be sorted into a decoder to obtain predicted sorting information of the pick-up and delivery position points to be sorted.

[0150] In fact, after obtaining the graph features, according to Figure 3 As shown, the graph features are input into the encoder to encode the graph features, thereby obtaining a high-dimensional representation of the graph features. The high-dimensional representation of the graph features can actually include the above-mentioned current location point feature vector and the pickup and delivery location point feature vector. The encoder can be a graph convolutional neural network (GCN) as an encoder, and the decoder can be a long short-term memory artificial neural network (LSTM) as a decoder. In the decoding process of the decoder, the Attention mechanism is used to predict the sorting information of the pickup and delivery location points to be sorted.

[0151] In this embodiment, in the step of inputting the feature vector of the current location point, the feature vector of the pick-up and delivery location point, and the processed pick-up and delivery location points to be sorted into the decoder to obtain the predicted sorting information of the pick-up and delivery location points to be sorted, the decoder obtains the predicted sorting information of the pick-up and delivery location points to be sorted in the following manner: based on the feature vector of the current location point, the feature vector of the pick-up and delivery location point, and the processed pick-up and delivery location points to be sorted, determine the target pick-up and delivery location points that are first in each round of sorting among the pick-up and delivery location points to be sorted one by one; based on the confirmation order of the target pick-up and delivery location points, obtain the predicted sorting information of the pick-up and delivery location points to be sorted.

[0152] More specifically, based on the feature vector of the current location point, the feature vector of the pick-up and delivery location point and the processed pick-up and delivery location points to be sorted, the target pick-up and delivery location point that is ranked first in each round among the pick-up and delivery location points to be sorted is determined one by one; based on the confirmation order of the target pick-up and delivery location points, the predicted sorting information of the pick-up and delivery location points to be sorted is obtained, including: first, according to the pick-up and delivery strategy of picking up first and delivering later, the first candidate pick-up and delivery location point and the first non-candidate pick-up and delivery location point among the pick-up and delivery location points to be sorted are determined; the candidate pick-up and delivery location points are the location points that can participate in the sorting in each round among the pick-up and delivery location points to be sorted; then, based on the vector relationship between the feature vector of the current location point and the pick-up and delivery location point feature vector corresponding to the first candidate pick-up and delivery location point, the pick-up and delivery location point that is ranked first among the first candidate pick-up and delivery location points is determined as the first target pick-up and delivery location point; after determining the first target pick-up and delivery location point, the pick-up and delivery location point that is ranked first among the pick-up and delivery location points to be sorted is determined the second candidate pick-up and delivery position point and the second non-candidate pick-up and delivery position point; after determining the second candidate pick-up and delivery position point and the second non-candidate pick-up and delivery position point, based on the processed pick-up and delivery position points to be sorted, determine whether there is a pick-up and delivery position point in the second candidate pick-up and delivery position points that has a corresponding association relationship label added with the first target pick-up and delivery position point; if so, the pick-up and delivery position point in the second candidate pick-up and delivery position points that has a corresponding association relationship label added with the first target pick-up and delivery position point is used as the second target pick-up and delivery position point; if there is no pick-up and delivery position point in the second candidate pick-up and delivery position points that has a corresponding association relationship label added with the first target pick-up and delivery position point, based on the vector relationship between the pick-up and delivery position point feature vector corresponding to the first target pick-up and delivery position point and the pick-up and delivery position point feature vector corresponding to the second candidate pick-up and delivery position point, determine the pick-up and delivery position point ranked first in the second candidate pick-up and delivery position points as the second target pick-up and delivery position point.

[0153] After determining the second target pick-up and delivery position point, determine the third candidate pick-up and delivery position point and the third non-candidate pick-up and delivery position point among the pick-up and delivery position points to be sorted; after determining the third candidate pick-up and delivery position point and the third non-candidate pick-up and delivery position point, based on the processed pick-up and delivery position points to be sorted, determine whether there is a pick-up and delivery position point in the third candidate pick-up and delivery position points that has a corresponding association relationship label added with the second target pick-up and delivery position point; if so, use the pick-up and delivery position point in the third candidate pick-up and delivery position points that has a corresponding association relationship label added with the second target pick-up and delivery position point as the third target pick-up and delivery position point; if there is no pick-up and delivery position point in the third candidate pick-up and delivery position points that has a corresponding association relationship label added with the second target pick-up and delivery position point, then based on the vector relationship between the pick-up and delivery position point feature vector corresponding to the second target pick-up and delivery position point and the pick-up and delivery position point feature vector corresponding to the third candidate pick-up and delivery position point, determine the pick-up and delivery position point ranked first among the third candidate pick-up and delivery position points as the third target pick-up and delivery position point; and so on; until all the pick-up and delivery position points in the pick-up and delivery position points to be sorted are sorted, and the predicted sorting information is obtained.

[0154] For example, if the current location is R0, and the pick-up and delivery locations to be sorted are P0, P1, P2, P3, D0, D1, D2, and D3, since each order to be delivered needs to be picked up before delivery, in the first round of sorting, since P0 is before D0, P1 is before D1, P2 is before D2, and P3 is before D3, the first candidate pick-up and delivery locations that can participate in the sorting in the first round of sorting are P0, P1, P2, and P3, and the first non-candidate pick-up and delivery locations are D0, D1, D2, and D3; non-candidate pick-up and delivery locations are locations that cannot participate in sorting or do not need to participate in sorting in each round of sorting among the pick-up and delivery locations to be sorted. The locations that cannot participate in sorting or do not need to participate in sorting can specifically be locations that have been sorted and delivery locations that belong to the same order to be delivered as the pickup locations that have not yet been sorted. For example, in the first round of sorting, since P0, P1, P2, and P3 have not been sorted, D0, D1, D2, and D3 do not need to participate in the first round of sorting.

[0155] After determining the first candidate pick-up and delivery position point among the pick-up and delivery position points to be sorted, based on the vector relationship between the feature vector of the current position point and the feature vector of the pick-up and delivery position point corresponding to the first candidate pick-up and delivery position point, determine the pick-up and delivery position point ranked first among the first candidate pick-up and delivery position points as the first target pick-up and delivery position point. For example, based on the vector relationship between the feature vector of RO and the feature vector of P0, the feature vector of P1, the feature vector of P2, and the feature vector of P3, determine P0 as the first target pick-up and delivery position point. The above vector relationship can use the distance between vectors as an example. In the present application, the pick-up and delivery position point feature vector corresponding to the candidate pick-up and delivery position point is actually the pick-up and delivery position point feature vector corresponding to the candidate pick-up and delivery position point related to the current position point.

[0156] After determining the first target pick-up and delivery position point, the second candidate pick-up and delivery position points that can participate in the sorting in the second round are determined to be P1, P2, P3, and D0, and the second non-candidate pick-up and delivery position points are D1, D2, D3, and P0; because based on the processed pick-up and delivery position points to be sorted, it can be determined that P2 and P0 in the second candidate pick-up and delivery position points have added corresponding association relationship labels, and therefore P2 is selected as the second target pick-up and delivery position point.

[0157] After determining the second target pick-up and delivery position point, the third candidate pick-up and delivery position points that can participate in the sorting in the third round of sorting are determined to be P1, P3, D0, and D2, and the third non-candidate pick-up and delivery position points are D1, D3, P0, and P2. Since it can be determined based on the processed pick-up and delivery position points to be sorted that there is no position point in the third candidate pick-up and delivery position points with a corresponding association relationship label added to P2, the pick-up and delivery position point feature vector corresponding to the second target pick-up and delivery position point and the pick-up and delivery position point feature vector corresponding to the third candidate pick-up and delivery position point are respectively determined. The pick-up and delivery position point ranked first among the third candidate pick-up and delivery position points is determined as the third target pick-up and delivery position point. For example, based on the vector relationship between the feature vector of P2 and the feature vector of P1, the feature vector of P3, the feature vector of D0, and the feature vector of D2, P1 is determined to be the third target pick-up and delivery position point.

[0158] After determining the third target pick-up and delivery position point, the fourth candidate pick-up and delivery position points that can participate in the sorting in the fourth round of sorting are determined to be P3, D0, D1, and D2, and the fourth non-candidate pick-up and delivery position points are D3, P0, P1, and P2. Since it can be determined based on the processed pick-up and delivery position points to be sorted that there is no position point in the fourth candidate pick-up and delivery position points with a corresponding association relationship label added to P1, the pick-up and delivery position point feature vector corresponding to the third target pick-up and delivery position point and the pick-up and delivery position point feature vector corresponding to the fourth candidate pick-up and delivery position point are respectively determined. The pick-up and delivery position point ranked first among the fourth candidate pick-up and delivery position points is determined as the fourth target pick-up and delivery position point. For example, based on the vector relationship between the feature vector of P1 and the feature vector of P3, the feature vector of D0, the feature vector of D1, and the feature vector of D2, D0 is determined to be the fourth target pick-up and delivery position point.

[0159] After determining the fourth target pick-up and delivery position point, the fifth candidate pick-up and delivery position points that can participate in the sorting in the fifth round of sorting are determined to be P3, D1, and D2, and the fifth non-candidate pick-up and delivery position points are D3, P0, P1, P2, and D0. Since it can be determined based on the processed pick-up and delivery position points to be sorted that D1 and D0 in the fifth candidate pick-up and delivery position points have corresponding association relationship labels added, D1 is used as the fifth target pick-up and delivery position point.

[0160] After determining the fifth target pick-up and delivery position point, the sixth candidate pick-up and delivery position points that can participate in the sorting in the sixth round of sorting are determined to be P3 and D2, and the sixth non-candidate pick-up and delivery position points are D1, D3, P0, P2, D0, and P1. Since it can be determined based on the processed pick-up and delivery position points to be sorted that there is no position point in the sixth candidate pick-up and delivery position points with a corresponding association relationship label added to D1, the pick-up and delivery position point feature vector corresponding to the fifth target pick-up and delivery position point and the pick-up and delivery position point feature vector corresponding to the sixth candidate pick-up and delivery position point are respectively determined. The pick-up and delivery position point ranked first among the sixth candidate pick-up and delivery position points is determined as the sixth target pick-up and delivery position point. For example, based on the vector relationship between the feature vector of D1 and the feature vector of P3 and the feature vector of D2, P3 is determined to be the sixth target pick-up and delivery position point.

[0161] According to the above method, the seventh candidate pick-up and delivery position points that can participate in the sorting in the seventh round of sorting are further determined to be D3 and D2, and the seventh non-candidate pick-up and delivery position points are D1, P3, P0, P2, D0, and P1. Since it can be determined based on the processed pick-up and delivery position points to be sorted that there is no position point in the seventh candidate pick-up and delivery position points with a corresponding association relationship label added to P3, the pick-up and delivery position point that is ranked first among the seventh candidate pick-up and delivery position points is determined as the seventh target pick-up and delivery position point based on the vector relationship between the pick-up and delivery position point feature vector corresponding to the sixth target pick-up and delivery position point and the pick-up and delivery position point feature vector corresponding to the seventh candidate pick-up and delivery position point. For example, based on the vector relationship between the feature vector of P3 and the feature vector of D3 and the feature vector of D2, D2 is determined to be the seventh target pick-up and delivery position point, and finally D3 is the eighth target pick-up and delivery position point. Therefore, the predicted sorting information of each pick-up and delivery position point to be sorted is finally R0, P0, P2, P1, D0, D1, P3, D2, and D3.

[0162] In this embodiment, it also includes: after determining the candidate pick-up and delivery position points and non-candidate pick-up and delivery position points in each round, the non-candidate pick-up and delivery position points are shielded by masking. The reason for shielding the non-candidate pick-up and delivery position points by masking is to facilitate the decoder to determine the corresponding target pick-up and delivery position points in each round of sorting. For ease of understanding, please refer to Figure 4 , which is a schematic diagram of the third scenario of the order pickup and delivery location point sorting method of the present application.

[0163] When predicting the first ranked position point in the first round, since the first candidate pick-up and delivery position points are P0, P1, P2, and P3, and the first non-candidate pick-up and delivery position points are D0, D1, D2, and D3, a mask method (i.e., mask) is used to block D0, D1, D2, and D3, for example, Figure 4As shown, the background colors of D0, D1, D2, and D3 are set to gray; similarly, the second non-candidate pick-up and delivery position point until the eighth non-candidate pick-up and delivery position point are subsequently shielded by a masking method.

[0164] Step S204: providing the predicted sorting information to the delivery terminal, so that the delivery terminal determines a delivery route for delivering multiple orders to be delivered based on the current location, the pickup and delivery location to be sorted, and the predicted sorting information.

[0165] After obtaining the predicted sorting information of the pickup and delivery locations to be sorted, the predicted sorting information is provided to the delivery terminal, so that the delivery terminal determines the delivery path for delivering multiple orders to be delivered based on the current location, the pickup and delivery locations to be sorted and the predicted sorting information.

[0166] Of course, in this embodiment, the server can also determine the delivery path for delivering multiple orders to be delivered based on the current location, the pickup and delivery location to be sorted, and the predicted sorting information, and then provide the planned delivery path information to the delivery terminal.

[0167] Specifically, based on the current location point, the pickup and delivery location points to be sorted, and the predicted sorting information, determining the delivery path for delivering multiple orders to be delivered can mean: arranging the pickup and delivery location points to be sorted after the current location point in sequence according to the sorting corresponding to the predicted sorting information, and determining each sorted location point; performing path planning based on each sorted location point, and determining the delivery path for delivering multiple orders to be delivered. Specifically, each sorted location point has corresponding geographic location information. After determining each sorted location point, the delivery path for delivering multiple orders to be delivered is planned based on the geographic location information corresponding to each sorted location point.

[0168] The present application provides a method for sorting order pick-up and delivery location points, which is applied to a service end. In this method, in response to detecting a sorting request message sent by a distribution terminal for requesting to sort the to-be-sorted pick-up and delivery location points of multiple to-be-delivered orders that require a target distribution capacity to perform pick-up and delivery tasks, multiple to-be-delivered orders are determined; then, based on the distribution capacity characteristics of the target distribution capacity and the order characteristics of the multiple to-be-delivered orders, graph features of a graph with the current location point of the target distribution capacity and the to-be-sorted pick-up and delivery location points as nodes are obtained; the graph features obtained by taking the current location point and the to-be-sorted pick-up and delivery location points as nodes can better reflect the mutual correlation between the current location point and each pick-up and delivery location point, and then the graph features, the multiple to-be-delivered orders are sorted. The order pick-up and delivery association relationship between them, the predicted sorting information of the pick-up and delivery location points to be sorted can also reflect the mutual association relationship between each location point; at the same time, the order pick-up and delivery association relationship includes at least one of the order merging relationship, the order-related pickup relationship, and the order-related delivery relationship. Since the order pick-up and delivery association relationship is taken into account when obtaining the above-mentioned predicted sorting information, and the delivery capacity will also consider the order pick-up and delivery association relationship when delivering multiple orders, in fact, this method takes into account the delivery behavior habits of the delivery capacity, that is, the obtained predicted sorting information is in line with the behavior habits of the delivery riders in the actual delivery process, so that the subsequent delivery path determined based on the predicted sorting information is more in line with reality; at the same time, it greatly facilitates the delivery process of the delivery riders, thereby improving the delivery efficiency.

[0169] Second embodiment

[0170] Corresponding to the first embodiment, the second embodiment of the present application provides a method for determining a delivery path. The execution subject of this embodiment is a delivery terminal, and the same parts of the second embodiment as the first embodiment are not introduced again. For details, please refer to the relevant parts of the first embodiment and the scenario embodiment.

[0171] Please refer to Figure 5 , which is a flow chart of the delivery path determination method provided in the second embodiment of the present application.

[0172] The delivery path determination method of the embodiment of the present application is applied to the delivery terminal, and includes the following steps:

[0173] Step S501: sending a sorting request message to the server for requesting to sort the to-be-sorted pick-up and delivery location points of a plurality of to-be-delivered orders that require a target delivery capacity to perform pick-up and delivery tasks.

[0174] A sorting request message is sent to the server so that the server responds to the request message and multiple orders to be delivered are determined; based on the delivery capacity characteristics of the target delivery capacity and the order characteristics of the multiple orders to be delivered, the graph characteristics of the graph with the current location of the target delivery capacity and the pick-up and delivery locations to be sorted as nodes are obtained; according to the graph characteristics and the order pick-up and delivery association relationship between the multiple orders to be delivered, the predicted sorting information of the pick-up and delivery locations to be sorted is obtained; the order pick-up and delivery association relationship includes at least one of an order merging relationship, an order-associated pickup relationship, and an order-associated delivery relationship.

[0175] Step S502: Obtain prediction ranking information provided by the server.

[0176] Step S503: Based on the current location, the pickup and delivery locations to be sorted, and the predicted sorting information, a delivery route for delivering multiple orders to be delivered is determined.

[0177] Specifically, based on the current location point, the pickup and delivery location points to be sorted, and the predicted sorting information, a delivery path for delivering multiple orders to be delivered is determined, including: arranging the pickup and delivery location points to be sorted after the current location point in sequence according to the sorting corresponding to the predicted sorting information, and determining each location point after sorting; performing path planning based on each location point after sorting, and determining a delivery path for delivering multiple orders to be delivered. Specifically, each location point after sorting corresponds to geographic location information. After determining each location point after sorting, a delivery path for delivering multiple orders to be delivered is planned based on the geographic location information corresponding to each location point after sorting. Of course, when a delivery rider receives a new delivery order or increases or decreases the pickup and delivery location points that need to perform pick-up and delivery tasks, a new delivery path can be replanned.

[0178] After determining the delivery path for delivering multiple orders to be delivered, the corresponding delivery path information can be displayed or broadcasted at the delivery terminal to facilitate navigation of the delivery rider during the delivery process.

[0179] The present application provides a method for determining a delivery path, which is applied to a delivery terminal. In the method, a sorting request message is sent to a server to request sorting of to-be-sorted pickup and delivery locations of multiple to-be-delivered orders that require a target delivery transport capacity to perform pickup and delivery tasks; thereafter, predicted sorting information provided by the server is obtained; based on the current location, the to-be-sorted pickup and delivery locations, and the predicted sorting information, a delivery path for delivering multiple to-be-delivered orders is determined. Since the order pickup and delivery association relationship is considered when obtaining the predicted sorting information, and the delivery transport capacity will also consider the order pickup and delivery association relationship when actually delivering multiple orders, in fact, the method takes into account the delivery behavior habits of the delivery transport capacity, that is, the obtained predicted sorting information is consistent with the behavior habits of the delivery riders in the actual delivery process, so that the delivery path determined based on the predicted sorting information is more in line with reality; at the same time, it greatly facilitates the delivery process of the delivery riders, thereby improving the delivery efficiency.

[0180] Third embodiment

[0181] Corresponding to the order pickup and delivery location point sorting method provided in the first embodiment of the present application, the third embodiment of the present application also provides an order pickup and delivery location point sorting device. Since the device embodiment is basically similar to the first embodiment, the description is relatively simple, and the relevant parts can refer to the partial description of the first embodiment. The device embodiment described below is only illustrative.

[0182] Please refer to Figure 6 , which is a schematic diagram of the order pickup and delivery location point sorting device provided in the third embodiment of the present application.

[0183] The order pickup and delivery location point sorting device 600 is applied to the server, and the device includes:

[0184] The to-be-delivered order determination unit 601 is configured to determine the multiple to-be-delivered orders in response to detecting a sorting request message sent by a delivery terminal for requesting to sort the to-be-sorted pick-up and delivery location points of the multiple to-be-delivered orders that require a target delivery transport capacity to perform pick-up and delivery tasks;

[0185] A graph feature obtaining unit 602 is used to obtain graph features of a graph with the current location of the target delivery capacity and the pickup and delivery location points to be sorted as nodes based on the delivery capacity features of the target delivery capacity and the order features of multiple orders to be delivered;

[0186] The first prediction sorting information obtaining unit 603 is used to obtain the prediction sorting information of the pickup and delivery location points to be sorted according to the graph features and the order pickup and delivery association relationship between the multiple orders to be delivered; the order pickup and delivery association relationship includes at least one of an order merging relationship, an order-associated pickup relationship, and an order-associated delivery relationship;

[0187] The predicted sorting information providing unit 604 is used to provide the predicted sorting information to the delivery terminal, so that the delivery terminal determines the delivery path for delivering the multiple orders to be delivered based on the current location point, the pickup and delivery location point to be sorted and the predicted sorting information.

[0188] Optionally, the graph feature obtaining unit is specifically used to:

[0189] Obtaining the time characteristics of the distribution capacity associated with time in the distribution capacity characteristics and the space characteristics of the distribution capacity associated with space in the distribution capacity characteristics;

[0190] Obtaining order time features associated with time in the order features, and order space features associated with space in the order features;

[0191] The graph feature is obtained according to the delivery capacity time feature, the delivery capacity space feature, the order time feature, and the order space feature.

[0192] Optionally, the graph features include node features and edge features;

[0193] The graph feature obtaining unit is specifically used for:

[0194] If the node represents the current location point, the time feature of the delivery capacity is concatenated with the spatial feature of the delivery capacity to obtain the node feature;

[0195] If the node represents a pickup location, concatenate the first order time feature associated with pickup in the order time feature and the first order space feature associated with pickup in the order space feature to obtain the node feature;

[0196] If the node represents a delivery location point, concatenate the second order time feature associated with the delivery in the order time feature and the second order space feature associated with the delivery in the order space feature to obtain the node feature;

[0197] The edge feature is obtained according to the node feature and the node association relationship between nodes.

[0198] Optionally, the graph feature obtaining unit is specifically used to:

[0199] For any pair of two nodes having a node association relationship, determine two target nodes having a node association relationship;

[0200] Based on the node characteristics of the two target nodes, determine the distance and theoretical time difference between the position points represented by the two target nodes, where the theoretical time difference is the theoretical time required from the position point represented by one target node to the position point represented by another target node;

[0201] The distance and the theoretical time difference are used as the edge features.

[0202] Optionally, the method further includes: an order pickup and delivery association determination unit, which is specifically used to:

[0203] Determine each pickup location point and each delivery location point of the plurality of orders to be delivered;

[0204] Determine the pickup location relationship between each pickup location point and the delivery location relationship between each delivery location point;

[0205] According to at least one of the pickup position relationship and the delivery position relationship, an order pickup and delivery association relationship between the multiple orders to be delivered is determined.

[0206] Optionally, the order pickup and delivery association determination unit is specifically configured to:

[0207] For any two orders among the multiple orders to be delivered, determine whether the position relationship between the two pickup locations of the two orders meets the requirement of being adjacent to each other and whether the theoretical pickup time difference between the two orders is less than a first preset time difference; if yes, confirm that the two orders have an order-associated pickup relationship;

[0208] For any two orders among the multiple orders to be delivered, determine whether the positional relationship between the two delivery location points of the two orders meets the requirement of adjacent locations and whether the theoretical delivery time difference between the two orders is less than a second preset time difference; if so, confirm that the two orders have an order-related delivery relationship.

[0209] Optionally, it further includes: an order merging relationship confirmation unit, which is specifically used to:

[0210] For any two orders among the multiple orders to be delivered, determine whether the two orders are merged into one waybill;

[0211] If two orders are merged into one waybill, it is confirmed that there is an order merging relationship between the two orders.

[0212] Optionally, it further includes: an association relationship label adding unit, which is specifically used to:

[0213] According to the order pick-up and delivery association relationship, an association relationship tag is added to the pick-up and delivery location points that have the order pick-up and delivery association relationship among the pick-up and delivery location points to be sorted, and the pick-up and delivery location points to which the association relationship tag is added and the pick-up and delivery location points to which the association relationship tag is not added are used as the processed pick-up and delivery location points to be sorted;

[0214] The prediction ranking information first obtaining unit is specifically used for:

[0215] According to the graph features and the processed pick-up and delivery position points to be sorted, predicted sorting information of the pick-up and delivery position points to be sorted is obtained.

[0216] Optionally, the association relationship label adding unit is specifically used to:

[0217] According to the order pickup and delivery association relationship, determining target orders to be delivered that have the order pickup and delivery association relationship among the multiple orders to be delivered;

[0218] Determine the pickup and delivery location points to be sorted that have the order pickup and delivery association relationship among the target orders to be delivered;

[0219] An association relationship label for indicating the type of the order pickup and delivery association relationship is added to the to-be-sorted pickup and delivery location point having the order pickup and delivery association relationship.

[0220] Optionally, the prediction ranking information first obtaining unit is specifically used to:

[0221] The graph features and the processed pick-up and delivery locations to be sorted are input into a trained target graph neural network model to obtain predicted sorting information of the pick-up and delivery locations to be sorted; the target graph neural network model is trained using the sample pick-up and delivery locations processed by adding sample labels, sample graph features, and actual sorting information of the sample pick-up and delivery locations; the sample labels refer to association labels added to the sample pick-up and delivery locations of multiple sample orders of the sample distribution capacity that have the order pick-up and delivery association relationship with each other; the sample graph features are obtained based on the sample distribution capacity features of the sample distribution capacity and the sample order features of the multiple sample orders, and the sample graph features refer to the graph features of the graph with the sample locations of the sample distribution capacity and the sample pick-up and delivery locations as nodes; the actual sorting information is the sorting information obtained by the sample distribution capacity after delivering the multiple sample orders.

[0222] Optionally, the target graph neural network model includes an encoder and a decoder; the graph features include node features and edge features, and the edge features are used to represent relationship features between nodes with node association relationships;

[0223] The prediction ranking information first obtaining unit is specifically used for:

[0224] Input the graph features into an encoder to obtain a current location point feature vector for vector representation of the features of the current location point and a pick-up and delivery location point feature vector for vector representation of the features of the pick-up and delivery location points to be sorted;

[0225] The feature vector of the current location point, the feature vector of the pick-up and delivery location point and the processed pick-up and delivery location points to be sorted are input into the decoder to obtain the predicted sorting information of the pick-up and delivery location points to be sorted.

[0226] Optionally, the prediction ranking information first obtaining unit is specifically used to:

[0227] Based on the feature vector of the current location point, the feature vector of the pick-up and delivery location point and the processed pick-up and delivery location points to be sorted, the target pick-up and delivery location points that are ranked first in each round among the pick-up and delivery location points to be sorted are determined one by one; based on the confirmation order of the target pick-up and delivery location points, the predicted sorting information of the pick-up and delivery location points to be sorted is obtained.

[0228] Optionally, the prediction ranking information first obtaining unit is specifically used to:

[0229] According to the order pick-up-first-then-delivery pick-up and delivery strategy, determine the first candidate pick-up and delivery location point and the first non-candidate pick-up and delivery location point among the pick-up and delivery location points to be sorted; the candidate pick-up and delivery location points are location points among the pick-up and delivery location points to be sorted that can participate in sorting in each round of sorting;

[0230] Based on the vector relationship between the feature vector of the current location point and the feature vector of the pickup and delivery location point corresponding to the first candidate pickup and delivery location point, determine the pickup and delivery location point ranked first among the first candidate pickup and delivery location points as the first target pickup and delivery location point;

[0231] Determine a second candidate pick-up and delivery position point and a second non-candidate pick-up and delivery position point among the pick-up and delivery position points to be sorted;

[0232] Based on the processed pick-up and delivery position points to be sorted, it is determined whether there is a pick-up and delivery position point in the second candidate pick-up and delivery position points that has a corresponding association relationship tag added with the first target pick-up and delivery position point; if so, the pick-up and delivery position point in the second candidate pick-up and delivery position points that has a corresponding association relationship tag added with the first target pick-up and delivery position point is used as the second target pick-up and delivery position point;

[0233] Determine a third candidate pick-up and delivery position point and a third non-candidate pick-up and delivery position point among the pick-up and delivery position points to be sorted;

[0234] Based on the processed pick-up and delivery position points to be sorted, determine whether there is a pick-up and delivery position point in the third candidate pick-up and delivery position point that has a corresponding association relationship tag added with the second target pick-up and delivery position point; if yes, use the pick-up and delivery position point in the third candidate pick-up and delivery position point that has a corresponding association relationship tag added with the second target pick-up and delivery position point as the third target pick-up and delivery position point;

[0235] And so on; until all the pick-up and delivery position points in the to-be-sorted pick-up and delivery position points are sorted, the predicted sorting information is obtained.

[0236] Optionally, the prediction ranking information first obtaining unit is further used to:

[0237] If there is no pick-up and delivery location point in the second candidate pick-up and delivery location points that has a corresponding association relationship tag added to the first target pick-up and delivery location point, then based on the vector relationship between the pick-up and delivery location point feature vector corresponding to the first target pick-up and delivery location point and the pick-up and delivery location point feature vector corresponding to the second candidate pick-up and delivery location point, determine the pick-up and delivery location point ranked first among the second candidate pick-up and delivery location points as the second target pick-up and delivery location point;

[0238] If there is no pick-up and delivery location point in the third candidate pick-up and delivery location points that has a corresponding association relationship tag added to the second target pick-up and delivery location point, then based on the vector relationship between the pick-up and delivery location point feature vector corresponding to the second target pick-up and delivery location point and the pick-up and delivery location point feature vector corresponding to the third candidate pick-up and delivery location point, determine the pick-up and delivery location point ranked first among the third candidate pick-up and delivery location points as the third target pick-up and delivery location point;

[0239] And so on; until all the pick-up and delivery position points in the to-be-sorted pick-up and delivery position points are sorted, the predicted sorting information is obtained.

[0240] Optionally, it further includes: a shielding unit, specifically used for:

[0241] After determining the candidate pick-up and delivery location points and non-candidate pick-up and delivery location points in each round, the non-candidate pick-up and delivery location points are blocked by using a mask.

[0242] Fourth embodiment

[0243] Corresponding to the delivery path determination method provided in the second embodiment of the present application, the fourth embodiment of the present application also provides a delivery path determination device. Since the device embodiment is basically similar to the second embodiment, the description is relatively simple, and the relevant parts can refer to the partial description of the second embodiment. The device embodiment described below is only illustrative.

[0244] Please refer to Figure 7 , which is a schematic diagram of a delivery path determination device provided in the fourth embodiment of the present application.

[0245] The delivery path determination device 700 is applied to a delivery terminal, and the device includes:

[0246] The sorting request message sending unit 701 is used to send a sorting request message to the server for requesting to sort the to-be-sorted pick-up and delivery locations of multiple to-be-delivered orders that require the target delivery capacity to perform pick-up and delivery tasks, so that the server determines the multiple to-be-delivered orders in response to the request message; based on the delivery capacity characteristics of the target delivery capacity and the order characteristics of the multiple to-be-delivered orders, obtain the graph characteristics of a graph with the current location of the target delivery capacity and the to-be-sorted pick-up and delivery locations as nodes; according to the graph characteristics and the order pick-up and delivery association relationship between the multiple to-be-delivered orders, obtain the predicted sorting information of the to-be-sorted pick-up and delivery locations; the order pick-up and delivery association relationship includes at least one of an order merging relationship, an order-associated pickup relationship, and an order-associated delivery relationship;

[0247] A second prediction ranking information obtaining unit 702 is used to obtain the prediction ranking information provided by the server;

[0248] The delivery path determination unit 703 is used to determine a delivery path for delivering the multiple orders to be delivered based on the current location point, the pickup and delivery location points to be sorted, and the predicted sorting information.

[0249] Optionally, the delivery path determination unit is specifically used to:

[0250] According to the order corresponding to the predicted order information, the pick-up and delivery location points to be ordered are arranged after the current location point in sequence, and each location point after the order is determined;

[0251] Path planning is performed based on the sorted location points to determine a delivery path for delivering the multiple orders to be delivered.

[0252] Fifth embodiment

[0253] Corresponding to the methods of the first and second embodiments of the present application, the fifth embodiment of the present application further provides an electronic device.

[0254] like Figure 8 As shown, Figure 8 A schematic diagram of an electronic device provided in accordance with the fifth embodiment of the present application.

[0255] In this embodiment, an optional hardware structure of the electronic device 800 can be as follows: Figure 8 As shown, it includes: at least one processor 801, at least one memory 802 and at least one communication bus 805; the memory 802 contains a program 803 and data 804.

[0256] The bus 805 may be a communication device for transmitting data between components inside the electronic device 800, such as an internal bus (e.g., a CPU-memory bus, where the processor is a central processing unit, CPU for short), an external bus (e.g., a universal serial bus port, a peripheral component interconnect express port), etc.

[0257] In addition, the electronic device also includes: at least one network interface 806, at least one peripheral interface 807. The network interface 806 provides wired or wireless communication related to an external network 808 (for example, the Internet, an intranet, a local area network, a mobile communication network, etc.); in some embodiments, the network interface 806 may include any number of network interface controllers (NICs), radio frequency (RF) modules, repeaters, transceivers, modems, routers, gateways, any combination of wired network adapters, wireless network adapters, Bluetooth adapters, infrared adapters, near field communication (NFC) adapters, cellular network chips, etc.

[0258] The peripheral interface 807 is used to connect to the peripheral device, and the peripheral device can be the peripheral device 1 ( Figure 8 809 in), peripheral 2 ( Figure 8 810) and peripheral 3 ( Figure 8 811 in the figure). Peripherals are peripheral devices, which may include but are not limited to cursor control devices (such as mice, touch pads or touch screens), keyboards, displays (such as cathode ray tube displays, liquid crystal displays). Displays or light emitting diode displays, video input devices (such as cameras or communication interfaces coupled to video files), etc.

[0259] The processor 801 may be a CPU, or an application specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present application.

[0260] The memory 802 may include a high-speed RAM (full name: Random Access Memory) memory, and may also include a non-volatile memory, such as at least one disk memory.

[0261] The processor 801 calls the program and data stored in the memory 802 to execute the methods of the first embodiment and the second embodiment of the present application.

[0262] Sixth embodiment

[0263] Corresponding to the methods of the first and second embodiments of the present application, the sixth embodiment of the present application further provides a computer storage medium, wherein the computer storage medium stores a computer program, and the computer program is run by a processor to execute the methods of the first and second embodiments of the present application.

[0264] Although the present application is disclosed as above in the form of a preferred embodiment, it is not intended to limit the present application. Any technical personnel in this field may make possible changes and modifications without departing from the spirit and scope of the present application. Therefore, the scope of protection of the present application shall be based on the scope defined by the claims of the present application.

[0265] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory. The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

[0266] 1. Computer-readable media includes permanent and non-permanent, removable and non-removable media that can be used to store information by any method or technology. Information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission medium that can be used to store information that can be accessed by a computing device. According to the definition in this article, computer readable media does not include non-transitory computer readable storage media, such as modulated data signals and carrier waves.

[0267] 2. Those skilled in the art should understand that the embodiments of the present application can be provided as methods, systems or computer program products. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present application can take 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.

[0268] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

Claims

1. A method for sorting order pick-up and delivery locations, applied to a server, characterized in that: The method comprises: In response to detecting a sorting request message sent by a distribution terminal for requesting to sort the to-be-sorted pick-up and delivery location points of a plurality of to-be-delivered orders requiring a target distribution transport capacity to perform pick-up and delivery tasks, determining the plurality of to-be-delivered orders; Based on the delivery capacity characteristics of the target delivery capacity and the order characteristics of the multiple orders to be delivered, obtaining the graph characteristics of a graph with the current location of the target delivery capacity and the pickup and delivery location points to be sorted as nodes; According to the graph features and the order pickup and delivery association relationship between the multiple orders to be delivered, the predicted sorting information of the pickup and delivery location points to be sorted is obtained; the order pickup and delivery association relationship includes at least one of an order merging relationship, an order-associated pickup relationship, and an order-associated delivery relationship; The predicted sorting information is provided to the delivery terminal, so that the delivery terminal determines a delivery route for delivering the multiple orders to be delivered based on the current location, the pickup and delivery location to be sorted, and the predicted sorting information.

2. The method according to claim 1, characterized in that The obtaining of the graph features of a graph having the current location of the target delivery capacity and the pickup and delivery location points to be sorted as nodes based on the delivery capacity features of the target delivery capacity and the order features of the multiple orders to be delivered includes: Obtaining the time characteristics of the distribution capacity associated with time in the distribution capacity characteristics and the space characteristics of the distribution capacity associated with space in the distribution capacity characteristics; Obtaining order time features associated with time in the order features, and order space features associated with space in the order features; The graph feature is obtained according to the delivery capacity time feature, the delivery capacity space feature, the order time feature, and the order space feature.

3. The method according to claim 2, characterized in that The graph features include node features and edge features; The obtaining of the graph feature according to the distribution capacity time feature, the distribution capacity space feature, the order time feature, and the order space feature includes: If the node represents the current location point, the time feature of the delivery capacity is concatenated with the spatial feature of the delivery capacity to obtain the node feature; If the node represents a pickup location, concatenate the first order time feature associated with pickup in the order time feature and the first order space feature associated with pickup in the order space feature to obtain the node feature; If the node represents a delivery location point, concatenate the second order time feature associated with the delivery in the order time feature and the second order space feature associated with the delivery in the order space feature to obtain the node feature; The edge feature is obtained according to the node feature and the node association relationship between nodes.

4. The method according to claim 3, characterized in that: The step of obtaining the edge feature according to the node feature and the node association relationship between the nodes includes: For any pair of two nodes having a node association relationship, determine two target nodes having a node association relationship; Based on the node characteristics of the two target nodes, determine the distance and theoretical time difference between the position points represented by the two target nodes, where the theoretical time difference is the theoretical time required from the position point represented by one target node to the position point represented by another target node; The distance and the theoretical time difference are used as the edge features.

5. The method according to claim 1, characterized in that Also includes: Determine each pickup location point and each delivery location point of the plurality of orders to be delivered; Determine the pickup location relationship between each pickup location point and the delivery location relationship between each delivery location point; According to at least one of the pickup position relationship and the delivery position relationship, an order pickup and delivery association relationship between the multiple orders to be delivered is determined.

6. A method for determining a delivery path, applied to a delivery terminal, characterized in that: The method comprises: Sending a sorting request message to a server for requesting sorting of the to-be-sorted pick-up and delivery locations of a plurality of to-be-delivered orders that require a target delivery capacity to perform pick-up and delivery tasks, so that the server determines the plurality of to-be-delivered orders in response to the request message; obtaining graph features of a graph with the current location of the target delivery capacity and the to-be-sorted pick-up and delivery locations as nodes based on the delivery capacity features of the target delivery capacity and the order features of the plurality of to-be-delivered orders; obtaining predicted sorting information of the to-be-sorted pick-up and delivery locations based on the graph features and the order pick-up and delivery association relationships between the plurality of to-be-delivered orders; the order pick-up and delivery association relationships at least include at least one of an order merging relationship, an order-associated pickup relationship, and an order-associated delivery relationship; Obtaining the predicted ranking information provided by the server; Based on the current location point, the pickup and delivery location points to be sorted and the predicted sorting information, a delivery path for delivering the multiple orders to be delivered is determined.

7. An order pick-up and delivery location point sorting device, applied to a service end, characterized in that: The device comprises: A to-be-delivered order determination unit, configured to determine the multiple to-be-delivered orders in response to detecting a sorting request message sent by a delivery terminal for requesting to sort the to-be-sorted pick-up and delivery location points of the multiple to-be-delivered orders that require a target delivery transport capacity to perform pick-up and delivery tasks; A graph feature obtaining unit, configured to obtain graph features of a graph having a current location of the target delivery capacity and the pickup and delivery location points to be sorted as nodes based on the delivery capacity features of the target delivery capacity and the order features of the multiple orders to be delivered; A first prediction sorting information obtaining unit is used to obtain prediction sorting information of the pickup and delivery location points to be sorted according to the graph features and the order pickup and delivery association relationship between the multiple orders to be delivered; the order pickup and delivery association relationship includes at least one of an order merging relationship, an order-associated pickup relationship, and an order-associated delivery relationship; The predicted sorting information providing unit is used to provide the predicted sorting information to the delivery terminal, so that the delivery terminal determines the delivery path for delivering the multiple orders to be delivered based on the current location point, the pick-up and delivery location point to be sorted and the predicted sorting information.

8. A distribution path determination device, applied to a distribution terminal, characterized in that: The device comprises: A sorting request message sending unit is used to send a sorting request message to a server for requesting sorting of the to-be-sorted pick-up and delivery locations of multiple to-be-delivered orders that require a target delivery capacity to perform pick-up and delivery tasks, so that the server determines the multiple to-be-delivered orders in response to the request message; based on the delivery capacity characteristics of the target delivery capacity and the order characteristics of the multiple to-be-delivered orders, obtain graph characteristics of a graph with the current location of the target delivery capacity and the to-be-sorted pick-up and delivery locations as nodes; based on the graph characteristics and the order pick-up and delivery association relationship between the multiple to-be-delivered orders, obtain predicted sorting information of the to-be-sorted pick-up and delivery locations; the order pick-up and delivery association relationship includes at least one of an order merging relationship, an order-associated pickup relationship, and an order-associated delivery relationship; A second prediction ranking information obtaining unit, used to obtain the prediction ranking information provided by the server; A delivery path determination unit is used to determine a delivery path for delivering the multiple orders to be delivered based on the current location point, the pickup and delivery location points to be sorted and the predicted sorting information.

9. An electronic device, characterized in that: include: processor; A memory for storing a computer program, wherein the computer program is executed by a processor to perform the method according to any one of claims 1 to 6.

10. A computer storage medium, characterized in that: The computer storage medium stores a computer program, which is executed by a processor to perform the method according to any one of claims 1 to 6.

Images